Stephen Wolfram Blog Stephen Wolfram's Personal Blog 2017-11-20T23:23:21Z http://blog.stephenwolfram.com/feed/atom/ WordPress Stephen Wolfram <![CDATA[What Is a Computational Essay?]]> http://blog.internal.stephenwolfram.com/?p=14921 2017-11-20T23:23:21Z 2017-11-14T18:42:11Z compessaythumbA Powerful Way to Express Ideas People are used to producing prose—and sometimes pictures—to express themselves. But in the modern age of computation, something new has become possible that I’d like to call the computational essay. I’ve been working on building the technology to support computational essays for several decades, but it’s only very recently [...]]]> compessaythumb

A Powerful Way to Express Ideas

People are used to producing prose—and sometimes pictures—to express themselves. But in the modern age of computation, something new has become possible that I’d like to call the computational essay.

I’ve been working on building the technology to support computational essays for several decades, but it’s only very recently that I’ve realized just how central computational essays can be to both the way people learn, and the way they communicate facts and ideas. Professionals of the future will routinely deliver results and reports as computational essays. Educators will routinely explain concepts using computational essays. Students will routinely produce computational essays as homework for their classes.

Here’s a very simple example of a computational essay:

Simple computational essay example

There are basically three kinds of things here. First, ordinary text (here in English). Second, computer input. And third, computer output. And the crucial point is that these all work together to express what’s being communicated.

The ordinary text gives context and motivation. The computer input gives a precise specification of what’s being talked about. And then the computer output delivers facts and results, often in graphical form. It’s a powerful form of exposition that combines computational thinking on the part of the human author with computational knowledge and computational processing from the computer.

But what really makes this work is the Wolfram Language—and the succinct representation of high-level ideas that it provides, defining a unique bridge between human computational thinking and actual computation and knowledge delivered by a computer.

In a typical computational essay, each piece of Wolfram Language input will usually be quite short (often not more than a line or two). But the point is that such input can communicate a high-level computational thought, in a form that can readily be understood both by the computer and by a human reading the essay.

It’s essential to all this that the Wolfram Language has so much built-in knowledge—both about the world and about how to compute things in it. Because that’s what allows it to immediately talk not just about abstract computations, but also about real things that exist and happen in the world—and ultimately to provide a true computational communication language that bridges the capabilities of humans and computers.

An Example

Let’s use a computational essay to explain computational essays.

Let’s say we want to talk about the structure of a human language, like English. English is basically made up of words. Let’s get a list of the common ones.

Generate a list of common words in English:

WordList[]

WordList[]

How long is a typical word? Well, we can take the list of common words, and make a histogram that shows their distribution of lengths.

Make a histogram of word lengths:

Histogram[StringLength[WordList[]]]

Histogram[StringLength[WordList[]]]

Do the same for French:

Histogram[StringLength[WordList[Language -> "French"]]]

Histogram[StringLength[WordList[Language -> "French"]]]

Notice that the word lengths tend to be longer in French. We could investigate whether this is why documents tend to be longer in French than in English, or how this relates to quantities like entropy for text. (Of course, because this is a computational essay, the reader can rerun the computations in it themselves, say by trying Russian instead of French.)

But as something different, let’s compare languages by comparing their translations for, say, the word “computer”.

Find the translations for “computer” in the 10 most common languages:

Take[WordTranslation["computer", All], 10]

Take[WordTranslation["computer", All], 10]

Find the first translation in each case:

First /@ Take[WordTranslation["computer", All], 10]

First /@ Take[WordTranslation["computer", All], 10]

Arrange common languages in “feature space” based on their translations for “computer”:

FeatureSpacePlot[First /@ Take[WordTranslation["computer", All], 40]]

FeatureSpacePlot[First /@ Take[WordTranslation["computer", All], 40]]

From this plot, we can start to investigate all sorts of structural and historical relationships between languages. But from the point of view of a computational essay, what’s important here is that we’re sharing the exposition between ordinary text, computer input, and output.

The text is saying what the basic point is. Then the input is giving a precise definition of what we want. And the output is showing what’s true about it. But take a look at the input. Even just by looking at the names of the Wolfram Language functions in it, one can get a pretty good idea what it’s talking about. And while the function names are based on English, one can use “code captions” to understand it in another language, say Japanese:

FeatureSpacePlot[First/@Take[WordTranslation["computer",All],40]]

FeatureSpacePlot[First /@ Take[WordTranslation["computer", All], 40]]

But let’s say one doesn’t know about FeatureSpacePlot. What is it? If it was just a word or phrase in English, we might be able to look in a dictionary, but there wouldn’t be a precise answer. But a function in the Wolfram Language is always precisely defined. And to know what it does we can start by just looking at its documentation. But much more than that, we can just run it ourselves to explicitly see what it does.

FeatureSpacePlot page

And that’s a crucial part of what’s great about computational essays. If you read an ordinary essay, and you don’t understand something, then in the end you really just have to ask the author to find out what they meant. In a computational essay, though, there’s Wolfram Language input that precisely and unambiguously specifies everything—and if you want to know what it means, you can just run it and explore any detail of it on your computer, automatically and without recourse to anything like a discussion with the author.

Practicalities

How does one actually create a computational essay? With the technology stack we have, it’s very easy—mainly thanks to the concept of notebooks that we introduced with the first version of Mathematica all the way back in 1988. A notebook is a structured document that mixes cells of text together with cells of Wolfram Language input and output, including graphics, images, sounds, and interactive content:

A typical notebook

In modern times one great (and very hard to achieve!) thing is that full Wolfram Notebooks run seamlessly across desktop, cloud and mobile. You can author a notebook in the native Wolfram Desktop application (Mac, Windows, Linux)—or on the web through any web browser, or on mobile through the Wolfram Cloud app. Then you can share or publish it through the Wolfram Cloud, and get access to it on the web or on mobile, or download it to desktop or, now, iOS devices.

Notebook environments

Sometimes you want the reader of a notebook just to look at it, perhaps opening and closing groups of cells. Sometimes you also want them to be able to operate the interactive elements. And sometimes you want them to be able to edit and run the code, or maybe modify the whole notebook. And the crucial point is that all these things are easy to do with the cloud-desktop-mobile system we’ve built.

A New Form of Student Work

Computational essays are great for students to read, but they’re also great for students to write. Most of the current modalities for student work are remarkably old. Write an essay. Give a math derivation. These have been around for millennia. Not that there’s anything wrong with them. But now there’s something new: write a computational essay. And it’s wonderfully educational.

A computational essay is in effect an intellectual story told through a collaboration between a human author and a computer. The computer acts like a kind of intellectual exoskeleton, letting you immediately marshall vast computational power and knowledge. But it’s also an enforcer of understanding. Because to guide the computer through the story you’re trying to tell, you have to understand it yourself.

When students write ordinary essays, they’re typically writing about content that in some sense “already exists” (“discuss this passage”; “explain this piece of history”; …). But in doing computation (at least with the Wolfram Language) it’s so easy to discover new things that computational essays will end up with an essentially inexhaustible supply of new content, that’s never been seen before. Students will be exploring and discovering as well as understanding and explaining.

When you write a computational essay, the code in your computational essay has to produce results that fit with the story you’re telling. It’s not like you’re doing a mathematical derivation, and then some teacher tells you you’ve got the wrong answer. You can immediately see what your code does, and whether it fits with the story you’re telling. If it doesn’t, well then maybe your code is wrong—or maybe your story is wrong.

What should the actual procedure be for students producing computational essays? At this year’s Wolfram Summer School we did the experiment of asking all our students to write a computational essay about anything they knew about. We ended up with 72 interesting essays—exploring a very wide range of topics.

In a more typical educational setting, the “prompt” for a computational essay could be something like “What is the typical length of a word in English” or “Explore word lengths in English”.

There’s also another workflow I’ve tried. As the “classroom” component of a class, do livecoding (or a live experiment). Create or discover something, with each student following along by doing their own computations. At the end of the class, each student will have a notebook they made. Then have their “homework” be to turn that notebook into a computational essay that explains what was done.

And in my experience, this ends up being a very good exercise—that really tests and cements the understanding students have. But there’s also something else: when students have created a computational essay, they have something they can keep—and directly use—forever.

And this is one of the great general features of computational essays. When students write them, they’re in effect creating a custom library of computational tools for themselves—that they’ll be in a position to immediately use at any time in the future. It’s far too common for students to write notes in a class, then never refer to them again. Yes, they might run across some situation where the notes would be helpful. But it’s often hard to motivate going back and reading the notes—not least because that’s only the beginning; there’s still the matter of implementing whatever’s in the notes.

But the point is that with a computational essay, once you’ve found what you want, the code to implement it is right there—immediately ready to be applied to whatever has come up.

Any Subject You Want

What can computational essays be about? Almost anything! I’ve often said that for any field of study X (from archaeology to zoology), there either is now, or soon will be, a “computational X”. And any “computational X” can immediately be explored and explained using computational essays.

But even when there isn’t a clear “computational X” yet,  computational essays can still be a powerful way to organize and present material. In some sense, the very fact that a sequence of computations are typically needed to “tell the story” in an essay helps define a clear backbone for the whole essay. In effect, the structured nature of the computational presentation helps suggest structure for the narrative—making it easier for students (and others) to write essays that are easy to read and understand.

But what about actual subject matter? Well, imagine you’re studying history—say the history of the English Civil War. Well, conveniently, the Wolfram Language has a lot of knowledge about history (as about so many other things) built in. So you can present the English Civil War through a kind of dialog with it. For example, you can ask it for the geography of battles:

GeoListPlot[\!\(\*
NamespaceBox["LinguisticAssistant",
DynamicModuleBox[{Typeset`query$$ = "English Civil War", 
      Typeset`boxes$$ = TemplateBox[{"\"English Civil War\"", 
RowBox[{"Entity", "[", 
RowBox[{"\"MilitaryConflict\"", ",", "\"EnglishCivilWar\""}], "]"}], 
        "\"Entity[\\\"MilitaryConflict\\\", \
\\\"EnglishCivilWar\\\"]\"", "\"military conflict\""}, "Entity"], 
      Typeset`allassumptions$$ = {{
       "type" -> "Clash", "word" -> "English Civil War", 
        "template" -> "Assuming \"${word}\" is ${desc1}. Use as \
${desc2} instead", "count" -> "3", 
        "Values" -> {{
          "name" -> "MilitaryConflict", 
           "desc" -> "a military conflict", 
           "input" -> "*C.English+Civil+War-_*MilitaryConflict-"}, {
          "name" -> "Word", "desc" -> "a word", 
           "input" -> "*C.English+Civil+War-_*Word-"}, {
          "name" -> "HistoricalEvent", "desc" -> "a historical event",
            "input" -> "*C.English+Civil+War-_*HistoricalEvent-"}}}, {
       "type" -> "SubCategory", "word" -> "English Civil War", 
        "template" -> "Assuming ${desc1}. Use ${desc2} instead", 
        "count" -> "4", 
        "Values" -> {{
          "name" -> "EnglishCivilWar", 
           "desc" -> "English Civil War (1642 - 1651)", 
           "input" -> "*DPClash.MilitaryConflictE.English+Civil+War-_*\
EnglishCivilWar-"}, {
          "name" -> "FirstEnglishCivilWar", 
           "desc" -> "English Civil War (1642 - 1646)", 
           "input" -> "*DPClash.MilitaryConflictE.English+Civil+War-_*\
FirstEnglishCivilWar-"}, {
          "name" -> "SecondEnglishCivilWar", 
           "desc" -> "Second English Civil War", 
           "input" -> "*DPClash.MilitaryConflictE.English+Civil+War-_*\
SecondEnglishCivilWar-"}, {
          "name" -> "ThirdEnglishCivilWar", 
           "desc" -> "Third English Civil War", 
           "input" -> "*DPClash.MilitaryConflictE.English+Civil+War-_*\
ThirdEnglishCivilWar-"}}}}, Typeset`assumptions$$ = {}, 
      Typeset`open$$ = {1, 2}, Typeset`querystate$$ = {
      "Online" -> True, "Allowed" -> True, 
       "mparse.jsp" -> 1.305362`6.5672759594240935, 
       "Messages" -> {}}}, 
DynamicBox[ToBoxes[
AlphaIntegration`LinguisticAssistantBoxes["", 4, Automatic, 
Dynamic[Typeset`query$$], 
Dynamic[Typeset`boxes$$], 
Dynamic[Typeset`allassumptions$$], 
Dynamic[Typeset`assumptions$$], 
Dynamic[Typeset`open$$], 
Dynamic[Typeset`querystate$$]], StandardForm],
ImageSizeCache->{265., {7., 17.}},
TrackedSymbols:>{
        Typeset`query$$, Typeset`boxes$$, Typeset`allassumptions$$, 
         Typeset`assumptions$$, Typeset`open$$, Typeset`querystate$$}],
DynamicModuleValues:>{},
UndoTrackedVariables:>{Typeset`open$$}],
BaseStyle->{"Deploy"},
DeleteWithContents->True,
Editable->False,
SelectWithContents->True]\)["Battles"]]

You could ask for a timeline of the beginning of the war (you don’t need to say “first 15 battles”, because if one cares, one can just read that from the Wolfram Language code):

TimelinePlot[Take[\!\(\*
NamespaceBox["LinguisticAssistant",
DynamicModuleBox[{Typeset`query$$ = "English Civil War", 
       Typeset`boxes$$ = TemplateBox[{"\"English Civil War\"", 
RowBox[{"Entity", "[", 
RowBox[{"\"MilitaryConflict\"", ",", "\"EnglishCivilWar\""}], "]"}], 
         "\"Entity[\\\"MilitaryConflict\\\", \\\"EnglishCivilWar\\\"]\
\"", "\"military conflict\""}, "Entity"], 
       Typeset`allassumptions$$ = {{
        "type" -> "Clash", "word" -> "English Civil War", 
         "template" -> "Assuming \"${word}\" is ${desc1}. Use as \
${desc2} instead", "count" -> "3", 
         "Values" -> {{
           "name" -> "MilitaryConflict", 
            "desc" -> "a military conflict", 
            "input" -> "*C.English+Civil+War-_*MilitaryConflict-"}, {
           "name" -> "Word", "desc" -> "a word", 
            "input" -> "*C.English+Civil+War-_*Word-"}, {
           "name" -> "HistoricalEvent", 
            "desc" -> "a historical event", 
            "input" -> "*C.English+Civil+War-_*HistoricalEvent-"}}}, {
        "type" -> "SubCategory", "word" -> "English Civil War", 
         "template" -> "Assuming ${desc1}. Use ${desc2} instead", 
         "count" -> "4", 
         "Values" -> {{
           "name" -> "EnglishCivilWar", 
            "desc" -> "English Civil War (1642 - 1651)", 
            "input" -> "*DPClash.MilitaryConflictE.English+Civil+War-_\
*EnglishCivilWar-"}, {
           "name" -> "FirstEnglishCivilWar", 
            "desc" -> "English Civil War (1642 - 1646)", 
            "input" -> "*DPClash.MilitaryConflictE.English+Civil+War-_\
*FirstEnglishCivilWar-"}, {
           "name" -> "SecondEnglishCivilWar", 
            "desc" -> "Second English Civil War", 
            "input" -> "*DPClash.MilitaryConflictE.English+Civil+War-_\
*SecondEnglishCivilWar-"}, {
           "name" -> "ThirdEnglishCivilWar", 
            "desc" -> "Third English Civil War", 
            "input" -> "*DPClash.MilitaryConflictE.English+Civil+War-_\
*ThirdEnglishCivilWar-"}}}}, Typeset`assumptions$$ = {}, 
       Typeset`open$$ = {1, 2}, Typeset`querystate$$ = {
       "Online" -> True, "Allowed" -> True, 
        "mparse.jsp" -> 1.305362`6.5672759594240935, 
        "Messages" -> {}}}, 
DynamicBox[ToBoxes[
AlphaIntegration`LinguisticAssistantBoxes["", 4, Automatic, 
Dynamic[Typeset`query$$], 
Dynamic[Typeset`boxes$$], 
Dynamic[Typeset`allassumptions$$], 
Dynamic[Typeset`assumptions$$], 
Dynamic[Typeset`open$$], 
Dynamic[Typeset`querystate$$]], StandardForm],
ImageSizeCache->{275., {7., 17.}},
TrackedSymbols:>{
         Typeset`query$$, Typeset`boxes$$, Typeset`allassumptions$$, 
          Typeset`assumptions$$, Typeset`open$$, 
          Typeset`querystate$$}],
DynamicModuleValues:>{},
UndoTrackedVariables:>{Typeset`open$$}],
BaseStyle->{"Deploy"},
DeleteWithContents->True,
Editable->False,
SelectWithContents->True]\)["Battles"], 15]]

You could start looking at how armies moved, or who won and who lost at different points. At first, you can write a computational essay in which the computations are basically just generating custom infographics to illustrate your narrative. But then you can go further—and start really doing “computational history”. You can start to compute various statistical measures of the progress of the war. You can find ways to quantitatively compare it to other wars, and so on.

Can you make a “computational essay” about art? Absolutely. Maybe about art history. Pick 10 random paintings by van Gogh:




van Gogh paintings output

EntityValue[RandomSample[\!\(\*
NamespaceBox["LinguisticAssistant",
DynamicModuleBox[{Typeset`query$$ = "van gogh", Typeset`boxes$$ = 
       TemplateBox[{"\"Vincent van Gogh\"", 
RowBox[{"Entity", "[", 
RowBox[{"\"Person\"", ",", "\"VincentVanGogh::9vq62\""}], "]"}], 
         "\"Entity[\\\"Person\\\", \\\"VincentVanGogh::9vq62\\\"]\"", 
         "\"person\""}, "Entity"], 
       Typeset`allassumptions$$ = {{
        "type" -> "Clash", "word" -> "van gogh", 
         "template" -> "Assuming \"${word}\" is ${desc1}. Use as \
${desc2} instead", "count" -> "4", 
         "Values" -> {{
           "name" -> "Person", "desc" -> "a person", 
            "input" -> "*C.van+gogh-_*Person-"}, {
           "name" -> "Movie", "desc" -> "a movie", 
            "input" -> "*C.van+gogh-_*Movie-"}, {
           "name" -> "SolarSystemFeature", 
            "desc" -> "a solar system feature", 
            "input" -> "*C.van+gogh-_*SolarSystemFeature-"}, {
           "name" -> "Word", "desc" -> "a word", 
            "input" -> "*C.van+gogh-_*Word-"}}}}, 
       Typeset`assumptions$$ = {}, Typeset`open$$ = {1, 2}, 
       Typeset`querystate$$ = {
       "Online" -> True, "Allowed" -> True, 
        "mparse.jsp" -> 0.472412`6.125865914333281, 
        "Messages" -> {}}}, 
DynamicBox[ToBoxes[
AlphaIntegration`LinguisticAssistantBoxes["", 4, Automatic, 
Dynamic[Typeset`query$$], 
Dynamic[Typeset`boxes$$], 
Dynamic[Typeset`allassumptions$$], 
Dynamic[Typeset`assumptions$$], 
Dynamic[Typeset`open$$], 
Dynamic[Typeset`querystate$$]], StandardForm],
ImageSizeCache->{227., {7., 17.}},
TrackedSymbols:>{
         Typeset`query$$, Typeset`boxes$$, Typeset`allassumptions$$, 
          Typeset`assumptions$$, Typeset`open$$, 
          Typeset`querystate$$}],
DynamicModuleValues:>{},
UndoTrackedVariables:>{Typeset`open$$}],
BaseStyle->{"Deploy"},
DeleteWithContents->True,
Editable->False,
SelectWithContents->True]\)["NotableArtworks"], 10], "Image"]

Then look at what colors they use (a surprisingly narrow selection):

ChromaticityPlot[%]

ChromaticityPlot[%]

Or maybe one could write a computational essay about actually creating art, or music.

What about science? You could rediscover Kepler’s laws by looking at properties of planets:

\!\(\*
NamespaceBox["LinguisticAssistant",
DynamicModuleBox[{Typeset`query$$ = "planets", Typeset`boxes$$ = 
     TemplateBox[{"\"planets\"", 
RowBox[{"EntityClass", "[", 
RowBox[{"\"Planet\"", ",", "All"}], "]"}], 
       "\"EntityClass[\\\"Planet\\\", All]\"", "\"planets\""}, 
      "EntityClass"], 
     Typeset`allassumptions$$ = {{
      "type" -> "Clash", "word" -> "planets", 
       "template" -> "Assuming \"${word}\" is ${desc1}. Use as \
${desc2} instead", "count" -> "4", 
       "Values" -> {{
         "name" -> "PlanetClass", "desc" -> " referring to planets", 
          "input" -> "*C.planets-_*PlanetClass-"}, {
         "name" -> "ExoplanetClass", 
          "desc" -> " referring to exoplanets", 
          "input" -> "*C.planets-_*ExoplanetClass-"}, {
         "name" -> "MinorPlanetClass", 
          "desc" -> " referring to minor planets", 
          "input" -> "*C.planets-_*MinorPlanetClass-"}, {
         "name" -> "Word", "desc" -> "a word", 
          "input" -> "*C.planets-_*Word-"}}}}, 
     Typeset`assumptions$$ = {}, Typeset`open$$ = {1, 2}, 
     Typeset`querystate$$ = {
     "Online" -> True, "Allowed" -> True, 
      "mparse.jsp" -> 0.400862`6.054539882441674, "Messages" -> {}}}, 
DynamicBox[ToBoxes[
AlphaIntegration`LinguisticAssistantBoxes["", 4, Automatic, 
Dynamic[Typeset`query$$], 
Dynamic[Typeset`boxes$$], 
Dynamic[Typeset`allassumptions$$], 
Dynamic[Typeset`assumptions$$], 
Dynamic[Typeset`open$$], 
Dynamic[Typeset`querystate$$]], StandardForm],
ImageSizeCache->{171., {7., 17.}},
TrackedSymbols:>{
       Typeset`query$$, Typeset`boxes$$, Typeset`allassumptions$$, 
        Typeset`assumptions$$, Typeset`open$$, Typeset`querystate$$}],
DynamicModuleValues:>{},
UndoTrackedVariables:>{Typeset`open$$}],
BaseStyle->{"Deploy"},
DeleteWithContents->True,
Editable->False,
SelectWithContents->True]\)[{"DistanceFromSun", "OrbitPeriod"}]

ListLogLogPlot[%]

ListLogLogPlot[%]

Maybe you could go on and check it for exoplanets. Or you could start solving the equations of motion for planets.

You could look at biology. Here’s the first beginning of the reference sequence for the human mitochondrion:

GenomeData[{"Mitochondrion", {1, 150}}]

GenomeData[{"Mitochondrion", {1, 150}}]

You can start off breaking it into possible codons:

StringPartition[%, 3]

StringPartition[%, 3]

There’s an immense amount of data about all kinds of things built into the Wolfram Language. But there’s also the Wolfram Data Repository, which contains all sorts of specific datasets. Like here’s a map of state fairgrounds in the US:

GeoListPlot[  ResourceData["U.S. State Fairgrounds"][All, "GeoPosition"]]

GeoListPlot[
 ResourceData["U.S. State Fairgrounds"][All, "GeoPosition"]]

And here’s a word cloud of the constitutions of countries that have been enacted since 2010:

WordCloud[
 StringJoin[
  Normal[ResourceData["World Constitutions"][
    Select[#YearEnacted > \!\(\*
NamespaceBox["LinguisticAssistant",
DynamicModuleBox[{Typeset`query$$ = "year 2010", Typeset`boxes$$ = 
           RowBox[{"DateObject", "[", 
RowBox[{"{", "2010", "}"}], "]"}], 
           Typeset`allassumptions$$ = {{
            "type" -> "MultiClash", "word" -> "", 
             "template" -> "Assuming ${word1} is referring to \
${desc1}. Use \"${word2}\" as ${desc2}.", "count" -> "2", 
             "Values" -> {{
               "name" -> "PseudoTokenYear", "word" -> "year 2010", 
                "desc" -> "a year", 
                "input" -> "*MC.year+2010-_*PseudoTokenYear-"}, {
               "name" -> "Unit", "word" -> "year", "desc" -> "a unit",
                 "input" -> "*MC.year+2010-_*Unit-"}}}}, 
           Typeset`assumptions$$ = {}, Typeset`open$$ = {1}, 
           Typeset`querystate$$ = {
           "Online" -> True, "Allowed" -> True, 
            "mparse.jsp" -> 0.542662`6.186074404594303, 
            "Messages" -> {}}}, 
DynamicBox[ToBoxes[
AlphaIntegration`LinguisticAssistantBoxes["", 4, Automatic, 
Dynamic[Typeset`query$$], 
Dynamic[Typeset`boxes$$], 
Dynamic[Typeset`allassumptions$$], 
Dynamic[Typeset`assumptions$$], 
Dynamic[Typeset`open$$], 
Dynamic[Typeset`querystate$$]], StandardForm],
ImageSizeCache->{86., {7., 18.}},
TrackedSymbols:>{
             Typeset`query$$, Typeset`boxes$$, 
              Typeset`allassumptions$$, Typeset`assumptions$$, 
              Typeset`open$$, Typeset`querystate$$}],
DynamicModuleValues:>{},
UndoTrackedVariables:>{Typeset`open$$}],
BaseStyle->{"Deploy"},
DeleteWithContents->True,
Editable->False,
SelectWithContents->True]\) &], "Text"]]]]

Quite often one’s interested in dealing not with public data, but with some kind of local data. One convenient source of this is the Wolfram Data Drop. In an educational setting, particular databins (or cloud objects in general) can be set so that they can be read (and/or added to) by some particular group. Here’s a databin that I accumulate for myself, showing my heart rate through the day. Here it is for today:

DataListPlot[TimeSeries[Databin[

DateListPlot[TimeSeries[YourDatabinHere]]

Of course, it’s easy to make a histogram too:

Histogram[TimeSeries[Databin[

Histogram[TimeSeries[YourDatabinHere]]

What about math? A key issue in math is to understand why things are true. The traditional approach to this is to give proofs. But computational essays provide an alternative. The nature of the steps in them is different—but the objective is the same: to show what’s true and why.

As a very simple example, let’s look at primes. Here are the first 50:

Table[Prime[n], {n, 50}]

Table[Prime[n], {n, 50}]

Let’s find the remainder mod 6 for all these primes:

Mod[Table[Prime[n], {n, 50}], 6]

Mod[Table[Prime[n], {n, 50}], 6]

But why do only 1 and 5 occur (well, after the trivial cases of the primes 2 and 3)? We can see this by computation. Any number can be written as 6n+k for some n and k:

Table[6 n + k, {k, 0, 5}]

Table[6 n + k, {k, 0, 5}]

But if we factor numbers written in this form, we’ll see that 6n+1 and 6n+5 are the only ones that don’t have to be multiples:

Factor[%]

Factor[%]

What about computer science? One could for example write a computational essay about implementing Euclid’s algorithm, studying its running time, and so on.

Define a function to give all steps in Euclid’s algorithm:

gcdlist[a_, b_] :=   NestWhileList[{Last[#], Apply[Mod, #]} &, {a, b}, Last[#] != 0 &, 1]

gcdlist[a_, b_] := 
 NestWhileList[{Last[#], Apply[Mod, #]} &, {a, b}, Last[#] != 0 &, 1]

Find the distribution of running lengths for the algorithm for numbers up to 200:

Histogram[Flatten[Table[Length[gcdlist[i, j]], {i, 200}, {j, 200}]]]

Histogram[Flatten[Table[Length[gcdlist[i, j]], {i, 200}, {j, 200}]]]

Or in modern times, one could explore machine learning, starting, say, by making a feature space plot of part of the MNIST handwritten digits dataset:

FeatureSpacePlot[RandomSample[Keys[ResourceData["MNIST"]], 50]]

FeatureSpacePlot[RandomSample[Keys[ResourceData["MNIST"]], 50]]

If you wanted to get deeper into software engineering, you could write a computational essay about the HTTP protocol. This gets an HTTP response from a site:

URLRead["https://www.wolframalpha.com"]

URLRead["https://www.wolfram.com"]

And this shows the tree structure of the elements on the webpage at that URL:

TreeForm[Import["http://www.wolframalpha.com", {"HTML", "XMLObject"}],   VertexLabeling -> False, AspectRatio -> 1/2]

TreeForm[Import["http://www.wolframalpha.com", {"HTML", "XMLObject"}],
  VertexLabeling -> False, AspectRatio -> 1/2]

Or—in a completely different direction—you could talk about anatomy:

AnatomyPlot3D[left foot]

AnatomyPlot3D[Entity["AnatomicalStructure", "LeftFoot"]]

What Makes a Good Computational Essay?

As far as I’m concerned, for a computational essay to be good, it has to be as easy to understand as possible. The format helps quite a lot, of course. Because a computational essay is full of outputs (often graphical) that are easy to skim, and that immediately give some impression of what the essay is trying to say. It also helps that computational essays are structured documents, that deliver information in well-encapsulated pieces.

But ultimately it’s up to the author of a computational essay to make it clear. But another thing that helps is that the nature of a computational essay is that it must have a “computational narrative”—a sequence of pieces of code that the computer can execute to do what’s being discussed in the essay. And while one might be able to write an ordinary essay that doesn’t make much sense but still sounds good, one can’t ultimately do something like that in a computational essay. Because in the end the code is the code, and actually has to run and do things.

So what can go wrong? Well, like English prose, Wolfram Language code can be unnecessarily complicated, and hard to understand. In a good computational essay, both the ordinary text, and the code, should be as simple and clean as possible. I try to enforce this for myself by saying that each piece of input should be at most one or perhaps two lines long—and that the caption for the input should always be just one line long. If I’m trying to do something where the core of it (perhaps excluding things like display options) takes more than a line of code, then I break it up, explaining each line separately.

Another important principle as far as I’m concerned is: be explicit. Don’t have some variable that, say, implicitly stores a list of words. Actually show at least part of the list, so people can explicitly see what it’s like. And when the output is complicated, find some tabulation or visualization that makes the features you’re interested in obvious. Don’t let the “key result” be hidden in something that’s tucked away in the corner; make sure the way you set things up makes it front and center.

Use the structured nature of notebooks. Break up computational essays with section headings, again helping to make them easy to skim. I follow the style of having a “caption line” before each input. Don’t worry if this somewhat repeats what a paragraph of text has said; consider the caption something that someone who’s just “looking at the pictures” might read to understand what a picture is of, before they actually dive into the full textual narrative.

The technology of Wolfram Notebooks makes it straightforward to put in interactive elements, like Manipulate, into computational essays. And sometimes this is very helpful, and perhaps even essential. But interactive elements shouldn’t be overused. Because whenever there’s an element that requires interaction, this reduces the ability to skim the essay.

Sometimes there’s a fair amount of data—or code—that’s needed to set up a particular computational essay. The cloud is very useful for handling this. Just deploy the data (or code) to the Wolfram Cloud, and set appropriate permissions so it can automatically be read whenever the code in your essay is executed.

Notebooks also allow “reverse closing” of cells—allowing an output cell to be immediately visible, even though the input cell that generated it is initially closed. This kind of hiding of code should generally be avoided in the body of a computational essay, but it’s sometimes useful at the beginning or end of an essay, either to give an indication of what’s coming, or to include something more advanced where you don’t want to go through in detail how it’s made.

OK, so if a computational essay is done, say, as homework, how can it be assessed? A first, straightforward question is: does the code run? And this can be determined pretty much automatically. Then after that, the assessment process is very much like it would be for an ordinary essay. Of course, it’s nice and easy to add cells into a notebook to give comments on what’s there. And those cells can contain runnable code—that for example can take results in the essay and process or check them.

Are there principles of good computational essays? Here are a few candidates:

0. Understand what you’re talking about (!)

1. Find the most straightforward and direct way to represent your subject matter

2. Keep the core of each piece of Wolfram Language input to a line or two

3. Use explicit visualization or other information presentation as much as possible

4. Try to make each input+caption independently understandable

5. Break different topics or directions into different subsections

Learning the Language

At the core of computational essays is the idea of expressing computational thoughts using the Wolfram Language. But to do that, one has to know the language. Now, unlike human languages, the Wolfram Language is explicitly designed (and, yes, that’s what I’ve been doing for the past 30+ years) to follow definite principles and to be as easy to learn as possible. But there’s still learning to be done.

One feature of the Wolfram Language is that—like with human languages—it’s typically easier to read than to write. And that means that a good way for people to learn what they need to be able to write computational essays is for them first to read a bunch of essays. Perhaps then they can start to modify those essays. Or they can start creating “notes essays”, based on code generated in livecoding or other classroom sessions.

As people get more fluent in writing the Wolfram Language, something interesting happens: they start actually expressing themselves in the language, and using Wolfram Language input to carry significant parts of the narrative in a computational essay.

When I was writing An Elementary Introduction to the Wolfram Language (which itself is written in large part as a sequence of computational essays) I had an interesting experience. Early in the book, it was decently easy to explain computational exercises in English (“Make a table of the first 10 squares”). But a little later in the book, it became a frustrating process.

It was easy to express what I wanted in the Wolfram Language. But to express it in English was long and awkward (and had a tendency of sounding like legalese). And that’s the whole point of using the Wolfram Language, and the reason I’ve spent 30+ years building it: because it provides a better, crisper way to express computational thoughts.

It’s sometimes said of human languages that the language you use determines how you think. It’s not clear how true this is of human languages. But it’s absolutely true of computer languages. And one of the most powerful things about the Wolfram Language is that it helps one formulate clear computational thinking.

Traditional computer languages are about writing code that describes the details of what a computer should do. The point of the Wolfram Language is to provide something much higher level—that can immediately talk about things in the world, and that can allow people as directly as possible to use it as a medium of computational thinking. And in a sense that’s what makes a good computational essay possible.

The Long Path to Computational Essays

Now that we have full-fledged computational essays, I realize I’ve been on a path towards them for nearly 40 years. At first I was taking interactive computer output and Scotch-taping descriptions into it:

Interactive computer output sketch

By 1981, when I built SMP, I was routinely writing documents that interspersed code and explanations:

Code interspersed with explanations

But it was only in 1986, when I started documenting what became Mathematica and the Wolfram Language, that I started seriously developing a style close to what I now favor for computational essays:

Wolfram Language Version 1 documentation

And with the release of Mathematica 1.0 in 1988 came another critical element: the invention of Wolfram Notebooks. Notebooks arrived in a form at least superficially very similar to the way they are today (and already in many ways more sophisticated than the imitations that started appearing 25+ years later!): collections of cells arranged into groups, and capable of containing text, executable code, graphics, etc.

Early Mac notebooks

At first notebooks were only possible on Mac and NeXT computers. A few years later they were extended to Microsoft Windows and X Windows (and later, Linux). But immediately people started using notebooks both to provide reports about they’d done, and to create rich expository and educational material. Within a couple of years, there started to be courses based on notebooks, and books printed from notebooks, with interactive versions available on CD-ROM at the back:

Notebook publication example

So in a sense the raw material for computational essays already existed by the beginning of the 1990s. But to really make computational essays come into their own required the development of the cloud—as well as the whole broad range of computational knowledge that’s now part of the Wolfram Language.

By 1990 it was perfectly possible to create a notebook with a narrative, and people did it, particularly about topics like mathematics. But if there was real-world data involved, things got messy. One had to make sure that whatever was needed was appropriately available from a distribution CD-ROM or whatever. We created a Player for notebooks very early, that was sometimes distributed with notebooks.

But in the last few years, particularly with the development of the Wolfram Cloud, things have gotten much more streamlined. Because now you can seamlessly store things in the cloud and use them anywhere. And you can work directly with notebooks in the cloud, just using a web browser. In addition, thanks to lots of user-assistance innovations (including natural language input), it’s become even easier to write in the Wolfram Language—and there’s ever more that can be achieved by doing so.

And the important thing that I think has now definitively happened is that it’s become lightweight enough to produce a good computational essay that it makes sense to do it as something routine—either professionally in writing reports, or as a student doing homework.

Ancient Educational History

The idea of students producing computational essays is something new for modern times, made possible by a whole stack of current technology. But there’s a curious resonance with something from the distant past. You see, if you’d learned a subject like math in the US a couple of hundred years ago, a big thing you’d have done is to create a so-called ciphering book—in which over the course of several years you carefully wrote out the solutions to a range of problems, mixing explanations with calculations. And the idea then was that you kept your ciphering book for the rest of your life, referring to it whenever you needed to solve problems like the ones it included.

Well, now, with computational essays you can do very much the same thing. The problems you can address are vastly more sophisticated and wide-ranging than you could reach with hand calculation. But like with ciphering books, you can write computational essays so they’ll be useful to you in the future—though now you won’t have to imitate calculations by hand; instead you’ll just edit your computational essay notebook and immediately rerun the Wolfram Language inputs in it.

I actually only learned about ciphering books quite recently. For about 20 years I’d had essentially as an artwork a curious handwritten notebook (created in 1818, it says, by a certain George Lehman, apparently of Orwigsburg, Pennsylvania), with pages like this:

Ciphering book

I now know this is a ciphering book—that on this page describes how to find the “height of a perpendicular object… by having the length of the shadow given”. And of course I can’t resist a modern computational essay analog, which, needless to say, can be a bit more elaborate.

Find the current position of the Sun as azimuth, altitude:

SunPosition[]

SunPosition[]

Find the length of a shadow for an object of unit height:

1/Tan[SunPosition[][[2]]]

1/Tan[SunPosition[][[2]]]

Given a 10-ft shadow, find the height of the object that made it:

Tan[SunPosition[][[2]]]10ft

Tan[SunPosition[][[2]]]10ft

The Path Ahead

I like writing textual essays (such as blog posts!). But I like writing computational essays more. Because at least for many of the things I want to communicate, I find them a purer and more efficient way to do it. I could spend lots of words trying to express an idea—or I can just give a little piece of Wolfram Language input that expresses the idea very directly and shows how it works by generating (often very visual) output with it.

When I wrote my big book A New Kind of Science (from 1991 to 2002), neither our technology nor the world was quite ready for computational essays in the form in which they’re now possible. My research for the book filled thousands of Wolfram Notebooks. But when it actually came to putting together the book, I just showed the results from those notebooks—including a little of the code from them in notes at the back of the book.

But now the story of the book can be told in computational essays—that I’ve been starting to produce. (Just for fun, I’ve been livestreaming some of the work I’m doing to create these.)  And what’s very satisfying is just how clearly and crisply the ideas in the book can be communicated in computational essays.

There is so much potential in computational essays. And indeed we’re now starting the project of collecting “topic explorations” that use computational essays to explore a vast range of topics in unprecedentedly clear and direct ways. It’ll be something like our Wolfram Demonstrations Project (that now has 11,000+ Wolfram Language–powered Demonstrations). Here’s a typical example I wrote:

The Central Limit Theorem

Computational essays open up all sorts of new types of communication. Research papers that directly present computational experiments and explorations. Reports that describe things that have been found, but allow other cases to be immediately explored. And, of course, computational essays define a way for students (and others) to very directly and usefully showcase what they’ve learned.

There’s something satisfying about both writing—and reading—computational essays. It’s as if in communicating ideas we’re finally able to go beyond pure human effort—and actually leverage the power of computation. And for me, having built the Wolfram Language to be a computational communication language, it’s wonderful to see how it can be used to communicate so effectively in computational essays.

It’s so nice when I get something sent to me as a well-formed computational essay. Because I immediately know that I’m going to get a straight story that I can actually understand. There aren’t going to be all sorts of missing sources and hidden assumptions; there’s just going to be Wolfram Language input that stands alone, and that I can take out and study or run for myself.

The modern world of the web has brought us a few new formats for communication—like blogs, and social media, and things like Wikipedia. But all of these still follow the basic concept of text + pictures that’s existed since the beginning of the age of literacy. With computational essays we finally have something new—and it’s going to be exciting to see all the things it makes possible.

]]>
7
Michael Gammon <![CDATA[Are All Fish the Same Shape if You Stretch Them? The Victorian Tale of On Growth and Form]]> http://blog.internal.stephenwolfram.com/?p=14647 2017-10-30T18:22:03Z 2017-10-25T14:13:18Z Are All Fish the Same Shape If You Stretch Them? The Victorian Tale of On Growth and FormIs there a global theory for the shapes of fishes? It’s the kind of thing I might feel encouraged to ask by my explorations of simple programs and the forms they produce. But for most of the history of biology, it’s not the kind of thing anyone would ever have asked. With one notable exception: [...]]]> Are All Fish the Same Shape If You Stretch Them? The Victorian Tale of On Growth and Form

Is there a global theory for the shapes of fishes? It’s the kind of thing I might feel encouraged to ask by my explorations of simple programs and the forms they produce. But for most of the history of biology, it’s not the kind of thing anyone would ever have asked. With one notable exception: D’Arcy Wentworth Thompson.

And it’s now 100 years since D’Arcy Thompson published the first edition of his magnum opus On Growth and Form—and tried to use ideas from mathematics and physics to discuss global questions of biological growth and form. Probably the most famous pages of his book are the ones about fish shapes:

D'Arcy Thomspon and fish sketches

Stretch one kind of fish, and it looks like another. Yes, without constraints on how you stretch, it’s not quite clear what this is telling one, and I don’t think it’s much. But just to ask the question is interesting, and On Growth and Form is full of interesting questions—together with all manner of curious and interesting answers.

D’Arcy Thompson was in many ways a quintessential British Victorian academic, steeped in the classics, and writing books with titles like A Glossary of Greek Fishes (i.e. how were fish described in classical Greek texts). But he was also a diligent natural scientist, and he became a serious enthusiast of mathematics and physics. And where Aristotle (whom D’Arcy Thompson had translated) used plain language, with perhaps a dash of logic, to try to describe the natural world, D’Arcy Thompson tried to use the language of mathematics and physics.

At Christmas time, according to his daughter, he used to entertain children by drawing pictures of dogs on rubber sheets and stretching them from poodles to dachshunds. But it was not until the age of 57 that he turned such pursuits into the piece of scholarship that is On Growth and Form.

The first edition of the book was published in 1917. In many ways it’s like a catalog of biological forms—a kind of geometrical analog of Aristotle’s books on natural history. It’s particularly big on aquatic life—from plankton to fish. Land animals do make a showing, though mostly as skeletons. And ordinary plants make only specific appearances. But throughout the book the emphasis is on “why does such-and-such a thing have the form or shape it does?”. And over and over again the answer that’s given is: “because it’s following such-and-such a physical phenomenon, or mathematical structure”.

Much of the story of the book is told in its pictures. There are growth curves—of haddock, trees, regenerated tadpole tails, etc. There’s a long discussion of the shapes of cells—and especially their connection with phenomena (like splashes, bubbles and foams) where surface tension is important. There are spirals—described mathematically, and appearing in shells and horns and leaf arrangements. And finally there’s a long discussion of the “theory of transformations”—about how different forms (like the shapes of fishes or primate skulls) might be related by various (mathematically rather undefined) “transformations”.

On Growth and Form pages

In D’Arcy Thompson’s time—as still to a large extent today—the dominant form of explanation in biology is Darwinism: essentially the idea that things are the way they are because they’ve somehow evolved to be that way, in order to maximize some kind of fitness. D’Arcy Thompson didn’t think that was the whole story, or even necessarily the most important part of the story. He thought instead that many natural forms are the way they are because it’s an inevitable feature of the physics of biological tissue, or the mathematics of geometrical forms.

Sometimes his explanations fall a little flat. Leaves aren’t really shaped much like polar plots of trigonometric functions. Jellyfish aren’t convincingly shaped like drops of ink in water. But what he says often rings true. Hexagonal arrangements of cells are like closest geometrical packings of disks. Sheep horns and nautilus shells form logarithmic (equiangular) spirals.

He uses basic geometry and algebra quite a bit—and even sometimes a little combinatorics or topology. But he never goes as far as calculus (and, as it happens, he never learned it), and he never considers ideas like recursive rules or nested structures. But for me—as for quite a few others over the years—D’Arcy Thompson’s book is an important inspiration for the concept that even though biological forms may at first look complicated, there can still be theories and explanations for them.

In modern times, though, there’s a crucial new idea, that D’Arcy Thompson did not have: the idea of using not traditional mathematics and physics, but instead computation and simple programs as a way to describe the rules by which things grow. And—as I discovered in writing my book A New Kind of Science—it’s remarkable to what extent that idea lets us understand the mechanisms by which complex biological forms are produced, and lets us finish the bold initiative that D’Arcy Thompson began a century ago in On Growth and Form.

Who Was D’Arcy Thompson?

D’Arcy Wentworth Thompson was born in Edinburgh on May 5, 1860. His father, who was also named D’Arcy Wentworth Thompson, had been born in 1829, aboard a ship captained by his father, that was transporting convicts to Tasmania. D’Arcy Senior was soon sent to boarding school in England, and eventually studied classics at Cambridge. Though academically distinguished, he was apparently passed over for a fellowship because of perceived eccentricity—and wound up as a (modernizing, if opinionated) schoolteacher in Edinburgh. Once there, he soon met the lively young Fanny Gamgee, daughter of Joseph Gamgee, an early and distinguished veterinary surgeon—and in 1859 they were married.

D’Arcy (junior) was born the next year—but unfortunately his mother contracted an infection during childbirth, and died within the week. The result was that D’Arcy (junior) ended up living with his mother’s parents, taken care of by one of his mother’s sisters. When D’Arcy (junior) was three years old, his father then got a university professorship (of ancient Greek) in Ireland, and moved there. Still, D’Arcy (junior) stayed in close touch with his father through letters, and, later, visits. And indeed his father seems to have doted on him, for example publishing two children’s books dedicated to him:

Fun and Earnest first pages

In a foreshadowing of his later interests, D’Arcy (junior) learned Latin from his father almost as soon as he was able to speak, and was continually exposed to animals of all sorts in the Gamgee household. There was also a certain math/physics theme. D’Arcy Thompson (senior)’s best friend in Edinburgh was Peter Guthrie Tait—a distinguished mathematical physicist (mechanics, thermodynamics, knot theory, …) and friend of Maxwell, Hamilton and Kelvin—and D’Arcy (junior) often hung out at his house. Joseph Gamgee was also engaged in various scientific pursuits, for example publishing the book On Horseshoeing and Lameness based in part on a statistical study he’d done with the then 10-year-old D’Arcy (junior). Meanwhile, D’Arcy Thompson (senior) began to travel, as D’Arcy (junior) would later do, for example visiting Harvard in 1867 to give the Lowell Lectures—which D’Arcy (junior) would also give, in 1936, 69 years later.

At the age of 11, D’Arcy went to the school where his father had previously taught. He did well in academic studies, but also organized a natural history (“Eureka”) club, where he and his friends collected all sorts of specimens. And by the end of his time at school, he published his first paper: the 11-page (with photographs) “Note on Ulendron and Halonia”, describing the regular pattern of growth scars on two kinds of fossil plants.

At 18, D’Arcy started at Edinburgh University as a medical student. His grandfather—while distinguished—was not wealthy, with the result that D’Arcy had to support himself by tutoring Greek and writing articles for the Edinburgh-published Encyclopedia Britannica (the 9th edition, from 1889, contains an extensive article by D’Arcy on John Ray, a British naturalist of the 1600s). But D’Arcy’s real passion at the time was the then-hot field of paleontology, and after two years he abandoned his medical studies—and left to instead study Natural Science at the place his father had been years earlier: Trinity College, Cambridge.

D’Arcy did well at Cambridge, had an interesting circle of friends (including the future co-author of Principia Mathematica, Alfred North Whitehead), and quickly became something of a fixture in the local natural history scene. This led Macmillan & Co. to commission D’Arcy (still an undergraduate) to produce his first book: a translation from German of Hermann Muller’s The Fertilisation of Flowers. The publisher thought that the book—which was a fairly traditional work of descriptive natural history, based in part on observing about 14,000 visits of insects to flowers—would be of popular interest, and (in one of his last published appearances) got no less than Charles Darwin to write a preface for it:

Fertilisation of Flowers 1
Fertilisation of Flowers 2
Fertilisation of Flowers 3
Fertilisation of Flowers 4

At Cambridge, D’Arcy hung out a lot at the new Museum of Zoology, and was particularly influenced by a young professor named Frank Balfour who studied comparative embryology, and for whom a new Department of Animal Morphology was being created—but who died trying to climb Mont Blanc right when D’Arcy was finishing Cambridge.

D’Arcy began to pursue all sorts of projects, giving lectures on topics such as “Aristotle on Cephalopods”, and making detailed studies of “hydroid zoophyte” specimens (aquatic animals like sea anemones that look like plants) brought back from expeditions to the Arctic and Antarctic. He applied for a fellowship in Cambridge, but—like his father before him—didn’t get it.

In 1884, though, the newly created and new-thinking (non-religious, co-ed, young professors, …) University College in Dundee, Scotland, advertised for a professor of biology (yes, combining zoology and botany!). D’Arcy applied, and got the job—with the result that at age 24 he became a professor, a role in which he would remain for nearly 64 years.

D’Arcy the Professor

D’Arcy was immediately popular as a teacher, and continued to do a certain amount of rather dry academic work (in 1885 he published A Bibliography of Protozoa, Sponges, Coelenterata, and Worms, which was, as advertised, a list of about 6000 publications on those subjects between 1861 and 1883). But his real passion was the creation of his own Museum of Zoology, and the accumulation of specimens for it.

He was soon excitedly writing that “within the last week, I have had a porpoise, two mongooses, a small shark, an eel 8ft long… a young ostrich and two bagfuls of monkeys: all dead of course.” His archive (among its 30,000 items) contains extensive evidence of all sorts of trading of specimens from around the world:

Trading notes 1
Trading notes 2
Trading notes 3
Trading notes 4
Trading notes 5
Trading notes 6
Trading notes 7
Trading notes 8

But in Dundee he found a particularly good local source of specimens. Dundee had long been a center of international textile trade, and had also developed a small whaling industry. And when it was discovered that by mixing jute with whale oil it could be turned into fabric, whaling in Dundee grew dramatically.

Some of the hunting they did was local. But whaling ships from Dundee went as far as Canada and Greenland (and once even to Antarctica). And befriending their captains, D’Arcy persuaded them to bring him back specimens (as skeletons, in jars, etc.) from their expeditions—with the result, for example, that his museum rapidly accumulated the best arctic collection around.

The museum always operated on a shoestring budget, and it was typical in 1886 when D’Arcy wrote that he’d personally been “working all day on a baby Ornithorhynchus” (platypus). In his early years as a professor, D’Arcy published only a few papers, mostly on very detailed matters—like the strangely shaped stomach of a type of possum, or the structure of the porpoise larynx, or the correct taxonomic placement of a duck-like dinosaur. And he always followed the prevailing Darwinian paradigm of trying to explain things either by their evolutionary connections, or by their fitness for a particular function.

Museum specimen

The Matter of the Alaskan Seals

In Dundee, D’Arcy joined various local clubs, like the Dundee Naturalists’ Society, the Dundee Working Men’s Field Club, the Homeric Club, and, later, also the Freemasons. He became quite active in university and community affairs, notably campaigning for a medical school (and giving all sorts of statistical evidence for its utility), as well as for education for the local poor. But mostly D’Arcy lived the life of an academic, centered around his teaching and his museum.

Still, as a responsible member of the community, he was called on in various ways, and in 1892, he joined his first government commission—formed to investigate a plague of voles in Scotland (conclusions included: “don’t shoot hawks and owls that eat voles”, and “it’s probably not a good idea to set loose a ‘virus’ to infect the voles”). Then in 1896—at the age of 36—D’Arcy was tapped for a piece of international scientific diplomacy.

It all had to do with seals, and the fur trade based on them. When Russia sold Alaska to the US in 1867 it also sold the rights to the seals which bred on some of the islands in the Bering Sea. But by the 1890s Canadian ships (under British protection) were claiming the right to catch seals in the open ocean—and too many seals were being killed for the population to be maintained. In 1893 a treaty was made to clarify the situation. But in 1896 there was a need to analyze more carefully what was going on (and, yes, to claim what ended up being $10M in damages for Canadian/British sealers).

Lord Salisbury, the British Prime Minister at the time, who happened to be an amateur botanist, knew of D’Arcy and asked him to travel to the Bering Sea to investigate. D’Arcy had by that point traveled a bit around Europe, but this was a complex trip. At first he went to Washington, DC, dropping in at the White House. Then across Canada, and then by Coast Guard ship (and dog sled) to the seals.

North American trip

D’Arcy did well at making friends with his American counterparts (who included the president of the then-a-decade-old Stanford University), and found that at least on the American-controlled islands (the Russian-controlled ones were a different story) seals were being herded a bit like sheep in Scotland, and that though there was “abundant need for care and prudent measures of conservation”, things were basically OK. In Washington, DC, D’Arcy gave a long speech, and helped broker a kind of “seal peace treaty”—that the British government was pleased enough with to give D’Arcy a (medieval-inspired) “Companion of the Bath” honor.

Statesman of Science

Being a professor in Dundee wasn’t a particularly high position in the pecking order of the time. And after his Bering Sea experience, D’Arcy started investigating moving up. He applied for various jobs (for example at the Natural History Museum in London), but perhaps in part because he didn’t have fancier academic credentials (like a PhD)—and also had spent so much of his time organizing things rather than doing research—he never got any of them.

He was nevertheless increasingly sought after as a kind of statesman of science. And in 1898 he was appointed to the Fishery Board for Scotland (a role in which he continued for 43 years), and the next year he was the British delegate to the first International Conference on Oceanography.

D’Arcy was a serious collector of data. He maintained a team of people at the fish market, keeping track of the catches brought in from boats:

Tide log

And then he took this data and created graphics and statistical analyses:

Thompson fish catch graph

And over the years he became well known as a negotiator of fishing rights, both locally and internationally. He was also a collector of oceanographic data. He saw to it that there were detailed tide measurements made:

Oceanographic data

And had the data analyzed and decomposed into harmonic components—much as it is today:

Data analysis

The Scottish government even provided for him a research ship (a steam trawler named the SS Goldseeker) in which he and his students would go around the Scottish coast, measuring ocean properties and collecting specimens.

D’Arcy the Classical Scholar

D’Arcy always had many interests. First and foremost was natural history. But after that came classics. And indeed, back in his undergraduate days, D’Arcy had already started working with his classicist father on translating Aristotle’s works on natural history into English.

One of the complexities of that task, however, was to know what species Aristotle meant by words he used in Greek. And this led D’Arcy into what became a lifelong project—the first output of which was his 1894 book Glossary of Greek Birds:

Greek Birds 1
Greek Birds 2
Greek Birds 3
Greek Birds 4

It’s an interesting exercise—trying to fit together clues to deduce just what modern bird some passage in classical Greek literature was talking about. Often D’Arcy succeeds. Sometimes by using natural history; sometimes by thinking about mythology or about configurations of things like constellations named for birds. But sometimes D’Arcy just has to describe something as “a remarkable bird, of three varieties, of which one croaks like a frog, one bleats like a goat, and the third barks like a dog”—and he doesn’t know the modern equivalent.

Over the years, D’Arcy continued his efforts to translate Aristotle, and finally in 1910 (8 years after his father’s death) he was able to publish what remains to this day the standard translation of Aristotle’s main work on zoology, his History of Animals.

This project established D’Arcy as a classical scholar—and in 1912 he even got an honorary PhD (D.Litt.) at Cambridge on the basis of it. He also began a long association with what’s known as Liddell & Scott, the still-standard dictionary of ancient Greek. (Liddell had been notable for being the father of Alice, of Wonderland fame.)

But D’Arcy’s interests in Greek science extended beyond natural history, and into astronomy and mathematics. D’Arcy explored such things as ancient methods for computing square roots—and also studied Greek geometry.

So in 1889, when D’Arcy was investigating Foraminifera (protozoa that live in sediment or in the ocean and often form spiral shells) he was able to bring his knowledge of Greek mathematics to bear, declaring that “I have taken to Mathematics… and discovered some unsuspected wonders in regard to the Spirals of the Foraminifera!”.

Towards Something Bigger

When he was 41, in 1901, D’Arcy married his stepmother’s niece, the then 29-year-old Ada Maureen Drury (yes, small world that it is, she was named after “Byron’s” Ada, because an ancestor had reputedly been a romantic interest of Byron’s). They bought a small house somewhat outside of town—and between 1902 and 1910 they had three children, all daughters.

By 1910, D’Arcy was 50 years old, and an elder statesman of science. He kept himself busy teaching, managing his museum, doing administrative and government work, and giving public lectures. A typical lecture—given at Oxford in 1913—was entitled “On Aristotle as a Biologist”. It was charming, eloquent, ponderous and Victorian:

Thompson lecture excerpt

In many ways, D’Arcy was first and foremost a collector. He collected natural history specimens. He collected Greek words. He collected academic references—and antiquarian books. And he collected facts and statements—many of which he typed onto index cards, now to be found in his archive:

Index cards 1
Index cards 1

Still, in his role as elder statesman, D’Arcy was called upon to make broad pronouncements. And in many ways the great achievement of the later part of his life was to connect the disparate things he collected and identify common themes that could connect them.

In 1908 he had published (in Nature) a 2-page paper entitled “On the Shapes of Eggs and the Causes Which Determine Them”. In a sense the paper was about the physics of egg formation. And what was significant was that instead of accounting for different egg shapes in terms of their evolutionary fitness, it talked about the physical mechanisms that could produce them.

Three years later D’Arcy gave a speech entitled “Magnalia Naturae: or the Greater Problems of Biology” in which he took this much further, and started discussing “the possibility of… supporting the observed facts of organic form on mathematical principles [so as to make] morphology… a true natural science… justified by its relation to mathematics”.

Magnalia Naturae

In 1910, Cambridge University Press had asked D’Arcy if he’d like to write a book about whales. He said that instead perhaps he should write a “little book” about “The Forms of Organisms” or “Growth and Form”—and he began the process of assembling what would become On Growth and Form. The book had elements that drew on D’Arcy’s whole range of interests. His archives contain some of what went into the assembly of the book, like the original drawings of fish-shape transformations (D’Arcy wasn’t a great sketch artist):

Fish sketches

There were also other, more impressionistic images—like the one illustrating transformations between zebra-related animals (quagga, etc.) or one showing tortoise (?) shell structure:

Sketches 1
Sketches 2

D’Arcy didn’t contact his publisher again for several years, but in 1915—in the middle of World War I—he wrote them again, saying that he had finally finished the book “on a larger scale”, and soon signed a publishing contract (that’s shockingly similar to the way modern ones look):

D'Arcy Thompson publishing contract

It took a couple more years, between D’Arcy’s last-minute changes, and paper shortages associated with the war—but finally in 1917 the book (which had by then swelled to 800 pages) was published.

The Book

On Growth and Form opens with a classic D’Arcy “Prefatory Note”: “This book of mine has little need of preface, for indeed it is ‘all preface’ from beginning to end.” He goes on to apologize for his lack of mathematical skill—and then launches, beginning with a discussion of the relation of the philosophies of Kant and Aristotle on the nature of science.

D'Arcy Thompson prefatory note

The reviews were positive, and surprisingly sensible, with the Times Literary Supplement for example writing:

On Growth and Form review

Further into Mathematics

D’Arcy was 57 years old by the time On Growth and Form was published—and he could have used it as a closing act in his career. But instead it seemed to make him more energetic—and seemed to encourage him to take mathematical methods as a kind of personal theme.

In his study of the shapes of biological cells, D’Arcy had gotten very interested in polyhedra and packings, and particularly in Archimedean solids (such as the tetrakaidecahedron). His archives contain all sorts of investigations of possible packings and their properties, together with actual cardboard polyhedra, still ready to assemble:

Shape sketch 1
Shape sketch 2
Shape sketch 3
Shape sketch 4
Shape sketch 5
Shape sketch 6
Shape sketch 7
Shape sketch 8

D’Arcy extended his interest in number theory, collecting properties of numbers a little like he’d collected so many other things:

Number theory notes 1
Number theory notes 2
Number theory notes 3
Number theory notes 4

He dipped into chemistry, thinking about it in terms of graphs, like those derived from polyhedra:

Chemistry notes 1
Chemistry notes 2

And even when he worked on history, D’Arcy used mathematical thinking, here studying the distribution of when famous people lived, in connection with writing about the Golden Ages:

Golden Age notes 1
Golden Age notes 2

As an administrator he brought in math as well, here analyzing what in today’s world would be called a grading curve—and comparing exam results between different years:

Learning curve 1
Learning curve 2
Learning curve 3

He worked extensively on tides and tide computations. He collected data from harbors. And came up with theories about the various components of tides, some of which turned out to be correct:

Tide notes 1
Tide notes 2
Tide notes 3

The mathematics he used was always a bit circumscribed—and for example he never learned calculus, even to the point of apparently getting confused about growth rates versus finite differences in plots in On Growth and Form. (There seems to be just a single sheet of calculus-like work by him in his archives, and it’s simply an exercise copied without solution from the famous Whittaker & Watson textbook.)

But what about systems based on pure computational rules—of the kind that, for example, I have spent so much time studying? Well, in the archive there are things like this—perhaps a version of a space-filling curve:

Computational rules 1
Computational rules 2

And back from 1897 there’s a curious cardboard object that D’Arcy described as a “reasoning machine”:

Reasoning machine 1
Reasoning machine 2

It’s not completely clear what this was (though its wheel still turns nicely!). It seemed to involve a diagrammatic way of determining the truth value of a logical expression, perhaps following the work of Jevons from a couple of decades earlier. But so far as I can tell it was D’Arcy’s sole excursion into the world of logic and rule-based processes—and he never connected anything like this to biology.

The Later D’Arcy

Before On Growth and Form, D’Arcy had published only quite sporadically. But after it, as he entered his sixties, he began to write prodigiously, publishing all over the place on a wide range of topics. He gave lectures, in person and on the radio. And he also began to receive all sorts of honors (he became Sir D’Arcy in 1937)—and was invited to events all over the world (he did a grand tour of the US in the 1930s, and was also received as a celebrity in places like the Soviet Union).

On Growth and Form was considered a commercial success. Its original print run was 500 copies (of which at least 113 are now in academic libraries around the world), and by 1923 it had sold out. The publisher (Cambridge University Press) wanted to reprint it. But D’Arcy insisted that it needed to be revised—and in the end it took until 1942 before he got the revisions done. The second edition added 300 pages to the book—including photographs of splashes (obtained directly from Harold Edgerton at MIT), analysis of teeth, and patterns on animal coats. But the main elements of the book remained exactly the same.

D’Arcy had published a second edition of his Glossary of Greek Birds in 1936 (more birds, more interpretations), and in 1947, based on notes he started collecting in 1879, he released a kind of sequel: his Glossary of Greek Fishes. (Oxford University Press, in the flap copy for the book, says charmingly that “… it is highly improbable that there is any other scholar who has studied Greek fishes over so long a period as Sir D’Arcy Thompson…”.)

Even into his eighties, D’Arcy continued to travel all over the place—with his archives containing some typical travel documents of the time:

Travel document 1
Travel document 2
Travel document 3
Travel document 4
Travel document 5
Travel document 6

His travel was interrupted by World War II (which is perhaps why the second edition of On Growth and Form finally got finished in 1942). But in 1947, with the war over, at the age of 87, D’Arcy went to India for several months, notably lecturing on the skeletal structure of birds while holding a somewhat impatient live hen in a box. But in India D’Arcy’s health began to fail, and after returning to Scotland, he died in June 1948—to the last corresponding about specimens for his museum.

Thompson biography

Aftermath

D’Arcy’s wife (who seemed in frail health through much of her 47-year marriage to D’Arcy) lived on for only 7 months after his death. None of D’Arcy’s daughters ever married. His oldest daughter Ruth became a music teacher and administrator at a girl’s boarding school, and in 1958 (when she was 56) published a biography of D’Arcy.

His middle daughter Molly moved to South Africa, wrote children’s and travel books, and lived to the age of 101, dying in 2010—while his youngest daughter Barbara wrote a book on healing and herbalism and died in a freak river accident in 1990.

On Growth and Form was D’Arcy’s most notable output, and it has been reprinted many times over the course of a hundred years. The museum D’Arcy created in Dundee was largely dismantled in the 1950s, but has now been to some extent reconstituted, complete with some of the very specimens D’Arcy collected, with labels duly signed “DWT” (yup, that’s me next to the same orangutan as in the old picture of the museum):

Crocodile skull
Specimen jars
Stephen Wolfram with primate skeleton

In 1917 D’Arcy moved from Dundee to the nearby but more distinguished and ancient university in St Andrews, where he took over another museum. It too fell upon hard times, but still exists in a reduced form.

Crab
Birds

And now some of the D’Arcy’s specimens are being 3D-scanned (yes, that’s the same crocodile):

Crocodile skull 3D scan

And on a main street in St. Andrews there’s still a plaque where D’Arcy lived:

Thompson home plaque

What Was D’Arcy Like?

Thompson portrait

D’Arcy had an imposing physical presence. He stood 6’3” and had a large head, on which he often wore a black fedora. He had piercing blue eyes, and in his youth, he had red hair—which he grew into a large beard when he was a young professor. He often wore a long coat, which could sometimes seem moth eaten. Later in his life, he would sometimes walk around town with a parrot on his shoulder.

He was renowned as an engaging speaker and lecturer—known both for his colorful and eloquent content (he could regale the audience with the tale of a walrus he had known, or equally well discuss Aristotle’s year by the seaside), and for the various physical (and biological) demonstrations he would use. Many stories are told of his eccentricities, especially by his former students. It is said, for example, that he once came to give a lecture to his students which began with him pulling a dead frog out of one pocket of his coat—and then a live one out of the other pocket. Despite having spent most of his life in Scotland, he didn’t have a Scottish accent.

He was charming and jolly, and even in his eighties he was given to dancing when he could. He was tactful and diplomatic, if not particularly good at sensing other people’s opinions. He presented himself with a certain modesty (for example always expressing his weakness in mathematics), and—perhaps to his detriment—did little to advocate for himself.

He led a fairly simple life, centered around his work and family. He worked hard, typically until midnight each day. He always liked to learn. He enjoyed children and the young, and would happily play with them. When he walked around town, he was universally recognized (the shoulder parrot helped!). He was happy to chat with anyone, and in later years, he carried candy in his pocket, which he gave out to children he encountered.

D’Arcy was a product of his age, but also of an unusual combination of influences. Like many of the members of his adopted family, D’Arcy aspired to be a scientist. But like his father, he aspired to be a classical scholar. He did diligent and detailed academic work for many years, in natural history, in classics, and in ancient science. But he also enjoyed presentation and lecturing. And it was in large part through his efforts to explain his academic work that he came to make the connections that would lead to On Growth and Form.

What Happened After

If you search the scientific literature today, you’ll find about 4000 publications citing On Growth and Form. Their number relative to the total scientific literature has remained remarkably fairly even over the years (with a peak around the publication of the second edition in 1942, and perhaps a dip in the 1960s when genetics began to dominate biology):

On Growth and Form citations graph

There’s quite a diversity in the topics, as this random sample of titles indicates:

Titles referencing On Growth and Form

Most concern specific biological systems; some are more general. Making word clouds from titles by decade, one sees that “growth” is the dominant theme—though centered in the 1990s there are signs of the discussion that was going on about the “philosophy of evolution”, and the interplay between natural selection and “developmental constraints”:

Word clouds - works referencing On Growth and Form

On Growth and Form has never really become mainstream in biology—or any other field. (It didn’t help that by the 1930s, biology was firmly going off in the direction of biochemistry and later molecular biology.) So how have people found out about On Growth and Form?

Indeed, as I write this, I’m wondering: how did I myself find out about On Growth and Form? I can tell I knew about it by 1983, because I referenced it (somewhat casually) in my first long paper about cellular automata and the patterns they generate. I also know that in 1982 I bought a copy of the (heavily abridged) version of On Growth and Form that was available then. (I was thrilled in 1992 when I chanced upon a complete second edition of On Growth and Form in a used bookstore; I’d never seen the whole book before.)

But how did I first become aware of D’Arcy, and On Growth and Form? My first hypothesis today was that it was in 1977, from the historical notes of Benoit Mandelbrot’s Fractals book (yes, D’Arcy had actually used the term “self-similar”, though only in connection with spirals). Then I thought perhaps it might have been around 1980, from the references to Alan Turing’s 1952 paper on the chemical basis of morphogenesis. I wondered if perhaps it was from hearing about catastrophe theory, and the work of René Thom, in the mid-1970s. But my best guess as of now is that it was actually around 1978, from a little book called Patterns in Nature, by a certain Peter S. Stevens, that heavily references On Growth and Form, and that I happened across in a bookstore.

I’ve almost never seen mentions of Patterns in Nature, but in some ways it’s a simplified and modernized On Growth and Form, full of photographs comparing biological and non-biological systems, together with diagrams about how various structures can be built. But what was the path from D’Arcy to Patterns in Nature? It’s a typical kind of history question that comes up.

The first thing I noticed is that Peter Stevens (born 1936) was trained as an architect, and spent most of his career around Harvard. In his book, he thanks his father, Stanley Stevens (1906–1973), who was a psychoacoustics expert, who was at Harvard from 1932 on, and who organized a “Science of Science” interdisciplinary discussion group there. But recall that D’Arcy visited Harvard to give the Lowell Lectures in 1936. So that’s no doubt how Stevens, Sr. knew about him.

But in any case, from his Harvard connections came, I believe, the references to D’Arcy by evolutionary biologist Stephen J. Gould, and by John Tyler Bonner, who was the person who created the abridged version of On Growth and Form (sadly, omitting for example the chapter on phyllotaxis). I suspect D’Arcy’s influence on Buckminster Fuller also came through Harvard connections. And maybe Benoit Mandelbrot heard about D’Arcy there too. (One would think that with On Growth and Form being out there as a published book, there wouldn’t be need for word-of-mouth communication, but particularly outside of mainstream areas of science, word of mouth remains surprisingly important.)

But what about Turing? How did he know about D’Arcy? Well, I have at least a guess here. D’Arcy had been good friends in high school with a certain John Scott Haldane, who would go on to be a well-known physiology researcher, and who had a son named J. B. S. Haldane, who became a major figure in evolutionary biology and in the presentation of science to the public. Haldane often referenced D’Arcy, and notably introduced him to Peter Medawar (who would win a Nobel Prize for immunology), of whom D’Arcy (in 1944) would say “I do believe that more than any man you have understood what I have tried to say!”.

Both Medawar and evolutionary biologist (and originator of the term “transhumanism”) Julian Huxley encouraged D’Arcy to think about continuity and gradients in connection with his shape transformations (e.g. of fish). I don’t know the whole story, but I suspect these two connected with C. H. Waddington, a developmental biologist (and inventor of the term “epigenetics”) who interacted with Turing in Cambridge. (Small world that it is, Waddington’s mathematician daughter is married to a distinguished mathematician named John Milnor, with whom I discussed D’Arcy in the early 1980s.) And when Turing came to write about morphogenesis in 1952, he referenced D’Arcy (and Waddington), then proceeded to base his theory on (morphogen) gradients.

In another direction, D’Arcy interacted with early mathematical biologists like Alfred Lotka and Vito Volterra and Nicolas Rashevsky. And though their work was heavily based on differential equations (which D’Arcy didn’t really believe in), he took pains to support them when he could.

On Growth and Form also seems to have been popular in the art and architecture community, with people as diverse as the architects Mies van der Rohe, Le Corbusier, the painter Jackson Pollock and the sculptor Henry Moore also mentioning its influence.

Modern Times

So now that it’s been 100 years since On Growth and Form was published, do we finally understand how biological organisms grow? Lots of work has certainly been done at the genetic and molecular scale, and great progress has been made. But when it comes to macroscopic growth, much less has been done. And a large part of the reason, I suspect, is that it’s needed a new paradigm in order to make progress.

D’Arcy’s work was, more than anything, concerned with analogy and (essentially Aristotelian-style) mechanism. He didn’t really pursue traditional “theory” in the sense of the exact sciences. In his day, though, such theory would normally have meant writing down mathematical equations to represent growth, and then solving them to see what would happen.

And the problem is that when one looks at biological forms, they often seem far too complex to be the results of traditional mathematical equations. But starting in the 1950s a new possibility emerged: perhaps one could model biological growth as following not mathematical equations but instead rules like a program for a computer.

And when I started my systematic investigation of the computational universe of possible programs in the early 1980s, I was immediately struck by how “biological” a lot of the forms created, say, by simple cellular automata seemed:

Cellular automata

And this is how I came to study On Growth and Form. I viewed it almost as a catalog of biological forms—that I wondered if one could explain with computational rules. I even started collecting specimens—in a very pale shadow of D’Arcy’s efforts (and with no animal skeletons!):

Seashells 1
Seashells 2

Occasionally I would find one that just seemed to cry out as being from something like a program:

Shell pattern

But more than that, I kept on exploring spaces of possible programs—and discovering that the range of forms they produced seem to align remarkably well with the actual range of forms one sees across biological organisms. (I looked particularly at shell shapes and patterns, as well as other pigmentation patterns, and various forms of plants.)

And in a sense what I found strongly supports a core idea of D’Arcy’s: that the forms of organisms are not so much determined by evolution, as by what it’s possible for processes to produce. D’Arcy thought about physical processes and mathematical forms; 60+ years later I was in a position to explore the more general space of computational processes.

And it so happened that, like D’Arcy, I ended up presenting my main results in a (big) book, that I called A New Kind of Science. My main purpose in the book was to describe what I’d learned from exploring the computational universe. And I devoted two sections (out of 114) respectively to “Growth of Plants and Animals” and “Biological Pigmentation Patterns”—producing something that looks a bit similar to On Growth and Form:

A New Kind of Science pages

So, in the end, what about the fish? Well, I think I’ve managed to understand something about the “morphospace” of possible mollusc shells. And I’ve made a start on leaves—though I’m hoping one of these years to be able to get a lot more data. I’ve also looked at animal skeletons a bit. But, yes, I at least still don’t know about the space of possible fish shapes. Though maybe somewhere inside our image identification neural net (which saw plenty of fish in its training) it already knows. And maybe it agrees with what D’Arcy thought—a hundred years ago.

(For help with facts and materials I’d like to thank Matthew Jarron, Maia Sheridan, Isabella Scott, Special Collections at the University of St Andrews Library and the On Growth and Form 100 conference in Dundee/St Andrews.)

Watch a livestream of some of the research for this post being done here.

]]>
10
Stephen Wolfram <![CDATA[It’s Another Impressive Release! Launching Version 11.2 Today]]> http://blog.internal.stephenwolfram.com/?p=14302 2017-10-17T14:33:55Z 2017-09-14T15:46:53Z It’s Another Impressive Release! Launching Version 11.2 TodayOur Latest R&D Output I’m excited today to announce the latest output from our R&D pipeline: Version 11.2 of the Wolfram Language and Mathematica—available immediately on desktop (Mac, Windows, Linux) and cloud. It was only this spring that we released Version 11.1. But after the summer we’re now ready for another impressive release—with all kinds [...]]]> It’s Another Impressive Release! Launching Version 11.2 Today

Our Latest R&D Output

I’m excited today to announce the latest output from our R&D pipeline: Version 11.2 of the Wolfram Language and Mathematica—available immediately on desktop (Mac, Windows, Linux) and cloud.

It was only this spring that we released Version 11.1. But after the summer we’re now ready for another impressive release—with all kinds of additions and enhancements, including 100+ entirely new functions:

New functions word cloud

We have a very deliberate strategy for our releases. Integer releases (like 11) concentrate on major complete new frameworks that we’ll be building on far into the future. “.1” releases (like 11.2) are intended as snapshots of the latest output from our R&D pipeline–delivering new capabilities large and small as soon as they’re ready.

Version 11.2 has a mixture of things in it—ranging from ones that provide finishing touches to existing major frameworks, to ones that are first hints of major frameworks under construction. One of my personal responsibilities is to make sure that everything we add is coherently designed, and fits into the long-term vision of the system in a unified way.

And by the time we’re getting ready for a release, I’ve been involved enough with most of the new functions we’re adding that they begin to feel like personal friends. So when we’re doing a .1 release and seeing what new functions are going to be ready for it, it’s a bit like making a party invitation list: who’s going to be able to come to the big celebration?

Years back there’d be a nice list, but it would be of modest length. Today, however, I’m just amazed at how fast our R&D pipeline is running, and how much comes out of it every month. Yes, we’ve been consistently building our Wolfram Language technology stack for more than 30 years—and we’ve got a great team. But it’s still a thrill for me to see just how much we’re actually able to deliver to all our users in a .1 release like 11.2.

Advances in Machine Learning

It’s hard to know where to begin. But let’s pick a current hot area: machine learning.

We’ve had functionality that would now be considered machine learning in the Wolfram Language for decades, and back in 2014 we introduced the “machine-learning superfunctions” Classify and Predict—to give broad access to modern machine learning. By early 2015, we had state-of-the-art deep-learning image identification in ImageIdentify, and then, last year, in Version 11, we began rolling out our full symbolic neural net computation system.

Our goal is to push the envelope of what’s possible in machine learning, but also to deliver everything in a nice, integrated way that makes it easy for a wide range of people to use, even if they’re not machine-learning experts. And in Version 11.2 we’ve actually used machine learning to add automation to our machine-learning capabilities.

So, in particular, Classify and Predict are significantly more powerful in Version 11.2. Their basic scheme is that you give them training data, and they’ll learn from it to automatically produce a machine-learning classifier or predictor. But a critical thing in doing this well is to know what features to extract from the data—whether it’s images, sounds, text, or whatever. And in Version 11.2 Classify and Predict have a variety of new kinds of built-in feature extractors that have been pre-trained on a wide range of kinds of data.

But the most obviously new aspect of Classify and Predict is how they select the core machine-learning method to use (as well as hyperparameters for it). (By the way, 11.2 also introduces things like optimized gradient-boosted trees.) And if you run Classify and Predict now in a notebook you’ll actually see them dynamically figuring out and optimizing what they’re doing (needless to say, using machine learning):

Classify and Predict animation

By the way, you can always press Stop to stop the training process. And with the new option TimeGoal you can explicitly say how long the training should be planned to be—from seconds to years.

As a field, machine learning is advancing very rapidly right now (in the course of my career, I’ve seen perhaps a dozen fields in this kind of hypergrowth—and it’s always exciting). And one of the things about our general symbolic neural net framework is that we’re able to take new advances and immediately integrate them into our long-term system—and build on them in all sorts of ways.

At the front lines of this is the function NetModel—to which new trained and untrained models are being added all the time. (The models are hosted in the cloud—but downloaded and cached for desktop or embedded use.) And so, for example, a few weeks ago NetModel got a new model for inferring geolocations of photographs—that’s based on basic research from just a few months ago:

NetModel["ResNet-101 Trained on YFCC100M Geotagged Data"]

NetModel["ResNet-101 Trained on YFCC100M Geotagged Data"]

Now if we give it a picture with sand dunes in it, its top inferences for possible locations seem to center around certain deserts:

GeoBubbleChart[NetModel["ResNet-101 Trained on YFCC100M Geotagged Data"]["<image suppressed>", {"TopProbabilities", 50}]]

GeoBubbleChart[
 NetModel["ResNet-101 Trained on YFCC100M Geotagged Data"][
  CloudGet["https://wolfr.am/dunes"], {"TopProbabilities", 50}]]

NetModel handles networks that can be used for all sorts of purposes—not only as classifiers, but also, for example, as feature extractors.

Building on NetModel and our symbolic neural network framework, we’ve also been able to add new built-in classifiers to use directly from Classify. So now, in addition to things like sentiment, we have NSFW, face age and facial expression (yes, an actual tiger isn’t safe, but in a different sense):

Classify["NSFWImage", "<image suppressed>"]

Classify["NSFWImage",CloudGet["https://wolfr.am/tiger"]]

Our built-in ImageIdentify function (whose underlying network you can access with NetModel) has been tuned and retrained for Version 11.2—but fundamentally it’s still a classifier. One of the important things that’s happening with machine learning is the development of new types of functions, supporting new kinds of workflows. We’ve got a lot of development going on in this direction, but for 11.2 one new (and fun) example is ImageRestyle—that takes a picture and applies the style of another picture to it:

ImageRestyle["<image suppressed>", "<image suppressed>"]

ImageRestyle[\[Placeholder],\[Placeholder]]

And in honor of this new functionality, maybe it’s time to get the image on my personal home page replaced with something more “styled”—though it’s a bit hard to know what to choose:

ImageRestyle[#, [] PerformanceGoal -> "Quality", TargetDevice -> "GPU"] & /@ {insert image,insert image,insert image,insert image,insert image,insert image}

ImageRestyle gallery

ImageRestyle[#, , PerformanceGoal -> "Quality", 
   TargetDevice -> 
    "GPU"] & /@ {\[Placeholder], \[Placeholder], \[Placeholder], \
\[Placeholder], \[Placeholder], \[Placeholder]}

By the way, another new feature of 11.2 is the ability to directly export trained networks and other machine-learning functionality. If you’re only interested in an actual network, you can get in MXNet format—suitable for immediate execution wherever MXNet is supported. In typical real situations, there’s some pre- and post-processing that’s needed as well—and the complete functionality can be exported in WMLF (Wolfram Machine Learning Format).

Cloud (and iOS) Notebooks

We invented the idea of notebooks back in 1988, for Mathematica 1.0—and over the past 29 years we’ve been steadily refining and extending how they work on desktop systems. About nine years ago we also began the very complex process of bringing our notebook interface to web browsers—to be able to run notebooks directly in the cloud, without any need for local installation.

It’s been a long, hard journey. But between new features of the Wolfram Language and new web technologies (like isomorphic React, Flow, MobX)—and heroic efforts of software engineering—we’re finally reaching the point where our cloud notebooks are ready for robust prime-time use. Like, try this one:

Notebook

We actually do continuous releases of the Wolfram Cloud—but with Version 11.2 of the Wolfram Language we’re able to add a final layer of polish and tuning to cloud notebooks.

You can create and compute directly on the web, and you can immediately “peel off” a notebook to run on the desktop. Or you can start on the desktop, and immediately push your notebook to the cloud, so it can be shared, embedded—and further edited or computed with—in the cloud.

By the way, when you’re using the Wolfram Cloud, you’re not limited to desktop systems. With the Wolfram Cloud App, you can work with notebooks on mobile devices too. And now that Version 11.2 is released, we’re able to roll out a new version of the Wolfram Cloud App, that makes it surprisingly realistic (thanks to some neat UX ideas) to write Wolfram Language code even on your phone.

Talking of mobile devices, there’s another big thing that’s coming: interactive Wolfram Notebooks running completely locally and natively on iOS devices—both tablets and phones. This has been another heroic software engineering project—which actually started nearly as long ago as the cloud notebook project.

The goal here is to be able to read and interact with—but not author—notebooks directly on an iOS device. And so now with the Wolfram Player App that will be released next week, you can have a notebook on your iOS device, and use Manipulate and other dynamic content, as well as read and navigate notebooks—with the whole interface natively adapted to the touch environment.

For years it’s been frustrating when people send me notebook attachments in email, and I’ve had to do things like upload them to the cloud to be able to read them on my phone. But now with native notebooks on iOS, I can immediately just read notebook attachments directly from email.

Mathematical Limits

Math was the first big application of the Wolfram Language (that’s why it was called Mathematica!)… and for more than 30 years we’ve been committed to aggressively pursuing R&D to expand the domain of math that can be made computational. And in Version 11.2 the biggest math advance we’ve made is in the area of limits.

Mathematica 1.0 back in 1988 already had a basic Limit function. And over the years Limit has gradually been enhanced. But in 11.2—as a result of algorithms we’ve developed over the past several years—it’s reached a completely new level.

The simple-minded way to compute a limit is to work out the first terms in a power series. But that doesn’t work when functions increase too rapidly, or have wild and woolly singularities. But in 11.2 the new algorithms we’ve developed have no problem handling things like this:

Limit[E^(E^x + x^2) (-Erf[E^-E^x - x] - Erf[x]), x -> \[Infinity]]

Limit[E^(E^x + x^2) (-Erf[E^-E^x - x] - Erf[x]), x -> \[Infinity]]
"Limit[3*x + Sqrt[9*x^2 + 4*x - Sin[x]], x -> -Infinity]

Limit[(3 x + Sqrt[9 x^2 + 4 x - Sin[x]]), x -> -\[Infinity]]

It’s very convenient that we have a test set of millions of complicated limit problems that people have asked Wolfram|Alpha about over the past few years—and I’m pleased to say that with our new algorithms we can now immediately handle more than 96% of them.

Limits are in a sense at the very core of calculus and continuous mathematics—and to do them correctly requires a huge tower of knowledge about a whole variety of areas of mathematics. Multivariate limits are particularly tricky—with the main takeaway from many textbooks basically being “it’s hard to get them right”. Well, in 11.2, thanks to our new algorithms (and with a lot of support from our algebra, functional analysis and geometry capabilities), we’re finally able to correctly do a very wide range of multivariate limits—saying whether there’s a definite answer, or whether the limit is provably indeterminate.

Version 11.2 also introduces two other convenient mathematical constructs: MaxLimit and MinLimit (sometimes known as lim sup and lim inf). Ordinary limits have a habit of being indeterminate whenever things get funky, but MaxLimit and MinLimit have definite values, and are what come up most often in applications.

So, for example, there isn’t a definite ordinary limit here:

Limit[Sin[x] + Cos[x/4], x -> Infinity]

Limit[Sin[x] + Cos[x/4], x -> \[Infinity]]

But there’s a MaxLimit, that turns out to be a complicated algebraic number:

MaxLimit[Sin[x] + Cos[x/4], x -> \[Infinity]] // FullSimplify

MaxLimit[Sin[x] + Cos[x/4], x -> \[Infinity]] // FullSimplify
N[%]

N[%]

Another new construct in 11.2 is DiscreteLimit, that gives limits of sequences. Like here’s it’s illustrating the Prime Number Theorem:

DiscreteLimit[Prime[n]/(n*Log[n]), n -> Infinity]

DiscreteLimit[Prime[n]/(n Log[n]), n -> \[Infinity]]

And here it’s giving the limiting value of the solution to a recurrence relation:

DiscreteLimit[RSolveValue[{x[n+1] == Sqrt[1 + x[n]+1/x[n]], x[1] == 3}, x[n], n],n->\[Infinity]]

DiscreteLimit[
 RSolveValue[{x[n + 1] == Sqrt[1 + x[n] + 1/x[n]], x[1] == 3}, x[n], 
  n], n -> \[Infinity]]

All Sorts of New Data

There’s always new data in the Wolfram Knowledgebase—flowing every second from all sorts of data feeds, and systematically being added by our curators and curation systems. The architecture of our cloud and desktop system allows both new data and new types of data (as well as natural language input for it) to be immediately available in the Wolfram Language as soon as it’s in the Wolfram Knowledgebase.

And between Version 11.1 and Version 11.2, there’ve been millions of updates to the Knowledgebase. There’ve also been some new types of data added. For example—after several years of development—we’ve now got well-curated data on all notable military conflicts, battles, etc. in history:

Entity["MilitaryConflict", "SecondPunicWar"][EntityProperty["MilitaryConflict", "Battles"]]

Entity["MilitaryConflict", "SecondPunicWar"][
 EntityProperty["MilitaryConflict", "Battles"]]
GeoListPlot[%]

GeoListPlot[%]

Another thing that’s new in 11.2 is greatly enhanced predictive caching of data in the Wolfram Language—making it much more efficient to compute with large volumes of curated data from the Wolfram Knowledgebase.

By the way, Version 11.2 is the first new version to be released since the Wolfram Data Repository was launched. And through the Data Repository, 11.2 has access to nearly 600 curated datasets across a very wide range of areas. 11.2 also now supports functions like ResourceSubmit, for programmatically submitting data for publication in the Wolfram Data Repository. (You can also publish data yourself just using CloudDeploy.)

There’s a huge amount of data and types of computations available in Wolfram|Alpha—that with great effort have been brought to the level where they can be relied on, at least for the kind of one-shot usage that’s typical in Wolfram|Alpha. But one of our long-term goals is to take as many areas as possible and raise the level even higher—to the point where they can be built into the core Wolfram Language, and relied on for systematic programmatic usage.

In Version 11.2 an area where this has happened is ocean tides. So now there’s a function TideData that can give tide predictions for any of the tide stations around the world. I actually found myself using this function in a recent livecoding session I did—where it so happened that I needed to know daily water levels in Aberdeen Harbor in 1913. (Watch the Twitch recording to find out why!)

TideData[Entity["City", {"Aberdeen", "Maryland", "UnitedStates"}], "WaterLevel", DateRange[DateObject[{1913, 1, 1}], DateObject[{1913, 12, 31}], "Day"]]

TideData[Entity[
  "City", {"Aberdeen", "Maryland", "UnitedStates"}], "WaterLevel", 
 DateRange[DateObject[{1913, 1, 1}], DateObject[{1913, 12, 31}], 
  "Day"]]
DateListPlot[%]

DateListPlot[%]

GeoImage

GeoGraphics and related functions have built-in access to detailed maps of the world. They’ve also had access to low-resolution satellite imagery. But in Version 11.2 there’s a new function GeoImage that uses an integrated external service to provide full-resolution satellite imagery:

GeoImage[GeoDisk[Entity["Building", "ThePentagon::qzh8d"], Quantity[0.4, "Miles"]]]

GeoImage[GeoDisk[Entity["Building", "ThePentagon::qzh8d"], 
  Quantity[0.4, "Miles"]]]
GeoImage[GeoDisk[Entity["Building", "Stonehenge::46k59"],    Quantity[250, "Feet"]]]

GeoImage[GeoDisk[Entity["Building", "Stonehenge::46k59"], 
  Quantity[250, "Feet"]]]

I’ve ended up using GeoImage in each of the two livecoding sessions I did just recently. Yes, in principle one could go to the web and find a satellite image of someplace, but it’s amazing what a different level of utility one reaches when one can programmatically get the satellite image right inside the Wolfram Language—and then maybe feed it to image processing, or visualization, or machine-learning functions. Like here’s a feature space plot of satellite images of volcanos in California:

FeatureSpacePlot[GeoImage /@ GeoEntities[Entity["AdministrativeDivision", {"California", "UnitedStates"}], "Volcano"]]

FeatureSpacePlot[
 GeoImage /@ 
  GeoEntities[
   Entity["AdministrativeDivision", {"California", "UnitedStates"}], 
   "Volcano"]]

We’re always updating and adding all sorts of geo data in the Wolfram Knowledgebase. And for example, as of Version 11.2, we’ve now got high-resolution geo elevation data for the Moon—which came in very handy for our recent precision eclipse computation project.

ListPlot3D[GeoElevationData[GeoDisk[Entity["MannedSpaceMission", "Apollo15"][EntityProperty["MannedSpaceMission", "LandingPosition"]], Quantity[10, "Miles"]]], Mesh -> None]" width="564" height="433

ListPlot3D[
 GeoElevationData[
  GeoDisk[Entity["MannedSpaceMission", "Apollo15"][
    EntityProperty["MannedSpaceMission", "LandingPosition"]], 
   Quantity[10, "Miles"]]], Mesh -> None]

Visualization

One of the obvious strengths of the Wolfram Language is its wide range of integrated and highly automated visualization capabilities. Version 11.2 adds some convenient new functions and options. An example is StackedListPlot, which, as its name suggests, makes stacked (cumulative) list plots:

StackedListPlot[RandomInteger[10, {3, 30}]]

StackedListPlot[RandomInteger[10, {3, 30}]]

There’s also StackedDateListPlot, here working with historical time series from the Wolfram Knowledgebase:

StackedDateListPlot[  EntityClass[   "Country", {    EntityProperty["Country", "Population"] -> TakeLargest[10]}][   Dated["Population", All],     "Association"], PlotLabels -> Automatic]

StackedDateListPlot[
 EntityClass[
  "Country", {
   EntityProperty["Country", "Population"] -> TakeLargest[10]}][
  Dated["Population", All],
    "Association"], PlotLabels -> Automatic]
StackedDateListPlot[  EntityClass[   "Country", {    EntityProperty["Country", "Population"] -> TakeLargest[10]}][   Dated["Population", All],   "Association"], PlotLabels -> Automatic, PlotLayout -> "Percentile"]

StackedDateListPlot[
 EntityClass[
  "Country", {
   EntityProperty["Country", "Population"] -> TakeLargest[10]}][
  Dated["Population", All],
  "Association"], PlotLabels -> Automatic, PlotLayout -> "Percentile"]

One of our goals in the Wolfram Language is to make good stylistic choices as automatic as possible. And in Version 11.2 we’ve, for example, added a whole collection of plot themes for AnatomyPlot3D. You can always explicitly give whatever styling you want. But we provide many default themes. You can pick a classic anatomy book look (by the way, all these 3D objects are fully manipulable and computable):

AnatomyPlot3D[Entity["AnatomicalStructure", "LeftHand"], PlotTheme -> "Classic"]

AnatomyPlot3D[Entity["AnatomicalStructure", "LeftHand"],PlotTheme -> "Classic"]

Or you can go for more of a Gray’s Anatomy look:

AnatomyPlot3D[Entity["AnatomicalStructure", "LeftHand"], PlotTheme -> "Vintage"]

AnatomyPlot3D[Entity["AnatomicalStructure", "LeftHand"], 
 PlotTheme -> "Vintage"]

Or you can have a “scientific” theme that tries to make different structures as distinct as possible:

AnatomyPlot3D[Entity["AnatomicalStructure", "LeftHand"], PlotTheme -> "Scientific"]

StackedDateListPlot[
 EntityClass[
  "Country", {
   EntityProperty["Country", "Population"] -> TakeLargest[10]}][
  Dated["Population", All],
  "Association"], PlotLabels -> Automatic, PlotLayout -> "Percentile"]

3D Computational Geometry

The Wolfram Language has very strong computational geometry capabilities—that work on both exact surfaces and approximate meshes. It’s a tremendous algorithmic challenge to smoothly handle constructive geometry in 3D—but after many years of work, Version 11.2 can do it:

RegionIntersection[MengerMesh[2, 3],<br />  BoundaryDiscretizeRegion[Ball[{1, 1, 1}]]]

RegionIntersection[MengerMesh[2, 3], 
 BoundaryDiscretizeRegion[Ball[{1, 1, 1}]]]

And of course, everything fits immediately into the rest of the system:

Volume[%]

Volume[%]

More Audio

Version 11 introduced a major new framework for large-scale audio processing in the Wolfram Language. We’re still developing all sorts of capabilities based on this framework, especially using machine learning. And in Version 11.2 there are a number of immediate enhancements. There are very practical things, like built-in support for AudioCapture under Linux. There’s also now the notion of a dynamic AudioStream, whose playback can be programmatically controlled.

Another new function is SpeechSynthesize, which creates audio from text:

SpeechSynthesize["hello"]

SpeechSynthesize["hello"]
Spectrogram[%]

Spectrogram[%]

Capture the Screen

The Wolfram Language tries to let you get data wherever you can. One capability added for Version 11.2 is being able to capture images of your computer screen. (Rasterize has been able to rasterize complete notebooks for a long time; CurrentNotebookImage now captures an image of what’s visible from a notebook on your screen.) Here’s an image of my main (first) screen, captured as I’m writing this post:

CurrentScreenImage[1]

CurrentScreen output

CurrentScreenImage[1]

Of course, I can now do computation on this image, just like I would on any other image. Here’s a map of the inferred “saliency” of different parts of my screen:

ImageSaliencyFilter["<image suppressed>"]//Colorize

ImageSaliencyFilter[CurrentScreenImage[1]]//Colorize

Language Features

Part of developing the Wolfram Language is adding major new frameworks. But another part is polishing the system, and implementing new functions that make doing things in the system ever easier, smoother and clearer.

Here are a few functions we’ve added in 11.2. The first is simple, but useful: TakeList—a function that successively takes blocks of elements from a list:

TakeList[Alphabet[], {2, 5, 3, 4}]

TakeList[Alphabet[], {2, 5, 3, 4}]

Then there’s FindRepeat (a “colleague” of FindTransientRepeat), that finds exact repeats in sequences—here for a Fibonacci sequence mod 10:

FindRepeat[Mod[Array[Fibonacci, 500], 10]]

FindRepeat[Mod[Array[Fibonacci, 500], 10]]

Here’s a very different kind of new feature: an addition to Capitalize that applies the heuristics for capitalizing “important words” to make something “title case”. (Yes, for an individual string this doesn’t look so useful; but it’s really useful when you’ve got 100 strings from different sources to make consistent.)

Capitalize["a new kind of science", "TitleCase"]

Capitalize["a new kind of science", "TitleCase"]

Talking of presentation, here’s a simple but very useful new output format: DecimalForm. Numbers are normally displayed in scientific notation when they get big, but DecimalForm forces “grade school” number format, without scientific notation:

Table[16.5^n, {n, 10}]

Table[16.5^n, {n, 10}]
DecimalForm[Table[16.5^n, {n, 10}]]

DecimalForm[Table[16.5^n, {n, 10}]]

Another language enhancement added in 11.2—though it’s really more of a seed for the future—is TwoWayRule, input as <->. Ever since Version 1.0 we’ve had Rule (->), and over the years we’ve found Rule increasingly useful as an inert structure that can symbolically represent diverse kinds of transformations and connections. Rule is fundamentally one-way: “left-hand side goes to right-hand side”. But one also sometimes needs a two-way version—and that’s what TwoWayRule provides.

Right now TwoWayRule can be used, for example, to enter undirected edges in a graph, or pairs of levels to exchange in Transpose. But in the future, it’ll be used more and more widely.

Graph[{1 <-> 2, 2 <-> 3, 3 <-> 1}]

Graph[{1 <-> 2, 2 <-> 3, 3 <-> 1}]

11.2 has all sorts of other language enhancements. Here’s an example of a somewhat different kind: the functions StringToByteArray and ByteArrayToString, which handle the somewhat tricky issue of converting between raw byte arrays and strings with various encodings (like UTF-8).

Initialization & System Operations

How do you get the Wolfram Language to automatically initialize itself in some particular way? All the way from Version 1.0, you’ve been able to set up an init.m file to run at initialization time. But finally now in Version 11.2 there’s a much more general and programmatic way of doing this—using InitializationValue and related constructs.

It’s made possible by the PersistentValue framework introduced in 11.1. And what’s particularly nice about it is that it allows a whole range of “persistence locations”—so you can store your initialization information on a per-session, per-computer, per-user, or also (new in 11.2) per-notebook way.

Talking about things that go all the way to Version 1.0, here’s a little story. Back in Version 1.0, Mathematica (as it then was) pretty much always used to display how much memory was still available on your computer (and, yes, you had to be very careful back then because there usually wasn’t much). Well, somewhere along the way, as virtual memory became widespread, people started thinking that “available memory” didn’t mean much, and we stopped displaying it. But now, after being gone for 25+ years, modern operating systems have let us bring it back—and there’s a new function MemoryAvailable in Version 11.2. And, yes, for my computer the result has gained about 5 digits relative to what it had in 1988:

MemoryAvailable[]

MemoryAvailable[ ]

Unified Asynchronous Tasks

There’ve been ways to do some kinds of asynchronous or “background” tasks in the Wolfram Language for a while, but in 11.2 there’s a complete systematic framework for it. There’s a thing called TaskObject that symbolically represents an asynchronous task. And there are basically now three ways such a task can be executed. First, there’s CloudSubmit, which submits the task for execution in the cloud. Then there’s LocalSubmit, which submits the task to be executed on your local computer, but in a separate subkernel. And finally, there’s SessionSubmit, which executes the task in idle time in your current Wolfram Language session.

When you submit a task, it’s off getting executed (you can schedule it to happen at particular times using ScheduledTask). The way you “hear back” from the task is through “handler functions”: functions that are set up when you submit the task to “handle” certain events that can occur during the execution of the task (completion, errors, etc.).

There are also functions like TaskSuspend, TaskAbort, TaskWait and so on, that let you interact with tasks “from the outside”. And, yes, when you’re doing big machine-learning trainings, for example, this comes in pretty handy.

Connectivity

We’re always keen to make the Wolfram Language as connected as it can be. And in Version 11.2 we’ve added a variety of features to achieve that. In Version 11 we introduced the Authentication option, which lets you give credentials in functions like URLExecute. Version 11 already allowed for PermissionsKey (a.k.a. an “app id”). In 11.2 you can now give an explicit username and password, and you can also use SecuredAuthenticationKey to provide OAuth credentials. It’s tricky stuff, but I’m pleased with how cleanly we’re able to represent it using the symbolic character of the Wolfram Language—and it’s really useful when you’re, for example, actually working with a bunch internal websites or APIs.

Back in Version 10 (2014) we introduced the very powerful idea of using APIFunction to provide a symbolic specification for a web API—that could be deployed to the cloud using CloudDeploy. Then in Version 10.2 we introduced MailReceiverFunction, which responds not to web requests, but instead to receiving mail messages. (By the way, in 11.2 we’ve considerably strengthened SendMail, notably adding various authentication and address validation capabilities.)

In Version 11, we introduced the channel framework, which allows for publish-subscribe interactions between Wolfram Language instances (and external programs)—enabling things like chat, as well as a host of useful internal services. Well, in our continual path of automating and unifying, we’re introducing in 11.2 ChannelReceiverFunction—which can be deployed to the cloud to respond to whatever messages are sent on a particular channel.

In the low-level software engineering of the Wolfram Language we’ve used sockets for a long time. A few years ago we started exposing some socket functionality within the language. And now in 11.2 we have a full socket framework. The socket framework supports both traditional TCP sockets, as well as modern ZeroMQ sockets.

External Programs

Ever since the beginning, the Wolfram Language has been able to communicate with external C programs—actually using its native WSTP (Wolfram Symbolic Transfer Protocol) symbolic expression transfer protocol. Years ago J/Link and .NetLink enabled seamless connection to Java and .Net programs. RLink did the same for R. Then there are things like LibraryLink, that allow direct connection to DLLs—or RunProcess for running programs from the shell.

But 11.2 introduces a new form of external program communication: ExternalEvaluate. ExternalEvaluate is for doing computation in languages which—like the Wolfram Language—support REPL-style input/output. The two first examples available in 11.2 are Python and NodeJS.

Here’s a computation done with NodeJS—though this would definitely be better done directly in the Wolfram Language:

ExternalEvaluate["NodeJS", "Math.sqrt(50)"]

ExternalEvaluate["NodeJS", "Math.sqrt(50)"]

Here’s a Python computation (yes, it’s pretty funky to use & for BitAnd):

ExternalEvaluate["Python", "[ i & 10 for i in range(10)]"]

ExternalEvaluate["Python", "[ i & 10 for i in range(10)]"]

Of course, the place where things start to get useful is when one’s accessing large external code bases or libraries. And what’s nice is that one can use the Wolfram Language to control everything, and to analyze the results. ExternalEvaluate is in a sense a very lightweight construct—and one can routinely use it even deep inside some piece of Wolfram Language code.

There’s an infrastructure around ExternalEvaluate, aimed at connecting to the correct executable, appropriately converting types, and so on. There’s also StartExternalSession, which allows you to start a single external session, and then perform multiple evaluations in it.

The Whole List

So is there still more to say about 11.2? Yes! There are lots of new functions and features that I haven’t mentioned at all. Here’s a more extensive list:

New features

But if you want to find out about 11.2, the best thing to do is to actually run it. I’ve actually been running pre-release versions of 11.2 on my personal machines for a couple of months. So by now I’m taking the new features and functions quite for granted—even though, earlier on, I kept on saying “this is really useful; how could we have not had this for 30 years?”. Well, realistically, it’s taken building everything we have so far—not only to provide the technical foundations, but also to seed the ideas, for 11.2. But now our work on 11.2 is done, and 11.2 is ready to go out into the world—and deliver the latest results from our decades of research and development.

To comment, please visit the copy of this post at the Wolfram Blog »

]]>
0
Stephen Wolfram <![CDATA[When Exactly Will the Eclipse Happen? A Multimillenium Tale of Computation]]> http://blog.internal.stephenwolfram.com/?p=14009 2017-09-11T20:14:54Z 2017-08-16T03:25:27Z eclipseblog-thumbPreparing for August 21, 2017 On August 21, 2017, there’s going to be a total eclipse of the Sun visible on a line across the US. But when exactly will the eclipse occur at a given location? Being able to predict astronomical events has historically been one of the great triumphs of exact science. But [...]]]> eclipseblog-thumb

Preparing for August 21, 2017

On August 21, 2017, there’s going to be a total eclipse of the Sun visible on a line across the US. But when exactly will the eclipse occur at a given location? Being able to predict astronomical events has historically been one of the great triumphs of exact science. But in 2017, how well can it actually be done?

The answer, I think, is well enough that even though the edge of totality moves at just over 1000 miles per hour it should be possible to predict when it will arrive at a given location to within perhaps a second. And as a demonstration of this, we’ve created a website to let anyone enter their geo location (or address) and then immediately compute when the eclipse will reach them—as well as generate many pages of other information.

PrecisionEclipse.com pages

It’s an Old Business

These days it’s easy to find out when the next solar eclipse will be; indeed built right into the Wolfram Language there’s just a function that tells you (in this form the output is the “time of greatest eclipse”):

SolarEclipse[]

SolarEclipse[]

It’s also easy to find out, and plot, where the region of totality will be:

GeoListPlot[SolarEclipse["TotalPhasePolygon"]]

GeoListPlot[SolarEclipse["TotalPhasePolygon"]]

Or to determine that the whole area of totality will be about 16% of the area of the US:

GeoArea[SolarEclipse["TotalPhasePolygon"]]/GeoArea[Entity["Country", "UnitedStates"]]

GeoArea[SolarEclipse["TotalPhasePolygon"]]/GeoArea[Entity["Country", "UnitedStates"]]

But computing eclipses is not exactly a new business. In fact, the Antikythera device from 2000 years ago even tried to do it—using 37 metal gears to approximate the motion of the Sun and Moon (yes, with the Earth at the center). To me there’s something unsettling—and cautionary—about the fact that the Antikythera device stands as such a solitary piece of technology, forgotten but not surpassed for more than 1600 years.

But right there on the bottom of the device there’s an arm that moves around, and when it points to an Η or Σ marking, it indicates a possible Sun or Moon eclipse. The way of setting dates on the device is a bit funky (after all, the modern calendar wouldn’t be invented for another 1500 years), but if one takes the simulation on the Wolfram Demonstrations Project (which was calibrated back in 2012 when the Demonstration was created), and turns the crank to set the device for August 21, 2017, here’s what one gets:

The Antikythera device

Antikythera device in action

And, yes, all those gears move so as to line the Moon indicator up with the Sun—and to make the arm on the bottom point right at an Η—just as it should for a solar eclipse. It’s amazing to see this computation successfully happen on a device designed 2000 years ago.

Of course the results are a lot more accurate today. Though, strangely, despite all the theoretical science that’s been done, the way we actually compute the position of the Sun and Moon is conceptually very much like the gears—and effectively epicycles—of the Antikythera device. It’s just that now we have the digital equivalent of hundreds of thousands of gears.

Why Do Eclipses Happen?

A total solar eclipse occurs when the Moon gets in front of the Sun from the point of view of a particular location on the Earth. And it so happens that at this point in the Earth’s history the Moon can just block the Sun because it has almost exactly the same angular diameter in the sky as the Sun (about 0.5° or 30 arc-minutes).

So when does the Moon get between the Sun and the Earth? Well, basically every time there’s a new moon (i.e. once every lunar month). But we know there isn’t an eclipse every month. So how come?

Sun, Moon and Earth aligned

Graphics[{Style[Disk[{0, 0}, .3/5], Yellow], 
  Style[Disk[{.8, 0}, .1/5], Gray], Style[Disk[{1, 0}, .15/5], Blue]}]

Well, actually, in the analogous situation of Ganymede and Jupiter, there is an eclipse every time Ganymede goes around Jupiter (which happens to be about once per week). Like the Earth, Jupiter’s orbit around the Sun lies in a particular plane (the “Plane of the Ecliptic”). And it turns out that Ganymede’s orbit around Jupiter also lies in essentially the same plane. So every time Ganymede reaches the “new moon” position (or, in official astronomy parlance, when it’s aligned “in syzygy”—pronounced sizz-ee-gee), it’s in the right place to cast its shadow onto Jupiter, and to eclipse the Sun wherever that shadow lands. (From Jupiter, Ganymede appears about 3 times the size of the Sun.)

But our moon is different. Its orbit doesn’t lie in the plane of the ecliptic. Instead, it’s inclined at about 5°. (How it got that way is unknown, but it’s presumably related to how the Moon was formed.) But that 5° is what makes eclipses so comparatively rare: they can only happen when there’s a “new moon configuration” (syzygy) right at a time when the Moon’s orbit passes through the Plane of the Ecliptic.

To show what’s going on, let’s draw an exaggerated version of everything. Here’s the Moon going around the Earth, colored red whenever it’s close to the Plane of the Ecliptic:

Lunar path and the Plane of the Ecliptic

Graphics3D[{With[{dt = 0, \[Theta] = 20 Degree},
   Table[{With[{p = {Sin[2 Pi (t + dt)/27.3] Cos[\[Theta]],
         Cos[2 Pi (t + dt)/27.3] Cos[\[Theta]],
         Cos[2 Pi (t + dt)/27.3] Sin[\[Theta]]}}, {Style[
        Line[{{0, 0, 0}, p}], Opacity[.1]],
       Style[Sphere[p, .05],
        Blend[{Red, GrayLevel[.8, .02]},
         Sqrt[Abs[Cos[2 Pi t/27.2]]]]]}],
     Style[Sphere[{0, 0, 0}, .1], Blue]}, {t, 0, 26}]], EdgeForm[Red],
   Style[InfinitePlane[{0, 0, 0}, {{1, 0, 0}, {0, 1, 0}}],
   Directive[Red, Opacity[.02]]]}, Lighting -> "Neutral",
 Boxed -> False]

Now let’s look at what happens over the course of about a year. We’re showing a dot for where the Moon is each day. And the dot is redder if the Moon is closer to the Plane of the Ecliptic that day. (Note that if this was drawn to scale, you’d barely be able to see the Moon’s orbit, and it wouldn’t ever seem to go backwards like it does here.)

Approximately one year of lunar orbits

With[{dt = 1}, 
 Graphics[{Style[Disk[{0, 0}, .1], Darker[Yellow]], 
   Table[{With[{p = .2 {Sin[2 Pi t/27.3], Cos[2 Pi t/27.3]} + {Sin[
           2 Pi t/365.25], Cos[2 Pi t/365.25]}}, {Style[
        Line[{{Sin[2 Pi t/365.25], Cos[2 Pi t/365.25]}, p}], 
        Opacity[.3]], 
       Style[Disk[p, .01], 
        Blend[{Red, GrayLevel[.8]}, 
         Sqrt[Abs[Cos[2 Pi (t + dt)/27.2]]]]]}], 
     Style[Disk[{Sin[2 Pi t/365.25], Cos[2 Pi t/365.25]}, .005], 
      Blue]}, {t, 360}]}]]

Now we can start to see how eclipses work. The basic point is that there’s a solar eclipse whenever the Moon is both positioned between the Earth and the Sun, and it’s in the Plane of the Ecliptic. In the picture, those two conditions correspond to the Moon being as far as possible towards the center, and as red as possible. So far we’re only showing the position of the (exaggerated) moon once per day. But to make things clearer, let’s show it four times a day—and now prune out cases where the Moon isn’t at least roughly lined up with the Sun:

Eclipse points in lunar orbit

With[{dt=1},Graphics[{Style[Disk[{0,0},.1],Darker[Yellow]],Table[{With[{p=.2 {Sin[2 Pi t/27.3],Cos[2 Pi t/27.3]}+{Sin[2 Pi t/365.25],Cos[2 Pi t/365.25]}},If[Norm[p]>.81,{},{Style[Line[{{Sin[2 Pi t/365.25],Cos[2 Pi t/365.25]},p}],Opacity[.3]],Style[Disk[p,.01],Blend[{Red,GrayLevel[.8]},Sqrt[Abs[Cos[2 Pi (t+dt)/27.2]]]]]}]],Style[Disk[{Sin[2 Pi t/365.25],Cos[2 Pi t/365.25]},.005],Blue]},{t,1,360,.25}]}]]

And now we can see that at least in this particular case, there are two points (indicated by arrows) where the Moon is lined up and in the plane of the ecliptic (so shown red)—and these points will then correspond to solar eclipses.

In different years, the picture will look slightly different, essentially because the Moon is starting at a different place in its orbit at the beginning of the year. Here are schematic pictures for a few successive years:

Schematic pictures for a few successive years

GraphicsGrid[
 Partition[
  Table[With[{dt = 1}, 
    Graphics[{Table[{With[{p = .2 {Sin[2 Pi t/27.3], 
              Cos[2 Pi t/27.3]} + {Sin[2 Pi t/365.25], 
             Cos[2 Pi t/365.25]}}, 
         If[Norm[p] > .81, {}, {Style[Line[{{0, 0}, p}], 
            Blend[{Red, GrayLevel[.8]}, 
             Sqrt[Abs[Cos[2 Pi (t + dt)/27.2]]]]], 
           Style[Line[{{Sin[2 Pi t/365.25], Cos[2 Pi t/365.25]}, p}], 
            Opacity[.3]], 
           Style[Disk[p, .01], 
            Blend[{Red, GrayLevel[.8]}, 
             Sqrt[Abs[Cos[2 Pi (t + dt)/27.2]]]]]}]], 
        Style[Disk[{Sin[2 Pi t/365.25], Cos[2 Pi t/365.25]}, .005], 
         Blue]}, {t, 1 + n *365.25, 360 + n*365.25, .25}], 
      Style[Disk[{0, 0}, .1], Darker[Yellow]]}]], {n, 0, 5}], 3]]

It’s not so easy to see exactly when eclipses occur here—and it’s also not possible to tell which are total eclipses where the Moon is exactly lined up, and which are only partial eclipses. But there’s at least an indication, for example, that there are “eclipse seasons” in different parts of the year where eclipses happen.

OK, so what does the real data look like? Here’s a plot for 20 years in the past and 20 years in the future, showing the actual days in each year when total and partial solar eclipses occur (the small dots everywhere indicate new moons):

Plot of solar eclipses over 40 years

coord[date_] := {DateValue[date, "Year"], 
  date - NextDate[DateObject[{DateValue[date, "Year"]}], "Instant"]}
ListPlot[coord /@ SolarEclipse[{Now - Quantity[20, "Years"], Now + Quantity[20, "Years"], All}], AspectRatio -> 1/3, Frame -> True]

The reason for the “drift” between successive years is just that the lunar month (29.53 days) doesn’t line up with the year, so the Moon doesn’t go through a whole number of orbits in the course of a year, with the result that at the beginning of a new year, the Moon is in a different phase. But as the picture makes clear, there’s quite a lot of regularity in the general times at which eclipses occur—and for example there are usually 2 eclipses in a given year—though there can be more (and in 0.2% of years there can be as many as 5, as there last were in 1935 ).

To see more detail about eclipses, let’s plot the time differences (in fractional years) between all successive solar eclipses for 100 years in the past and 100 years in the future:

ListLinePlot[Differences[SolarEclipse[{Now - Quantity[100, "Years"], Now + Quantity[100, "Years"], All}]]/Quantity[1, "Years"], Mesh -> All, PlotRange -> All, Frame -> True, AspectRatio -> 1/3, FrameTicks -> {None, Automatic}]

ListLinePlot[Differences[SolarEclipse[{Now - Quantity[100, "Years"], Now + Quantity[100, "Years"], All}]]/Quantity[1, "Years"], Mesh -> All, PlotRange -> All, Frame -> True, AspectRatio -> 1/3, FrameTicks -> {None, Automatic}]

And now let’s plot the same time differences, but just for total solar eclipses:

ListLinePlot[Differences[SolarEclipse[{Now - Quantity[100, "Years"], Now + Quantity[100, "Years"], All}, EclipseType -> "Total"]]/Quantity[1, "Years"], Mesh -> All, PlotRange -> All, Frame -> True, AspectRatio -> 1/3, FrameTicks -> {None, Automatic}]

ListLinePlot[Differences[SolarEclipse[{Now - Quantity[100, "Years"], Now + Quantity[100, "Years"], All}, EclipseType -> "Total"]]/Quantity[1, "Years"], Mesh -> All, PlotRange -> All, Frame -> True, AspectRatio -> 1/3, FrameTicks -> {None, Automatic}]

There’s obviously a fair amount of overall regularity here, but there are also lots of little fine structure and irregularities. And being able to correctly predict all these details has basically taken science the better part of a few thousand years.

Ancient History

It’s hard not to notice an eclipse, and presumably even from the earliest times people did. But were eclipses just reflections—or omens—associated with random goings-on in the heavens, perhaps in some kind of soap opera among the gods? Or were they things that could somehow be predicted?

A few thousand years ago, it wouldn’t have been clear what people like astrologers could conceivably predict. When will the Moon be at a certain place in the sky? Will it rain tomorrow? What will the price of barley be? Who will win a battle? Even now, we’re not sure how predictable all of these are. But the one clear case where prediction and exact science have triumphed is astronomy.

At least as far as the Western tradition is concerned, it all seems to have started in ancient Babylon—where for many hundreds of years, careful observations were made, and, in keeping with the ways of that civilization, detailed records were kept. And even today we still have thousands of daily official diary entries written in what look like tiny chicken scratches preserved on little clay tablets (particularly from Ninevah, which happens now to be in Mosul, Iraq). “Night of the 14th: Cold north wind. Moon was in front of α Leonis. From 15th to 20th river rose 1/2 cubit. Barley was 1 kur 5 siit. 25th, last part of night, moon was 1 cubit 8 fingers behind ε Leonis. 28th, 74° after sunrise, solar eclipse…”

Babylonian clay tabletIf one looks at what happens on a particular day, one probably can’t tell much. But by putting observations together over years or even hundreds of years it’s possible to see all sorts of repetitions and regularities. And back in Babylonian times the idea arose of using these to construct an ephemeris—a systematic table that said where a particular heavenly body such as the Moon was expected to be at any particular time.

(Needless to say, reconstructing Babylonian astronomy is a complicated exercise in decoding what’s by now basically an alien culture. A key figure in this effort was a certain Otto Neugebauer, who happened to work down the hall from me at the Institute for Advanced Study in Princeton in the early 1980s. I would see him almost every day—a quiet white-haired chap, with a twinkle in his eye—and just sometimes I’d glimpse his huge filing system of index cards which I now realize was at the center of understanding Babylonian astronomy.)

One thing the Babylonians did was to measure surprisingly accurately the repetition period for the phases of the Moon—the so-called synodic month (or “lunation period”) of about 29.53 days. And they noticed that 235 synodic months was very close to 19 years—so that about every 19 years dates and phases of the Moon repeat their alignment, forming a so-called Metonic cycle (named after Meton of Athens, who described it in 432 BC).

It probably helps that the random constellations in the sky form a good pattern against which to measure the precise position of the Moon (it reminds me of the modern fashion of wearing fractals to make motion capture for movies easier). But the Babylonians noticed all sorts of details of the motion of the Moon. They knew about its “anomaly”: its periodic speeding up and slowing down in the sky (now known to be a consequence of its slightly elliptical orbit). And they measured the average period of this—the so-called anomalistic month—to be about 27.55 days. They also noticed that the Moon went above and below the Plane of the Ecliptic (now known to be because of the inclination of its orbit)—with an average period (the so-called draconic month) that they measured as about 27.21 days.

And by 400 BC they’d noticed that every so-called saros of about 18 years 11 days all these different periods essentially line up (223 synodic months, 239 anomalistic months and 242 draconic months)—with the result that the Moon ends up at about the same position relative to the Sun. And this means that if there was an eclipse at one saros, then one can make the prediction that there’s going to be an eclipse at the next saros too.

When one’s absolutely precise about it, there are all sorts of effects that prevent precise repetition at each saros. But over timescales of more than 1300 years, there are in fact still strings of eclipses separated from each other by one saros. (Over the course of such a saros series, the locations of the eclipses effectively scan across the Earth; the upcoming eclipse is number 22 in a series of 77 that began in 1639 AD with an eclipse near the North Pole and will end in 3009 AD with an eclipse near the South Pole.)

Any given moment in time will be in the middle of quite a few saros series (right now it’s 40)—and successive eclipses will always come from different series. But knowing about the saros cycle is a great first step in predicting eclipses—and it’s for example what the Antikythera device uses. In a sense, it’s a quintessential piece of science: take many observations, then synthesize a theory from them, or a least a scheme for computation.

It’s not clear what the Babylonians thought about abstract, formal systems. But the Greeks were definitely into them. And by 300 BC Euclid had defined his abstract system for geometry. So when someone like Ptolemy did astronomy, they did it a bit like Euclid—effectively taking things like the saros cycle as axioms, and then proving from them often surprisingly elaborate geometrical theorems, such as that there must be at least two solar eclipses in a given year.

Ptolemy’s Almagest from around 150 AD is an impressive piece of work, containing among many other things some quite elaborate procedures—and explicit tables—for predicting eclipses. (Yes, even in the later printed version, numbers are still represented confusingly by letters, as they always were in ancient Greek.)

Ptolemy's Almagest

In Ptolemy’s astronomy, Earth was assumed to be at the center of everything. But in modern terms that just meant he was choosing to use a different coordinate system—which didn’t affect most of the things he wanted to do, like working out the geometry of eclipses. And unlike the mainline Greek philosophers he wasn’t so much trying to make a fundamental theory of the world, but just wanted whatever epicycles and so on he needed to explain what he observed.

The Dawn of Modern Science

For more than a thousand years Ptolemy’s theory of the Moon defined the state of the art. In the 1300s Ibn al-Shatir revised Ptolemy’s models, achieving somewhat better accuracy. In 1472 Regiomontanus (Johannes Müller), systematizer of trigonometry, published more complete tables as part of his launch of what was essentially the first-ever scientific publishing company. But even in 1543 when Nicolaus Copernicus introduced his Sun-centered model of the solar system, the results he got were basically the same as Ptolemy’s, even though his underlying description of what was going on was quite different.

It’s said that Tycho Brahe got interested in astronomy in 1560 at age 13 when he saw a solar eclipse that had been predicted—and over the next several decades his careful observations uncovered several effects in the motion of the Moon (such as speeding up just before a full moon)—that eventually resulted in perhaps a factor 5 improvement in the prediction of its position. To Tycho eclipses were key tests, and he measured them carefully, and worked hard to be able to predict their timing more accurately than to within a few hours. (He himself never saw a total solar eclipse, only partial ones.)

Tycho's pages

Armed with Tycho’s observations, Johannes Kepler developed his description of orbits as ellipses—introducing concepts like inclination and eccentric anomaly—and in 1627 finally produced his Rudolphine Tables, which got right a lot of things that had been got wrong before, and included all sorts of detailed tables of lunar positions, as well as vastly better predictions for eclipses.

Kepler's Rudolphine Tables

Using Kepler’s Rudolphine Tables (and a couple of pages of calculations)—the first known actual map of a solar eclipse was published in 1654. And while there are some charming inaccuracies in overall geography, the geometry of the eclipse isn’t too bad.

The first map of an eclipse

 

Whether it was Ptolemy’s epicycles or Kepler’s ellipses, there were plenty of calculations to do in determining the motions of heavenly bodies (and indeed the first known mechanical calculator—excepting the Antikythera device—was developed by a friend of Kepler’s, presumably for the purpose). But there wasn’t really a coherent underlying theory; it was more a matter of describing effects in ways that could be used to make predictions.

So it was a big step forward in 1687 when Isaac Newton published his Principia, and claimed that with his laws for motion and gravity it should be possible—essentially from first principles—to calculate everything about the motion of the Moon. (Charmingly, in his “Theory of the World” section he simply asserts as his Proposition XXII “That all the motions of the Moon… follow from the principles which we have laid down.”)

Newton was proud of the fact that he could explain all sorts of known effects on the basis of his new theory. But when it came to actually calculating the detailed motion of the Moon he had a frustrating time. And even after several years he still couldn’t get the right answer—in later editions of the Principia adding the admission that actually “The apse of the Moon is about twice as swift” (i.e. his answer was wrong by a factor of 2).

Still, in 1702 Newton was happy enough with his results that he allowed them to be published, in the form of a 20-page booklet on the “Theory of the Moon”, which proclaimed that “By this Theory, what by all Astronomers was thought most difficult and almost impossible to be done, the Excellent Mr. Newton hath now effected, viz. to determine the Moon’s Place even in her Quadratures, and all other Parts of her Orbit, besides the Syzygys, so accurately by Calculation, that the Difference between that and her true Place in the Heavens shall scarce be two Minutes…”

Newton's "Theory of the Moon" booklet

 

Newton didn’t explain his methods (and actually it’s still not clear exactly what he did, or how mathematically rigorous it was or wasn’t). But his booklet effectively gave a step-by-step algorithm to compute the position of the Moon. He didn’t claim it worked “at the syzygys” (i.e. when the Sun, Moon and Earth are lined up for an eclipse)—though his advertised error of two arc-minutes was still much smaller than the angular size of the Moon in the sky.

But it wasn’t eclipses that were the focus then; it was a very practical problem of the day: knowing the location of a ship out in the open ocean. It’s possible to determine what latitude you’re at just by measuring how high the Sun gets in the sky. But to determine longitude you have to correct for the rotation of the Earth—and to do that you have to accurately keep track of time. But back in Newton’s day, the clocks that existed simply weren’t accurate enough, especially when they were being tossed around on a ship.

But particularly after various naval accidents, the problem of longitude was deemed important enough that the British government in 1714 established a “Board of Longitude” to offer prizes to help get it solved. One early suggestion was to use the regularity of the moons of Jupiter discovered by Galileo as a way to tell time. But it seemed that a simpler solution (not requiring a powerful telescope) might just be to measure the position of our moon, say relative to certain fixed stars—and then to back-compute the time from this.

But to do this one had to have an accurate way to predict the motion of the Moon—which is what Newton was trying to provide. In reality, though, it took until the 1760s before tables were produced that were accurate enough to be able to determine time to within a minute (and thus distance to within 15 miles or so). And it so happens that right around the same time a marine chronometer was invented that was directly able to keep good time.

The Three-Body Problem

One of Newton’s great achievements in the Principia was to solve the so-called two-body problem, and to show that with an inverse square law of gravity the orbit of one body around another must always be what Kepler had said: an ellipse.

In a first approximation, one can think of the Moon as just orbiting the Earth in a simple elliptical orbit. But what makes everything difficult is that that’s just an approximation, because in reality there’s also a gravitational pull on the Moon from the Sun. And because of this, the Moon’s orbit is no longer a simple fixed ellipse—and in fact it ends up being much more complicated. There are a few definite effects one can describe and reason about. The ellipse gets stretched when the Earth is closer to the Sun in its own orbit. The orientation of the ellipse precesses like a top as a result of the influence of the Sun. But there’s no way in the end to work out the orbit by pure reasoning—so there’s no choice but to go into the mathematics and start solving the equations of the three-body problem.

In many ways this represented a new situation for science. In the past, one hadn’t ever been able to go far without having to figure out new laws of nature. But here the underlying laws were supposedly known, courtesy of Newton. Yet even given these laws, there was difficult mathematics involved in working out the behavior they implied.

Over the course of the 1700s and 1800s the effort to try to solve the three-body problem and determine the orbit of the Moon was at the center of mathematical physics—and attracted a veritable who’s who of mathematicians and physicists.

An early entrant was Leonhard Euler, who developed methods based on trigonometric series (including much of our current notation for such things), and whose works contain many immediately recognizable formulas:

Euler's methods

 

In the mid-1740s there was a brief flap—also involving Euler’s “competitors” Clairaut and d’Alembert—about the possibility that the inverse-square law for gravity might be wrong. But the problem turned out to be with the calculations, and by 1748 Euler was using sums of about 20 trigonometric terms and proudly proclaiming that the tables he’d produced for the three-body problem had predicted the time of a total solar eclipse to within minutes. (Actually, he had said there’d be 5 minutes of totality, whereas in reality there was only 1—but he blamed this error on incorrect coordinates he’d been given for Berlin.)

Mathematical physics moved rapidly over the next few decades, with all sorts of now-famous methods being developed, notably by people like Lagrange. And by the 1770s, for example, Lagrange’s work was looking just like it could have come from a modern calculus book (or from a Wolfram|Alpha step-by-step solution):

Lagrange's methods

 

Particularly in the hands of Laplace there was increasingly obvious success in deriving the observed phenomena of what he called “celestial mechanics” from mathematics—and in establishing the idea that mathematics alone could indeed generate new results in science.

At a practical level, measurements of things like the position of the Moon had always been much more accurate than calculations. But now they were becoming more comparable—driving advances in both. Meanwhile, there was increasing systematization in the production of ephemeris tables. And in 1767 the annual publication began of what was for many years the standard: the British Nautical Almanac.

The almanac quoted the position of the Moon to the arc-second, and systematically achieved at least arc-minute accuracy. The primary use of the almanac was for navigation (and it was what started the convention of using Greenwich as the “prime meridian” for measuring time). But right at the front of each year’s edition were the predicted times of the eclipses for that year—in 1767 just two solar eclipses:

Nautical Almanac

 

The Math Gets More Serious

At a mathematical level, the three-body problem is about solving a system of three ordinary differential equations that give the positions of the three bodies as a function of time. If the positions are represented in standard 3D Cartesian coordinates
ri={xi, yi, zi}, the equations can be stated in the form:

Three equations

 

The {x,y,z} coordinates here aren’t, however, what traditionally show up in astronomy. For example, in describing the position of the Moon one might use longitude and latitude on a sphere around the Earth. Or, given that one knows the Moon has a roughly elliptical orbit, one might instead choose to describe its motions by variables that are based on deviations from such an orbit. In principle it’s just a matter of algebraic manipulation to restate the equations with any given choice of variables. But in practice what comes out is often long and complex—and can lead to formulas that fill many pages.

But, OK, so what are the best kind of variables to use for the three-body problem? Maybe they should involve relative positions of pairs of bodies. Or relative angles. Or maybe positions in various kinds of rotating coordinate systems. Or maybe quantities that would be constant in a pure two-body problem. Over the course of the 1700s and 1800s many treatises were written exploring different possibilities.

But in essentially all cases the ultimate approach to the three-body problem was the same. Set up the problem with the chosen variables. Identify parameters that, if set to zero, would make the problem collapse to some easy-to-solve form. Then do a series expansion in powers of these parameters, keeping just some number of terms.

By the 1860s Charles Delaunay had spent 20 years developing the most extensive theory of the Moon in this way. He’d identified five parameters with respect to which to do his series expansions (eccentricities, inclinations, and ratios of orbit sizes)—and in the end he generated about 1800 pages like this (yes, he really needed Mathematica!):

Delaunay's series expansions

 

But the sad fact was that despite all this effort, he didn’t get terribly good answers. And eventually it became clear why. The basic problem was that Delaunay wanted to represent his results in terms of functions like sin and cos. But in his computations, he often wanted to do series expansions with respect to the frequencies of those functions. Here’s a minimal case:

Series[Sin[(ω + δ)*t], {δ, 0, 3}]

Series[Sin[(ω + δ)*t], {δ, 0, 3}]

And here’s the problem. Take a look even at the second term. Yes, the δ parameter may be small. But how about the t parameter, standing for time? If you don’t want to make predictions very far out, that’ll stay small. But what if you want to figure out what will happen further in the future?

Well eventually that term will get big. And higher-order terms will get even bigger. But unless the Moon is going to escape its orbit or something, the final mathematical expressions that represent its position can’t have values that are too big. So in these expressions the so-called secular terms that increase with t must somehow cancel out.

But the problem is that at any given order in the series expansion, there’s no guarantee that will happen in a numerically useful way. And in Delaunay’s case—even though with immense effort he often went to 7th order or beyond—it didn’t.

One nice feature of Delaunay’s computation was that it was in a sense entirely algebraic: everything was done symbolically, and only at the very end were actual numerical values of parameters substituted in.

But even before Delaunay, Peter Hansen had taken a different approach—substituting numbers as soon as he could, and dropping terms based on their numerical size rather than their symbolic form. His presentations look less pure (notice things like all those t1800, where t is the time in years), and it’s more difficult to tell what’s going on. But as a practical matter, his results were much better, and in fact were used for many national almanacs from about 1862 to 1922, achieving errors as small as 1 or 2 arc-seconds at least over periods of a decade or so. (Over longer periods, the errors could rapidly increase because of the lack of terms that had been dropped as a result of what amounted to numerical accidents.)

Hansen's method

 

Both Delaunay and Hansen tried to represent orbits as series of powers and trigonometric functions (so-called Poisson series). But in the 1870s, George Hill in the US Nautical Almanac Office proposed instead using as a basis numerically computed functions that came from solving an equation for two-body motion with a periodic driving force of roughly the kind the Sun exerts on the Moon’s orbit. A large-scale effort was mounted, and starting in 1892 Ernest W. Brown (who had moved to the US, but had been a student of George Darwin, Charles Darwin’s physicist son) took charge of the project and in 1918 produced what would stand for many years as the definitive “Tables of the Motion of the Moon”.

Brown’s tables consist of hundreds of pages like this—ultimately representing the position of the Moon as a combination of about 1400 terms with very precise coefficients:

Brown's tables

 

He says right at the beginning that the tables aren’t particularly intended for unique events like eclipses, but then goes ahead and does a “worked example” of computing an eclipse from 381 BC, reported by Ptolemy:

Brown's tables 2

 

It was an impressive indication of how far things had come. But ironically enough the final presentation of Brown’s tables had the same sum-of-trigonometric-functions form that one would get from having lots of epicycles. At some level it’s not surprising, because any function can ultimately be represented by epicycles, just as it can be represented by a Fourier or other series. But it’s a strange quirk of history that such similar forms were used.

Can the Three-Body Problem Be Solved?

It’s all well and good that one can find approximations to the three-body problem, but what about just finding an outright solution—like as a mathematical formula? Even in the 1700s, there’d been some specific solutions found—like Euler’s collinear configuration, and Lagrange’s equilateral triangle. But a century later, no further solutions had been found—and finding a complete solution to the three-body problem was beginning to seem as hopeless as trisecting an angle, solving the quintic, or making a perpetual motion machine. (That sentiment was reflected for example in a letter Charles Babbage wrote Ada Lovelace in 1843 mentioning the “horrible problem [of] the three bodies”—even though this letter was later misinterpreted by Ada’s biographers to be about a romantic triangle, not the three-body problem of celestial mechanics.)

In contrast to the three-body problem, what seemed to make the two-body problem tractable was that its solutions could be completely characterized by “constants of the motion”—quantities that stay constant with time (in this case notably the direction of the axis of the ellipse). So for many years one of the big goals with the three-body problem was to find constants of the motion.

In 1887, though, Heinrich Bruns showed that there couldn’t be any such constants of the motion, at least expressible as algebraic functions of the standard {x,y,z} position and velocity coordinates of the three bodies. Then in the mid-1890s Henri Poincaré showed that actually there couldn’t be any constants of the motion that were expressible as any analytic functions of the positions, velocities and mass ratios.

One reason that was particularly disappointing at the time was that it had been hoped that somehow constants of the motion would be found in n-body problems that would lead to a mathematical proof of the long-term stability of the solar system. And as part of his work, Poincaré also saw something else: that at least in particular cases of the three-body problem, there was arbitrarily sensitive dependence on initial conditions—implying that even tiny errors in measurement could be amplified to arbitrarily large changes in predicted behavior (the classic “chaos theory” phenomenon).

But having discovered that particular solutions to the three-body problem could have this kind of instability, Poincaré took a different approach that would actually be characteristic of much of pure mathematics going forward: he decided to look not at individual solutions, but at the space of all possible solutions. And needless to say, he found that for the three-body problem, this was very complicated—though in his efforts to analyze it he invented the field of topology.

Poincaré’s work all but ended efforts to find complete solutions to the three-body problem. It also seemed to some to explain why the series expansions of Delaunay and others hadn’t worked out—though in 1912 Karl Sundman did show that at least in principle the three-body problem could be solved in terms of an infinite series, albeit one that converges outrageously slowly.

But what does it mean to say that there can’t be a solution to the three-body problem? Galois had shown that there couldn’t be a solution to the generic quintic equation, at least in terms of radicals. But actually it’s still perfectly possible to express the solution in terms of elliptic or hypergeometric functions. So why can’t there be some more sophisticated class of functions that can be used to just “solve the three-body problem”?

Here are some pictures of what can actually happen in the three-body problem, with various initial conditions:

Three body problem pictures

eqns = {Subscript[m, 1]*
     Derivative[2][Subscript[r, 1]][
      t] == -((Subscript[m, 1]*
          Subscript[m, 2]*(Subscript[r, 1][t] - Subscript[r, 2][t]))/
        Norm[Subscript[r, 1][t] - Subscript[r, 2][t]]^3) - (Subscript[
         m, 1]*Subscript[m, 
         3]*(Subscript[r, 1][t] - Subscript[r, 3][t]))/
      Norm[Subscript[r, 1][t] - Subscript[r, 3][t]]^3, 
   Subscript[m, 2]*
     Derivative[2][Subscript[r, 2]][
      t] == -((Subscript[m, 1]*
          Subscript[m, 2]*(Subscript[r, 2][t] - Subscript[r, 1][t]))/
        Norm[Subscript[r, 2][t] - Subscript[r, 1][t]]^3) - (Subscript[
         m, 2]*Subscript[m, 
         3]*(Subscript[r, 2][t] - Subscript[r, 3][t]))/
      Norm[Subscript[r, 2][t] - Subscript[r, 3][t]]^3, 
   Subscript[m, 3]*
     Derivative[2][Subscript[r, 3]][
      t] == -((Subscript[m, 1]*
          Subscript[m, 3]*(Subscript[r, 3][t] - Subscript[r, 1][t]))/
        Norm[Subscript[r, 3][t] - Subscript[r, 1][t]]^3) - (Subscript[
         m, 2]*Subscript[m, 
         3]*(Subscript[r, 3][t] - Subscript[r, 2][t]))/
      Norm[Subscript[r, 3][t] - Subscript[r, 2][t]]^3};
(SeedRandom[#]; {Subscript[m, 1], Subscript[m, 2], Subscript[m, 3]} = 
    RandomReal[{0, 1}, 3]; 
   inits = Table[{Subscript[r, i][0] == RandomReal[{-1, 1}, 3], 
      Subscript[r, i]'[0] == RandomReal[{-1, 1}, 3]}, {i, 3}]; 
   sols = NDSolve[{eqns, inits}, {Subscript[r, 1], Subscript[r, 2], 
      Subscript[r, 3]}, {t, 0, 100}]; 
   ParametricPlot3D[{Subscript[r, 1][t], Subscript[r, 2][t], 
      Subscript[r, 3][t]} /. sols, {t, 0, 100}, 
    Ticks -> None]) & /@ {776, 5742, 6711, 2300, 5281, 9225}

And looking at these immediately gives some indication of why it’s not easy to just “solve the three-body problem”. Yes, there are cases where what happens is fairly simple. But there are also cases where it’s not, and where the trajectories of the three bodies continue to be complicated and tangled for a long time.

So what’s fundamentally going on here? I don’t think traditional mathematics is the place to look. But I think what we’re seeing is actually an example of a general phenomenon I call computational irreducibility that I discovered in the 1980s in studying the computational universe of possible programs.

Many programs, like many instances of the three-body problem, behave in quite simple ways. But if you just start looking at all possible simple programs, it doesn’t take long before you start seeing behavior like this:

Cellular automaton array

ArrayPlot[ CellularAutomaton[{#, 3, 1}, {{2}, 0}, 100], 
   ImageSize -> {Automatic, 100}] & /@ {5803305107286, 2119737824118, 
  5802718895085, 4023376322994, 6252890585925}

How can one tell what’s going to happen? Well, one can just keep explicitly running each program and seeing what it does. But the question is: is there some systematic way to jump ahead, and to predict what will happen without tracing through all the steps?

The answer is that in general there isn’t. And what I call the Principle of Computational Equivalence suggests that pretty much whenever one sees complex behavior, there won’t be.

Here’s the way to think about this. The system one’s studying is effectively doing a computation to work out what its behavior will be. So to jump ahead we’d in a sense have to do a more sophisticated computation. But what the Principle of Computational Equivalence says is that actually we can’t—and that whether we’re using our brains or our mathematics or a Turing machine or anything else, we’re always stuck with computations of the same sophistication.

So what about the three-body problem? Well, I strongly suspect that it’s an example of computational irreducibility: that in effect the computations it’s doing are as sophisticated as any computations that we can do, so there’s no way we can ever expect to systematically jump ahead and solve the problem. (We also can’t expect to just define some new finite class of functions that can just be evaluated to give the solution.)

I’m hoping that one day someone will rigorously prove this. There’s some technical difficulty, because the three-body problem is usually formulated in terms of real numbers that immediately have an infinite number of digits—but to compare with ordinary computation one has to require finite processes to set up initial conditions. (Ultimately one wants to show for example that there’s a “compiler” that can go from any program, say for a Turing machine, and can generate instructions to set up initial conditions for a three-body problem so that the evolution of the three-body problem will give the same results as running that program—implying that the three-body problem is capable of universal computation.)

I have to say that I consider Newton in a sense very lucky. It could have been that it wouldn’t have been possible to work out anything interesting from his theory without encountering the kind of difficulties he had with the motion of the Moon—because one would always be running into computational irreducibility. But in fact, there was enough computational reducibility and enough that could be computed easily that one could see that the theory was useful in predicting features of the world (and not getting wrong answers, like with the apse of the Moon)—even if there were some parts that might take two centuries to work out, or never be possible at all.

Newton himself was certainly aware of the potential issue, saying that at least if one was dealing with gravitational interactions between many planets then “to define these motions by exact laws admitting of easy calculation exceeds, if I am not mistaken, the force of any human mind”. And even today it’s extremely difficult to know what the long-term evolution of the solar system will be.

It’s not particularly that there’s sensitive dependence on initial conditions: we actually have measurements that should be precise enough to determine what will happen for a long time. The problem is that we just have to do the computation—a bit like computing the digits of π—to work out the behavior of the n-body problem that is our solar system.

Existing simulations show that for perhaps a few tens of millions of years, nothing too dramatic can happen. But after that we don’t know. Planets could change their order. Maybe they could even collide, or be ejected from the solar system. Computational irreducibility implies that at least after an infinite time it’s actually formally undecidable (in the sense of Gödel’s Theorem or the Halting Problem) what can happen.

One of my children, when they were very young, asked me whether when dinosaurs existed the Earth could have had two moons. For years when I ran into celestial mechanics experts I would ask them that question—and it was notable how difficult they found it. Most now say that at least at the time of the dinosaurs we couldn’t have had an extra moon—though a billion years earlier it’s not clear.

We used to only have one system of planets to study. And the fact that there were (then) 9 of them used to be a classic philosopher’s example of a truth about the world that just happens to be the way it is, and isn’t “necessarily true” (like 2+2=4). But now of course we know about lots of exoplanets. And it’s beginning to look as if there might be a theory for things like how many planets a solar system is likely to have.

At some level there’s presumably a process like natural selection: some configurations of planets aren’t “fit enough” to be stable—and only those that are survive. In biology it’s traditionally been assumed that natural selection and adaptation is somehow what’s led to the complexity we see. But actually I suspect much of it is instead just a reflection of what generally happens in the computational universe—both in biology and in celestial mechanics. Now in celestial mechanics, we haven’t yet seen in the wild any particularly complex forms (beyond a few complicated gap structures in rings, and tumbling moons and asteroids). But perhaps elsewhere we’ll see things like those obviously tangled solutions to the three-body problem—that come closer to what we’re used to in biology.

It’s remarkable how similar the issues are across so many different fields. For example, the whole idea of using “perturbation theory” and series expansions that has existed since the 1700s in celestial mechanics is now also core to quantum field theory. But just like in celestial mechanics there’s trouble with convergence (maybe one should try renormalization or resummation in celestial mechanics). And in the end one begins to realize that there are phenomena—no doubt like turbulence or the three-body problem—that inevitably involve more sophisticated computations, and that need to be studied not with traditional mathematics of the kind that was so successful for Newton and his followers but with the kind of science that comes from exploring the computational universe.

Approaching Modern Times

But let’s get back to the story of the motion of the Moon. Between Brown’s tables, and Poincaré’s theoretical work, by the beginning of the 1900s the general impression was that whatever could reasonably be computed about the motion of the Moon had been computed.

Occasionally there were tests. Like for example in 1925, when there was a total solar eclipse visible in New York City, and the New York Times perhaps overdramatically said that “scientists [were] tense… wondering whether they or moon is wrong as eclipse lags five seconds behind”. The fact is that a prediction accurate to 5 seconds was remarkably good, and we can’t do all that much better even today. (By the way, the actual article talks extensively about “Professor Brown”—as well as about how the eclipse might “disprove Einstein” and corroborate the existence of “coronium”—but doesn’t elaborate on the supposed prediction error.)

Newspaper article

 

As a practical matter, Brown’s tables were not exactly easy to use: to find the position of the Moon from them required lots of mechanical desk calculator work, as well as careful transcription of numbers. And this led Leslie Comrie in 1932 to propose using a punch-card-based IBM Hollerith automatic tabulator—and with the help of Thomas Watson, CEO of IBM, what was probably the first “scientific computing laboratory” was established—to automate computations from Brown’s tables.

Automatic tabulation

 

(When I was in elementary school in England in the late 1960s—before electronic calculators—I always carried around, along with my slide rule, a little book of “4-figure mathematical tables”. I think I found it odd that such a book would have an author—and perhaps for that reason I still remember the name: “L. J. Comrie”.)

By the 1950s, the calculations in Brown’s tables were slowly being rearranged and improved to make them more suitable for computers. But then with John F. Kennedy’s 1962 “We choose to go the Moon”, there was suddenly urgent interest in getting the most accurate computations of the Moon’s position. As it turned out, though, it was basically just a tweaked version of Brown’s tables, running on a mainframe computer, that did the computations for the Apollo program.

At first, computers were used in celestial mechanics purely for numerical computation. But by the mid-1960s there were also experiments in using them for algebraic computation, and particularly to automate the generation of series expansions. Wallace Eckert at IBM started using FORMAC to redo Brown’s tables, while in Cambridge David Barton and Steve Bourne (later the creator of the “Bourne shell” (sh) in Unix) built their own CAMAL computer algebra system to try extending the kind of thing Delaunay had done. (And by 1970, Delaunay’s 7th-order calculations had been extended to 20th order.)

When I myself started to work on computer algebra in 1976 (primarily for computations in particle physics), I’d certainly heard about CAMAL, but I didn’t know what it had been used for (beyond vaguely “celestial mechanics”). And as a practicing theoretical physicist in the late 1970s, I have to say that the “problem of the Moon” that had been so prominent in the 1700s and 1800s had by then fallen into complete obscurity.

I remember for example in 1984 asking a certain Martin Gutzwiller, who was talking about quantum chaos, what his main interest actually was. And when he said “the problem of the Moon”, I was floored; I didn’t know there still was any problem with the Moon. As it turns, in writing this post I found out that Gutzwiller was actually the person who took over from Eckert and spent nearly two decades working on trying to improve the computations of the position of the Moon.

Why Not Just Solve It?

Traditional approaches to the three-body problem come very much from a mathematical way of thinking. But modern computational thinking immediately suggests a different approach. Given the differential equations for the three-body problem, why not just directly solve them? And indeed in the Wolfram Language there’s a built-in function NDSolve for numerically solving systems of differential equations.

So what happens if one just feeds in equations for a three-body problem? Well, here are the equations:

eqns = {Subscript[m,       1] (Subscript[r, 1]^\[Prime]\[Prime])[t] == -((       Subscript[m, 1] Subscript[m,         2] (Subscript[r, 1][t] - Subscript[r, 2][t]))/       Norm[Subscript[r, 1][t] - Subscript[r, 2][t]]^3) - (      Subscript[m, 1] Subscript[m,        3] (Subscript[r, 1][t] - Subscript[r, 3][t]))/      Norm[Subscript[r, 1][t] - Subscript[r, 3][t]]^3,     Subscript[m,       2] (Subscript[r, 2]^\[Prime]\[Prime])[t] == -((       Subscript[m, 1] Subscript[m,         2] (Subscript[r, 2][t] - Subscript[r, 1][t]))/       Norm[Subscript[r, 2][t] - Subscript[r, 1][t]]^3) - (      Subscript[m, 2] Subscript[m,        3] (Subscript[r, 2][t] - Subscript[r, 3][t]))/      Norm[Subscript[r, 2][t] - Subscript[r, 3][t]]^3,     Subscript[m,       3] (Subscript[r, 3]^\[Prime]\[Prime])[t] == -((       Subscript[m, 1] Subscript[m,         3] (Subscript[r, 3][t] - Subscript[r, 1][t]))/       Norm[Subscript[r, 3][t] - Subscript[r, 1][t]]^3) - (      Subscript[m, 2] Subscript[m,        3] (Subscript[r, 3][t] - Subscript[r, 2][t]))/      Norm[Subscript[r, 3][t] - Subscript[r, 2][t]]^3};

eqns = {Subscript[m, 
     1] (Subscript[r, 1]^\[Prime]\[Prime])[t] == -((
      Subscript[m, 1] Subscript[m, 
       2] (Subscript[r, 1][t] - Subscript[r, 2][t]))/
      Norm[Subscript[r, 1][t] - Subscript[r, 2][t]]^3) - (
     Subscript[m, 1] Subscript[m, 
      3] (Subscript[r, 1][t] - Subscript[r, 3][t]))/
     Norm[Subscript[r, 1][t] - Subscript[r, 3][t]]^3, 
   Subscript[m, 
     2] (Subscript[r, 2]^\[Prime]\[Prime])[t] == -((
      Subscript[m, 1] Subscript[m, 
       2] (Subscript[r, 2][t] - Subscript[r, 1][t]))/
      Norm[Subscript[r, 2][t] - Subscript[r, 1][t]]^3) - (
     Subscript[m, 2] Subscript[m, 
      3] (Subscript[r, 2][t] - Subscript[r, 3][t]))/
     Norm[Subscript[r, 2][t] - Subscript[r, 3][t]]^3, 
   Subscript[m, 
     3] (Subscript[r, 3]^\[Prime]\[Prime])[t] == -((
      Subscript[m, 1] Subscript[m, 
       3] (Subscript[r, 3][t] - Subscript[r, 1][t]))/
      Norm[Subscript[r, 3][t] - Subscript[r, 1][t]]^3) - (
     Subscript[m, 2] Subscript[m, 
      3] (Subscript[r, 3][t] - Subscript[r, 2][t]))/
     Norm[Subscript[r, 3][t] - Subscript[r, 2][t]]^3};

Now as an example let’s set the masses to random values:

{Subscript[m, 1], Subscript[m, 2], Subscript[m, 3]} = RandomReal[{0, 1}, 3]

{Subscript[m, 1], Subscript[m, 2], Subscript[m, 3]} = RandomReal[{0, 1}, 3]

And let’s define the initial position and velocity for each body to be random as well:

inits = Table[{Subscript[r, i][0] == RandomReal[{-1, 1}, 3], Derivative[1][Subscript[r, i]][0] == RandomReal[{-1, 1}, 3]}, {i, 3}]

inits = Table[{Subscript[r, i][0] == RandomReal[{-1, 1}, 3], Derivative[1][Subscript[r, i]][0] == RandomReal[{-1, 1}, 3]}, {i, 3}]

Now we can just use NDSolve to get the solutions (it gives them as implicit approximate numerical functions of t):

sols = NDSolve[{eqns, inits}, {Subscript[r, 1], Subscript[r, 2], Subscript[r, 3]}, {t, 0, 100}]

sols = NDSolve[{eqns, inits}, {Subscript[r, 1], Subscript[r, 2], Subscript[r, 3]}, {t, 0, 100}]

And now we can plot them. And now we’ve got a solution to a three-body problem, just like that!

ParametricPlot3D[Evaluate[{Subscript[r, 1][t], Subscript[r, 2][t], Subscript[r, 3][t]} /. First[sols]], {t, 0, 100}]

ParametricPlot3D[Evaluate[{Subscript[r, 1][t], Subscript[r, 2][t], Subscript[r, 3][t]} /. First[sols]], {t, 0, 100}]

Well, obviously this is using the Wolfram Language and a huge tower of modern technology. But would it have been possible even right from the beginning for people to generate direct numerical solutions to the three-body problem, rather than doing all that algebra? Back in the 1700s, Euler already knew what’s now called Euler’s method for finding approximate numerical solutions to differential equations. So what if he’d just used that method to calculate the motion of the Moon?

The method relies on taking a sequence of discrete steps in time. And if he’d used, say, a step size of a minute, then he’d have had to take 40,000 steps to get results for a month, but he should have been able to successfully reproduce the position of the Moon to about a percent. If he’d tried to extend to 3 months, however, then he would already have had at least a 10% error.

Any numerical scheme for solving differential equations in practice eventually builds up some kind of error—but the more one knows about the equations one’s solving, and their expected solutions, the more one’s able to preprocess and adapt things to minimize the error. NDSolve has enough automatic adaptivity built into it that it’ll do pretty well for a surprisingly long time on a typical three-body problem. (It helps that the Wolfram Language and NDSolve can handle numbers with arbitrary precision, not just machine precision.)

But if one looks, say, at the total energy of the three-body system—which one can prove from the equations should stay constant—then one will typically see an error slowly build up in it. One can avoid this if one effectively does a change of variables in the equations to “factor out” energy. And one can imagine doing a whole hierarchy of algebraic transformations that in a sense give the numerical scheme as much help as possible.

And indeed since at least the 1980s that’s exactly what’s been done in practical work on the three-body problem, and the Earth-Moon-Sun system. So in effect it’s a mixture of the traditional algebraic approach from the 1700s and 1800s, together with modern numerical computation.

The Real Earth-Moon-Sun Problem

OK, so what’s involved in solving the real problem of the Earth-Moon-Sun system? The standard three-body problem gives a remarkably good approximation to the physics of what’s happening. But it’s obviously not the whole story.

For a start, the Earth isn’t the only planet in the solar system. And if one’s trying to get sufficiently accurate answers, one’s going to have to take into account the gravitational effect of other planets. The most important is Jupiter, and its typical effect on the orbit of the Moon is at about the 10-5 level—sufficiently large that for example Brown had to take it into account in his tables.

The next effect is that the Earth isn’t just a point mass, or even a precise sphere. Its rotation makes it bulge at the equator, and that affects the orbit of the Moon at the 10-6 level.

Orbits around the Earth ultimately depend on the full mass distribution and gravitational field of the Earth (which is what Sputnik-1 was nominally launched to map)—and both this, and the reverse effect from the Moon, come in at the 10-8 level. At the 10-9 level there are then effects from tidal deformations (“solid tides”) on the Earth and moon, as well as from gravitational redshift and other general relativistic phenomena.

To predict the position of the Moon as accurately as possible one ultimately has to have at least some model for these various effects.

But there’s a much more immediate issue to deal with: one has to know the initial conditions for the Earth, Sun and Moon, or in other words, one has to know as accurately as possible what their positions and velocities were at some particular time.

And conveniently enough, there’s now a really good way to do that, because Apollo 11, 14 and 15 all left laser retroreflectors on the Moon. And by precisely timing how long it takes a laser pulse from the Earth to round-trip to these retroreflectors, it’s now possible in effect to measure the position of the Moon to millimeter accuracy.

OK, so how do modern analogs of the Babylonian ephemerides actually work? Internally they’re dealing with the equations for all the significant bodies in the solar system. They do symbolic preprocessing to make their numerical work as easy as possible. And then they directly solve the differential equations for the system, appropriately inserting models for things like the mass distribution in the Earth.

They start from particular measured initial conditions, but then they repeatedly insert new measurements, trying to correct the parameters of the model so as to optimally reproduce all the measurements they have. It’s very much like a typical machine learning task—with the training data here being observations of the solar system (and typically fitting just being least squares).

But, OK, so there’s a model one can run to figure out something like the position of the Moon. But one doesn’t want to have to explicitly do that every time one needs to get a result; instead one wants in effect just to store a big table of pre-computed results, and then to do something like interpolation to get any particular result one needs. And indeed that’s how it’s done today.

How It’s Really Done

Back in the 1960s NASA started directly solving differential equations for the motion of planets. The Moon was more difficult to deal with, but by the 1980s that too was being handled in a similar way. Ongoing data from things like the lunar retroreflectors was added, and all available historical data was inserted as well.

The result of all this was the JPL Development Ephemeris (JPL DE). In addition to new observations being used, the underlying system gets updated every few years, typically to get what’s needed for some spacecraft going to some new place in the solar system. (The latest is DE432, built for going to Pluto.)

But how is the actual ephemeris delivered? Well, for every thousand years covered, the ephemeris has about 100 megabytes of results, given as coefficients for Chebyshev polynomials, which are convenient for interpolation. And for any given quantity in any given coordinate system over a particular period of time, one accesses the appropriate parts of these results.

OK, but so how does one find an eclipse? Well, it’s an iterative process. Start with an approximation, perhaps from the saros cycle. Then interpolate the ephemeris and look at the result. Then keep iterating until one finds out just when the Moon will be in the appropriate position.

But actually there’s some more to do. Because what’s originally computed are the positions of the barycenters (centers of mass) of the various bodies. But now one has to figure out how the bodies are oriented.

The Earth rotates, and we know its rate quite precisely. But the Moon is basically locked with the same face pointing to the Earth, except that in practice there are small “librations” where the Moon wobbles a little back and forth—and these turn out to be particularly troublesome to predict.

Computing the Eclipse

OK, so let’s say one knows where the Earth, Moon and Sun are. How does one then figure out where on the Earth the eclipse will actually hit? Well, there’s some further geometry to do. Basically, the Moon generates a cone of shadow in a direction defined by the location of the Sun, and what’s then needed is to figure out how the surface of the Earth intersects that cone.

In 1824 Friedrich Bessel suggested in effect inverting the problem by using the shadow cone to define a coordinate system in which to specify the positions of the Sun and Moon. The resulting so-called Besselian elements provide a convenient summary of the local geometry of an eclipse—with respect to which its path can be defined.

OK, but so how does one figure out at what time an eclipse will actually reach a given point on Earth? Well, first one has to be clear on one’s definition of time. And there’s an immediate issue with the speed of light and special relativity. What does it mean to say that the positions of the Earth and Sun are such-and-such at such-and-such a time? Because it takes light about 8 minutes to get to the Earth from the Sun, we only get to see where the Sun was 8 minutes ago, not where it is now.

And what we need is really a classic special relativity setup. We essentially imagine that the solar system is filled with a grid of clocks that have been synchronized by light pulses. And what a modern ephemeris does is to quote the results for positions of bodies in the solar system relative to the times on those clocks. (General relativity implies that in different gravitational fields the clocks will run at different rates, but for our purposes this is a tiny effect. But what isn’t a tiny effect is including retardation in the equations for the n-body problem—making them become delay differential equations.)

But now there’s another issue. If one’s observing the eclipse, one’s going to be using some timepiece (phone?) to figure out what time it is. And if it’s working properly that timepiece should show official “civil time” that’s based on UTC—which is what NTP internet time is synchronized to. But the issue is that UTC has a complicated relationship to the time used in the astronomical ephemeris.

The starting point is what’s called UT1: a definition of time in which one day is the average time it takes the Earth to rotate once relative to the Sun. But the point is that this average time isn’t constant, because the rotation of the Earth is gradually slowing down, primarily as a result of interactions with the Moon. But meanwhile, UTC is defined by an atomic clock whose timekeeping is independent of any issues about the rotation of the Earth.

There’s a convention for keeping UT1 aligned with UTC: if UT1 is going to get more than 0.9 seconds away from UTC, then a leap second is added to UTC. One might think this would be a tiny effect, but actually, since 1972, a total of 27 leap seconds have been added. Exactly when a new leap second will be needed is unpredictable; it depends on things like what earthquakes have occurred. But we need to account for leap seconds if we’re going to get the time of the eclipse correct to the second relative to UTC or internet time.

There are a few other effects that are also important in the precise observed timing of the eclipse. The most obvious is geo elevation. In doing astronomical computations, the Earth is assumed to be an ellipsoid. (There are many different definitions, corresponding to different geodetic “datums”—and that’s an issue in defining things like “sea level”, but it’s not relevant here.) But if you’re at a different height above the ellipsoid, the cone of shadow from the eclipse will reach you at a different time. And the size of this effect can be as much as 0.3 seconds for every 1000 feet of height.

All of the effects we’ve talked about we’re readily able to account for. But there is one remaining effect that’s a bit more difficult. Right at the beginning or end of totality one typically sees points of light on the rim of the Moon. Known as Baily’s beads, these are the result of rays of light that make it to us between mountains on the Moon. Figuring out exactly when all these rays are extinguished requires taking geo elevation data for the Moon, and effectively doing full 3D ray tracing. The effect can last as long as a second, and can cause the precise edge of totality to move by as much as a mile. (One can also imagine effects having to do with the corona of the Sun, which is constantly changing.)

But in the end, even though the shadow of the Moon on the Earth moves at more than 1000 mph, modern science successfully makes it possible to compute when the shadow will reach a particular point on Earth to an accuracy of perhaps a second. And that’s what our precisioneclipse.com website is set up to do.

Eclipse Experiences

I saw my first partial solar eclipse more than 50 years ago. And I’ve seen one total solar eclipse before in my life—in 1991. It was the longest eclipse (6 minutes 53 seconds) that’ll happen for more than a century.

There was a certain irony to my experience, though, especially in view of our efforts now to predict the exact arrival time of next week’s eclipse. I’d chartered a plane and flown to a small airport in Mexico (yes, that’s me on the left with the silly hat)—and my friends and I had walked to a beautiful deserted beach, and were waiting under a cloudless sky for the total eclipse to begin.

1991 eclipse trip photos

 

I felt proud of how prepared I was—with maps marking to the minute when the eclipse should arrive. But then I realized: there we were, out on a beach with no obvious signs of modern civilization—and nobody had brought any properly set timekeeping device (and in those days my cellphone was just a phone, and didn’t even have signal there).

And so it was that I missed seeing a demonstration of an impressive achievement of science. And instead I got to experience the eclipse pretty much the way people throughout history have experienced eclipses—even if I did know that the Moon would continue gradually eating into the Sun and eventually cover it, and that it wouldn’t make the world end.

There’s always something sobering about astronomical events, and about realizing just how tiny human scales are compared to them. Billions of eclipses have happened over the course of the Earth’s history. Recorded history has covered only a few thousand of them. On average, there’s an eclipse at any given place on Earth roughly every 400 years; in Jackson, WY, where I’m planning to see next week’s eclipse, it turns out the next total eclipse will be 727 years from now—in 2744.

In earlier times, civilizations built giant monuments to celebrate the motions of the Sun and moon. For the eclipse next week what we’re making is a website. But that website builds on one of the great epics of human intellectual history—stretching back to the earliest times of systematic science, and encompassing contributions from a remarkable cross-section of the most celebrated scientists and mathematicians from past centuries.

It’ll be about 9538 days since the eclipse I saw in 1991. The Moon will have traveled some 500 million miles around the Earth, and the Earth some 15 billion miles around the Sun. But now—in a remarkable triumph of science—we’re computing to the second when they’ll be lined up again.

]]>
9
Stephen Wolfram <![CDATA[High-School Summer Camp: A Two-Week Path to Computational Thinking]]> http://blog.internal.stephenwolfram.com/?p=13854 2017-09-14T19:29:18Z 2017-08-02T17:24:28Z High-School Summer Camp: A Two-Week Path to Computational ThinkingThe Summer Camp Was a Success! How far can one get in teaching computational thinking to high-school students in two weeks? Judging by the results of this year’s Wolfram High-School Summer Camp the answer is: remarkably far. I’ve been increasingly realizing what an immense and unique opportunity there now is to teach computational thinking with [...]]]> High-School Summer Camp: A Two-Week Path to Computational Thinking

The Summer Camp Was a Success!

How far can one get in teaching computational thinking to high-school students in two weeks? Judging by the results of this year’s Wolfram High-School Summer Camp the answer is: remarkably far.

I’ve been increasingly realizing what an immense and unique opportunity there now is to teach computational thinking with the whole stack of technology we’ve built up around the Wolfram Language. But it was a thrill to see just how well this seems to actually work with real high-school students—and to see the kinds of projects they managed to complete in only two weeks.

Wolfram Summer Camp 2017

We’ve been doing our high-school summer camp for 5 years now (as well as our 3-week Summer School for more experienced students for 15 years). And every time we do the camp, we figure out a little more. And I think that by now we really have it down—and we’re able to take even students who’ve never really been exposed to computation before, and by the end of the camp have them doing serious computational thinking—and fluently implementing their ideas by writing sometimes surprisingly sophisticated Wolfram Language code (as well as creating well-written notebooks and “computational essays” that communicate about what they’ve done).

Over the coming year, we’re going to be dramatically expanding our Computational Thinking Initiative, and working to bring analogs of the Summer Camp experience to as many students as possible. But the Summer Camp provides fascinating and important data about what’s possible.

The Setup for the Camp

So how did the Summer Camp actually work? We had a lot of applicants for the 40 slots we had available this year. Some had been pointed to the camp by parents, teachers, or previous attendees. But a large fraction had just seen mention of it in the Wolfram|Alpha sidebar. There were students from a range of kinds of schools around the US, and overseas (though we still have to figure out how to get more applicants from underserved populations). Our team had done interviews to pick the final students—and I thought the ones they’d selected were terrific.

Students at the Wolfram Summer Camp

The students’ past experience was quite diverse. Some were already accomplished programmers (almost always self-taught). Others had done a CS class or two. But quite a few had never really done anything computational before—even though they were often quite advanced in various STEM areas such as math. But almost regardless of background, it was striking to me how new the core concepts of computational thinking seemed to be to so many of the students.

How does one take an idea or a question about almost anything, and find a way to formulate it for a computer? To be fair, it’s only quite recently, with all the knowledge and automation that we’ve been able to build into the Wolfram Language, that it’s become realistic for kids to do these kinds of things for real. So it’s not terribly surprising that in their schools or elsewhere our students hadn’t really been exposed to such things before. But it’s now possible—and that means there’s a great new opportunity to seriously teach computational thinking to kids, and to position them to pursue the amazing range of directions that computational thinking is opening up.

It’s important, by the way, to distinguish between “computational thinking” and straight “coding”. Computational thinking is about formulating things in computational terms. Coding is about the actual mechanics of telling a computer what to do. One of our great goals with the Wolfram Language is to automate the process of coding as much as possible so people can concentrate on pure computational thinking. When one’s using lower-level languages, like C++ and Java, there’s no choice but to be involved with the detailed mechanics of coding. But with the Wolfram Language the exciting thing is that it’s possible to teach pure high-level computational thinking, without being forced to deal with the low-level mechanics of coding.

What does this mean in practice? I think it’s very empowering for students: as soon as they “get” a concept, they can immediately apply it, and do real things with it. And it was pretty neat at the Summer Camp to see how easily even students who’d never written programs before were able to express surprisingly sophisticated computational ideas in the Wolfram Language. Sometimes it seemed like students who’d learned a low-level language before were actually at a disadvantage. Though for me it was interesting a few times to witness the “aha” moment when a student realized that they didn’t have to break down their computations into tiny steps the way they’d been taught—and that they could turn some big blob of code they’d written into one simple line that they could immediately understand and extend.

Suggesting Projects

The Summer Camp program involves several hours each day of lectures and workshops aimed at bringing students up to speed with computational thinking and how to express it in the Wolfram Language. But the real core of the program is every student doing an individual, original, computational thinking project.

And, yes, this is a difficult thing to orchestrate. But over the years we’ve been doing our Summer School and Summer Camp we’ve developed a very successful way of setting this up. There are a bunch of pieces to it, and the details depend on the level of the students. But here let’s talk about high-school students, and this year’s Summer Camp.

Right before the camp we (well, actually, I) came up with a list of about 70 potential projects. Some are quite specific, some are quite open-ended, and some are more like “metaprojects” (e.g. pick a dataset in the Wolfram Data Repository and analyze it). Some are projects that could at least in some form already have been done quite a few years ago. But many projects have only just become possible—this year particularly as a result of all our recent advances in machine learning.

summer camp list

 

I tried to have a range of nominal difficulty levels for the projects. I say “nominal” because even a project that can in principle be done in an easy way can also always be done in a more elaborate and sophisticated way. I wanted to have projects that ranged from the extremely well defined and precise (implement a precise algorithm of this particular type), to ones that involved wrangling data or machine learning training, to ones that were basically free-form and where the student got to define the objective.

Many of the projects in this list might seem challenging for high-school students. But my calculation (which in fact worked out well) was that with the technology we now have, all of them are within range.

It’s perhaps interesting to compare the projects with what I suggested for this year’s Summer School. The Summer School caters to more experienced students—typically at the college, graduate school or postdoc level. And so I was able to suggest projects that require deeper mathematical or software engineering knowledge—or are just bigger, with a higher threshold to achieve a reasonable level of success.

summer school list

 

Matching Projects to Students

Before students start picking projects, it’s important that they understand what a finished project should look like, and what’s involved in doing it. So at the very beginning of the camp, the instructors went through projects from previous camps, and discussed what the “output” of a project should be. Maybe it’ll be an active website; maybe an interactive Demonstration; maybe it’ll be a research paper. It’s got to be possible to make a notebook that describes the project and its results, and to make a post about it for Wolfram Community.

After talking about the general idea of projects, and giving examples of previous ones, the instructors did a quick survey of this year’s suggestions list, filling in a few details of what the imagined projects actually were. After this, the students were asked to pick their top three projects from our list, and then invent two more potential projects of their own.

It’s always an interesting challenge to find the right project for each student—and it’s something I’ve personally been involved in at our Summer Camp for the past several years. (And, yes, it helps that I have decades of experience in organizing professional and research projects and figuring out the best people to do them.)

It’s taken us a few iterations, but here’s the approach we’ve found works well. First, we randomly break the students up into groups of a dozen or so. Then we meet with each group, going around the room and asking each student a little about themselves, their interests and goals—and their list of projects.

After we’re finished with each group, we meet separately and try to come up with a project for each student. Sometimes it’ll be one of the projects straight from our list. Sometimes it’ll be a project that the student themself suggested. And sometimes it’ll be some creative combination of these, or even something completely different based on what they said they were interested in.

After we think we’ve come up with a good project, the next step is to meet individually with each student and actually suggest it to them. It’s very satisfying that a lot of the time the students seem really enthused about the projects we end up suggesting. But sometimes it becomes clear that a project just isn’t a good fit—and then sometimes we modify it in real time, but more often we circle back later with a different suggestion.

Once the projects are set, we assign an appropriate mentor to each student, taking into account both the student and the subject of the project. And then things are off and running. We have various checkpoints, like that students have to write up descriptions of their projects and post them on the internal Summer Camp site.

I personally wasn’t involved in the actual execution of the projects (though I did have a chance to check in on a few of them). So it was pretty interesting for me to see at the end of the camp what had actually happened. It’s worth mentioning that our scheme is that mentors can make suggestions about projects, but all the final code in a project should be created by the student. And if one version of the project ends up being too difficult, it’s up to the mentor to simplify it. So however the final project comes out, it really is the student’s work.

Much of the time, the Summer Camp will be the first time students have ever done an original project. It could potentially seem daunting. But I think the fact that we give so many examples of other projects, and that everyone else at the camp is also doing a project, really helps. And in the end experiencing the whole process of going from the idea for a project to a real, finished project is incredibly educational—and seems to have a big effect on many of our students.

A Few Case Studies

OK, so that’s the theory. So what actually happened at this year’s Summer Camp? Well, here are all the projects the students did, with the titles they gave them:

final projects list

 

It’s a very interesting, impressive, and diverse list. But let me pick out a few semi-randomly to discuss in a bit more detail. Consider these as “case studies” for what high-school students can accomplish with the Wolfram Language in a couple of weeks at a summer camp.

Routing Airplanes around Mountains

One young man at our camp had quite a lot of math background, and told me he was interested in airplanes and flying, and had designed his own remote-control plane. I started thinking about all sorts of drone survey projects. But he didn’t have a drone with him—and we had to come up with a project that could actually be done in a couple of weeks. So I ended up suggesting the following: given two points on Earth, find how an airplane can get from one to the other by the shortest path that never needs to go above a certain altitude. (And, yes, a small-scale version of this would be relevant to things like drone surveying too.)

Here’s how the student did this project. First, he realized that one could think of possible flight paths as edges on a graph whose nodes are laid out on a grid on the Earth. Then he used the built-in GeoElevationData to delete nodes that couldn’t be visited because the elevation at that point was above the cutoff. Then he just used FindShortestPath to find the shortest path in the graph from the start to the end.

I thought this was a pretty clever solution. It was a nice piece of computational thinking to realize that the elements of paths could be thought of as edges on a graph with nodes removed. Needless to say, there were some additional details to get a really good result. First, the student added in diagonal connections on the grid, with appropriate weightings to still get the correct shortest path computation. And then he refined the path by successively merging line segments to better approximate a great-circle path, at each step using computational geometry to check that the path wouldn’t go through a “too-high” region.

Finding the shortest flight path

Finding Kiwi Calls

You never know what people are going to come to Summer Camp with. A young man from New Zealand came to our camp with some overnight audio recordings from outside his house featuring occasional periods of (quite strange-sounding) squawking that were apparently the calls of one or more kiwi birds. What the young man wanted to do was automatic “kiwi voice recognition”, finding the calls, and perhaps distinguishing different birds.

I said I thought this wouldn’t be a particularly easy project, but he should try it anyway. Looking at what happened, it’s clear the project started out well. It was easy to pull out all intervals in his audio that weren’t just silence. But that broke up everything, including kiwi calls, into very small blocks. He solved that by the following interesting piece of code, that uses pattern matching to combine symbolic audio objects:

Wolfram Language code for kiwi code project

At this point it might just have worked to use unsupervised machine learning and FeatureSpacePlot to distinguish kiwi from non-kiwi sound clips. But machine learning is still quite a hit-or-miss business—and in this case it wasn’t a hit. So what did the student do? Well, he built himself a tiny lightweight user interface in a notebook, then started manually classifying sound clips. (Various instructors commented that it was fortunate he brought headphones…)

After classifying 200 clips, he used Classify to automatically classify all the other clips. He did a variety of transformations to the data—applying signal processing, generating a spectrogram, etc. And in the end he got his kiwi classifier to 82% accuracy: enough to make a reasonable first pass on finding kiwi calls—and going down a path to computational ornithology.

Biomechanics of Walking and Running

One young woman said she’d recently gotten a stress fracture in her foot that she was told was related to the force she was putting on it while running. She asked if she could make a computational model of what was going on. I have to say I was pessimistic about being able to do that in two weeks—and I suggested instead a project that I thought would be more manageable, involving studying possible gaits (walk, trot, etc.) for creatures with different numbers of legs. But I encouraged her to spend a little time seeing if she could do her original project—and I suggested that if she got to the stage of actually modeling bones, she could use our built-in anatomical data.

The next I knew it was a day before the end of the Summer Camp, and I was looking at what had happened with the projects… and I was really impressed! She’d found a paper with an appropriate model, understood it, and implemented it, and now she had an interactive demonstration of the force on a foot during walking or running. She’d even used the anatomical data to show a 3D image of what was happening.

She explained that when one walks there are two peaks in the force, but when one runs, there’s only one. And when I set her interactive demonstration for my own daily walking regimen I found out that (as she said was typical) I put a maximum force of about twice my weight on my foot when I walk.

Biomechanics of Walking and Running

Banana Ripeness Classifier

At first I couldn’t tell if he was really serious… but one young man insisted he wanted to use machine learning to tell when a piece of fruit is ripe. As it happens, I had used pretty much this exact example in a blog post some time ago discussing the use of machine learning in smart contracts. So I said, “sure, why don’t you try it”. I saw the student a few times during the Summer Camp, curiously always carrying a banana. And what I discovered at the end of the camp was that that very banana was a key element of his project.

At first he searched the web for images of bananas described as “underripe”, “overripe”, etc., then arranged them using FeatureSpacePlot:

Banana ripeness

Then he realized that he could get more quantitative by first looking at where in color space the pixels of the banana image lay. The result was that he was actually able to define a “banana ripeness scale”, where, as he described it: “A value of one was assigned to bananas that were on the brink of spoilage. A value of zero was assigned to a green banana fresh off a tree. A value of 0.5 was assigned to the ‘perfect’ banana.” It’s a nice example of how something everyday and qualitative can be made computational.

For his project, the student made a “Banana Classifier” app that he deployed through the Wolfram Cloud. And he even had an actual banana to test it on!

Banana classifier

Number Internationalization

One of my suggested projects was to implement “international or historical numeral systems”—the analogs of things like Roman numerals but for different cultures and times. One young woman fluent in Korean said she’d like to do this project, starting with Korean.

As it happens, our built-in IntegerName function converts to traditional Korean numerals. So she set herself the task of converting from Korean numerals. It’s an interesting algorithmic exercise, and she solved it with some nice, elegant code.

Korean to Hindu-Arabic numerals

By that point, she was on a roll… so she decided to go on to Burmese, and Thai. She tried to figure out Burmese from web sources… only to discover they were inconsistent… with the result that she ended up contacting a person who had an educational video about Burmese numerals, and eventually unscrambled the issue, wrote code to represent it, and then corrected the Wikipedia page about Burmese numerals. All in all, a great example of real-world algorithm curation. Oh, and she set up the conversions as a Wolfram Language microsite on the web.

Is That a Joke?

Can machine learning tell if something is funny? One young man at the Summer Camp wanted to find out. So for his project he used our Reddit API connection to pull jokes from the Jokes subreddit, and (presumably) non-jokes from the AskReddit subreddit. It took a bit of cleanup and data wrangling… but then he was able to feed his training data straight into the Classify function, and generated a classifier from which he then built a website.

It’s a little hard to know how well it works outside of “Reddit-style humor”—but his anecdotal study at the Summer Camp suggested about a 90% success rate.

Is this a joke machine

Making and Checking Checksums

Different projects involve different kinds of challenges. Sometimes the biggest challenge is just to define the project precisely enough. Other times it’s to get—or clean—the data that’s needed. Still other times, it’s to find a way to interpret voluminous output. And yet other times, it’s to see just how elegantly some particular idea can be implemented.

One math-oriented young woman at the camp picked “implementing checksum algorithms” from my list. Such algorithms (used for social security numbers, credit card numbers, etc.) are very nicely and precisely defined. But how simply and elegantly can they be implemented in the Wolfram Language? It’s a good computational thinking exercise—that requires really understanding both the algorithms and the language. And for me it’s nice to be able to immediately read off from the young woman’s code just how these checksum algorithms work…

Checksums algorithm

4D Plotting on a Tesseract

How should one plot a function in 4D? I had a project in my list about this, though I have to admit I hadn’t really figured out how it should be done. But, fortunately, a young man at the Summer Camp was keen to try to work on it. And with an interesting mixture of computational and mathematical thinking, he created ParametricPlot4D—then did a bunch of math to figure out how to render the results in what seemed like two useful ways: as an orthogonal projection, and as a stereographic projection. A Manipulate makes the results interactive—and they look pretty neat…

Plotting a tesseract

State-Level Mortality

In addition to my explicit list of project suggestions, I had a “meta suggestion”: take any dataset, for example from the new Wolfram Data Repository, and try to analyze and understand it. One student took a dataset about meteorite impacts; another about the recent Ebola outbreak in Africa. One young woman said she was interested in actuarial science—so I suggested that she look at something quintessentially actuarial: mortality data.

I suggested that maybe she could look at the (somewhat macabrely named) Death Master File. I wasn’t sure how far she’d get with it. But at the end of the camp I found out that she’d processed 90 million records—and successfully reduced them to derive aggregate survival curves for 25 different states and make an interactive Demonstration of the results. (Letting me conclude, for example, that my current probability of living to age 100 is 28% higher in Massachusetts than in Indiana…)

Survival image

“OCRing” Regular Tiling

Each year when I make up a list of projects for the Summer Camp I wonder if there’ll be particular favorites. My goal is actually to avoid this, and to have as uniform a distribution of interest in the projects as possible. But this year “Use Machine Learning to Identify Polyhedra” ended up being a minor favorite. And one consequence was that a student had already started working on the project even before we’d talked to him—even though by that time the project was already assigned to someone else.

But actually the “recovery” was better than the original. Because we figured out a really nice alternative project that was very well suited to the student. The project was to take images of regular tilings, say from a book, and to derive a computational representation of them, suitable, say, for LatticeData.

The student came up with a pretty sophisticated approach, largely based on image processing, but with a dash of computational geometry, combinatorics and even some cluster analysis thrown in. First, he used fairly elaborate image processing to identify the basic unit in the tiling. Then he figured out how this unit was arranged to form the final tiling. It ended up being about 102 lines of fairly dense algorithmic code—but the result was a quite robust “tiling OCR” system, that he also deployed on the web.

Analyzing regular tilings

Finding Buildings in Satellite Images

In my list I had a project “Identify buildings from satellite images”. A few students thought it sounded interesting, but as I thought about it some more, I got concerned that it might be really difficult. Still, one of our students was a capable young man who already seemed to know a certain amount about machine learning. So I encouraged him to give it a try. He ended up doing an impressive job.

He started by getting training data by comparing satellite images with street maps that marked buildings (and, conveniently, starting with the upcoming version of the Wolfram Language, not only streets maps but also satellite images are built in):

Buildings

Then he used NetChain to build a neural net (based on the classic LeNet network, but modified). And then he started trying to classify parts of images as “building” or “not building”.

FinalNetUpdate3[ ]

The results weren’t at all bad. But so far they were only answering the question “is there a building in that square?”, not “where is there a building?”. So then—in a nice piece of computational thinking—the student came up with a further idea: just have a window pan across the image, at each step estimating the probability of building vs. not-building. The result was a remarkably accurate heat map of where buildings might be.

TempMapDensity[ , FinalNetUpdate3, 50, 25]

It’d be a nice machine learning result for anyone. But as something done by a high-school student in two weeks I think it’s really impressive. And another great example of what’s now possible at an educational level with our whole Wolfram Language technology stack.

Beyond the Summer Camp

OK, so our Summer Camp was a success, and, with luck, the students from it are now successfully “launched” as independent computational thinkers. (The test, as far as I’m concerned, is whether when confronted with something in their education or their lives, they routinely turn to computational thinking, and just “write a program to solve the problem”. I’m hopeful that many of them now will. And, by the way, they immediately have “marketable skills”—like being able to do all sorts of data-science-related things.)

But how can we scale up what we’ve achieved with the Summer Camp? Well, we have a whole Computational Thinking Initiative that we’ve been designing to do just that. We’ll be rolling out different parts over the next little while, but one aspect will be doing other camps, and enabling other people to also do camps.

We’ve now got what amounts to an operations manual for how to “do a camp”. But suffice it to say that the core of it is to have instructors with good knowledge of the Wolfram Language (e.g. to the level of our Certified Instructor program), access to a bunch of great students, and use of a suitable venue. Two weeks seems to be a good length, though longer would work too. (Shorter will probably not be sufficient for students without prior experience to get to the point of doing a real project.)

Our camp is for high-school students (mainly aged 15 through 17). I think it would also be possible to do a successful camp for advanced middle-school students (maybe aged 12 and 13). And, of course, our long-running Summer School provides a very successful model for older students.

Beyond camps, we’ve had for some time a mentorships program which we will be streamlining and scaling up—helping students to work on longer-term projects. We’re also planning a variety of events and venues in which students can showcase their computational thinking work.

But for now it’s just exciting to see what was achieved in two weeks at this year’s Summer Camp. Yes, with the tech stack we now have, high-school students really can do serious computational thinking—that will make them not only immediately employable, but also positioned for what I think will be some of the most interesting career directions of the next few decades.

]]>
0
Stephen Wolfram <![CDATA[The Practical Business of Ontology: A Tale from the Front Lines]]> http://blog.internal.stephenwolfram.com/?p=13822 2017-07-20T19:23:09Z 2017-07-19T18:27:52Z ontoloty-thumbThe Philosophy of Chemicals “We’ve just got to decide: is a chemical like a city or like a number?” I spent my day yesterday—as I have for much of the past 30 years—designing new features of the Wolfram Language. And yesterday afternoon one of my meetings was a fast-paced discussion about how to extend the [...]]]> ontoloty-thumb

The Philosophy of Chemicals

“We’ve just got to decide: is a chemical like a city or like a number?” I spent my day yesterday—as I have for much of the past 30 years—designing new features of the Wolfram Language. And yesterday afternoon one of my meetings was a fast-paced discussion about how to extend the chemistry capabilities of the language.

At some level the problem we were discussing was quintessentially practical. But as so often turns out to be the case for things we do, it ultimately involves some deep intellectual issues. And to actually get the right answer—and to successfully design language features that will stand the test of time—we needed to plumb those depths, and talk about things that usually wouldn’t be considered outside of some kind of philosophy seminar.

Thinker

Part of the issue, of course, is that we’re dealing with things that haven’t really ever come up before. Traditional computer languages don’t try to talk directly about things like chemicals; they just deal with abstract data. But in the Wolfram Language we’re trying to build in as much knowledge about everything as possible, and that means we have to deal with actual things in the world, like chemicals.

We’ve built a whole system in the Wolfram Language for handling what we call entities. An entity could be a city (like New York City), or a movie, or a planet—or a zillion other things. An entity has some kind of name (“New York City”). And it has definite properties (like population, land area, founding date, …).

We’ve long had a notion of chemical entities—like water, or ethanol, or tungsten carbide. Each of these chemical entities has properties, like molecular mass, or structure graph, or boiling point.

And we’ve got many hundreds of thousands of chemicals where we know lots of properties. But all of these are in a sense concrete chemicals: specific compounds that we could put in a test tube and do things with.

But what we were trying to figure out yesterday is how to handle abstract chemicals—chemicals that we just abstractly construct, say by giving an abstract graph representing their chemical structures. Should these be represented by entities, like water or New York City? Or should they be considered more abstract, like lists of numbers, or, for that matter, mathematical graphs?

Well, of course, among the abstract chemicals we can construct are chemicals that we already represent by entities, like sucrose or aspirin or whatever. But here there’s an immediate distinction to make. Are we talking about individual molecules of sucrose or aspirin? Or about these things as bulk materials?

At some level it’s a confusing distinction. Because, we might think, once we know the molecular structure, we know everything—it’s just a matter of calculating it out. And some properties—like molar mass—are basically trivial to calculate from the molecular structure. But others—like melting point—are very far from trivial.

OK, but is this just a temporary problem that one shouldn’t base a long-term language design on? Or is it something more fundamental that will never change? Well, conveniently enough, I happen to have done a bunch of basic science that essentially answers this: and, yes, it’s something fundamental. It’s connected to what I call computational irreducibility. And for example, the precise value of, say, the melting point for an infinite amount of some material may actually be fundamentally uncomputable. (It’s related to the undecidability of the tiling problem; fitting in tiles is like seeing how molecules will arrange to make a solid.)

So by knowing this piece of (rather leading-edge) basic science, we know that we can meaningfully make a distinction between bulk versions of chemicals and individual molecules. Clearly there’s a close relation between, say, water molecules, and bulk water. But there’s still something fundamentally and irreducibly different about them, and about the properties we can compute for them.

At Least the Atoms Should Be OK

Alright, so let’s talk about individual molecules. Obviously they’re made of atoms. And it seems like at least when we talk about atoms, we’re on fairly solid ground. It might be reasonable to say that any given molecule always has some definite collection of atoms in it—though maybe we’ll want to consider “parametrized molecules” when we talk about polymers and the like.

But at least it seems safe to consider types of atoms as entities. After all, each type of atom corresponds to a chemical element, and there are only a limited number of those on the periodic table. Now of course in principle one can imagine additional “chemical elements”; one could even think of a neutron star as being like a giant atomic nucleus. But again, there’s a reasonable distinction to be made: almost certainly there are only a limited number of fundamentally stable types of atoms—and most of the others have ridiculously short lifetimes.

There’s an immediate footnote, however. A “chemical element” isn’t quite as definite a thing as one might imagine. Because it’s always a mixture of different isotopes. And, say, from one tungsten mine to another, that mixture might change, giving a different effective atomic mass.

And actually this is a good reason to represent types of atoms by entities. Because then one just has to have a single entity representing tungsten that one can use in talking about molecules. And only if one wants to get properties of that type of atom that depend on qualifiers like which mine it’s from does one have to deal with such things.

In a few cases (think heavy water, for example), one will need to explicitly talk about isotopes in what is essentially a chemical context. But most of the time, it’s going to be enough just to specify a chemical element.

To specify a chemical element you just have to give its atomic number Z. And then textbooks will tell you that to specify a particular isotope you just have to say how many neutrons it contains. But that ignores the unexpected case of tantalum. Because, you see, one of the naturally occurring forms of tantalum (180mTa) is actually an excited state of the tantalum nucleus, which happens to be very stable. And to properly specify this, you have to give its excitation level as well as its neutron count.

In a sense, though, quantum mechanics saves one here. Because while there are an infinite number of possible excited states of a nucleus, quantum mechanics says that all of them can be characterized just by two discrete values: spin and parity.

Every isotope—and every excited state—is different, and has its own particular properties. But the world of possible isotopes is much more orderly than, say, the world of possible animals. Because quantum mechanics says that everything in the world of isotopes can be characterized just by a limited set of discrete quantum numbers.

We’ve gone from molecules to atoms to nuclei, so why not talk about particles too? Well, it’s a bigger can of worms. Yes, there are the well-known particles like electrons and protons that are pretty easy to talk about—and are readily represented by entities in the Wolfram Language. But then there’s a zoo of other particles. Some of them—just like nuclei—are pretty easy to characterize. You can basically say things like: “it’s a particular excited state of a charm-quark-anti-charm-quark system” or some such. But in particle physics one’s dealing with quantum field theory, not just quantum mechanics. And one can’t just “count elementary particles”; one also has to deal with the possibility of virtual particles and so on. And in the end the question of what kinds of particles can exist is a very complicated one—rife with computational irreducibility. (For example, what stable states there can be of the gluon field is a much more elaborate version of something like the tiling problem I mentioned in connection with melting points.)

Maybe one day we’ll have a complete theory of fundamental physics. And maybe it’ll even be simple. But exciting as that will be, it’s not going to help much here. Because computational irreducibility means that there’s essentially an irreducible distance between what’s underneath, and what phenomena emerge.

And in creating a language to describe the world, we need to talk in terms of things that can actually be observed and computed about. We need to pay attention to the basic physics—not least so we can avoid setups that will lead to confusion later. But we also need to pay attention to the actual history of science, and actual things that have been measured. Yes, there are, for example, an infinite number of possible isotopes. But for an awful lot of purposes it’s perfectly useful just to set up entities for ones that are known.

The Space of Possible Chemicals

But is it the same in chemistry? In nuclear physics, we think we know all the reasonably stable isotopes that exist—so any additional and exotic ones will be very short-lived, and therefore probably not important in practical nuclear processes. But it’s a different story in chemistry. There are tens of millions of chemicals that people have studied (and, for example, put into papers or patents). And there’s really no limit on the molecules that one might want to consider, and that might be useful.

But, OK, so how can we refer to all these potential molecules? Well, in a first approximation we can specify their chemical structures, by giving graphs in which every node is an atom, and every edge is a bond.

What really is a “bond”? While it’s incredibly useful in practical chemistry, it’s at some level a mushy concept—some kind of semiclassical approximation to a full quantum mechanical story. There are some standard extra bits: double bonds, ionization states, etc. But in practice chemistry is very successfully done just by characterizing molecular structures by appropriately labeled graphs of atoms and bonds.

OK, but should chemicals be represented by entities, or by abstract graphs? Well, if it’s a chemical one’s already heard of, like carbon dioxide, an entity seems convenient. But what if it’s a new chemical that’s never been discussed before? Well, one could think about inventing a new entity to represent it.

Any self-respecting entity, though, better have a name. So what would the name be? Well, in the Wolfram Language, it could just be the graph that represents the structure. But maybe one wants something that seems more like an ordinary textual name—a string. Well, there’s always the IUPAC way of naming chemicals with names like 1,1′-{[3-(dimethylamino)propyl]imino}bis-2-propanol. Or there’s the more computer-friendly SMILES version: CC(CN(CCCN(C)C)CC(C)O)O. And whatever underlying graph one has, one can always generate one of these strings to represent it.

There’s an immediate problem, though: the string isn’t unique. In fact, however one chooses to write down the graph, it can’t always be unique. A particular chemical structure corresponds to a particular graph. But there can be many ways to draw the graph—and many different representations for it. And in fact even the (“graph isomorphism”) problem of determining whether two representations correspond to the same graph can be difficult to solve.

What Is a Chemical in the End?

OK, so let’s imagine we represent a chemical structure by a graph. At first, it’s an abstract thing. There are atoms as nodes in the graph, but we don’t know how they’d be arranged in an actual molecule (and e.g. how many angstroms apart they’d be). Of course, the answer isn’t completely well defined. Are we talking about the lowest-energy configuration of the molecule? (What if there are multiple configurations of the same energy?) Is the molecule supposed to be on its own, or in water, or whatever? How was the molecule supposed to have been made? (Maybe it’s a protein that folded a particular way when it came off the ribosome.)

Well, if we just had an entity representing, say, “naturally occurring hemoglobin”, maybe we’d be better off. Because in a sense that entity could encapsulate all these details.

But if we want to talk about chemicals that have never actually been synthesized it’s a bit of a different story. And it feels as if we’d be better off just with an abstract representation of any possible chemical.

But let’s talk about some other cases, and analogies. Maybe we should just treat everything as an entity. Like every integer could be an entity. Yes, there are an infinite number of them. But at least it’s clear what names they should be given. With real numbers, things are already messier. For example, there’s no longer the same kind of uniqueness as with integers: 0.99999… is really the same as 1.00000…, but it’s written differently.

What about sequences of integers, or, for that matter, mathematical formulas? Well, every possible sequence or every possible formula could conceivably be a different entity. But this wouldn’t be particularly useful, because much of what one wants to do with sequences or formulas is to go inside them, and transform their structure. But what’s convenient about entities is that they’re each just “single things” that one doesn’t have to “go inside”.

So what’s the story with “abstract chemicals”? It’s going to be a mixture. But certainly one’s going to want to “go inside” and transform the structure. Which argues for representing the chemical by a graph.

But then there’s potentially a nasty discontinuity. We’ve got the entity of carbon dioxide, which we already know lots of properties about. And then we’ve got this graph that abstractly represents the carbon dioxide molecule.

We might worry that this would be confusing both to humans and programs. But the first thing to realize is that we can distinguish what these two things are representing. The entity represents the bulk naturally occurring version of the chemical—whose properties have potentially been measured. The graph represents an abstract theoretical chemical, whose properties would have to be computed.

But obviously there’s got to be a bridge. Given a concrete chemical entity, one of the properties will be the graph that represents the structure of the molecule. And given a graph, one will need some kind of ChemicalIdentify function, that—a bit like GeoIdentify or maybe ImageIdentify—tries to identify from the graph what chemical entity (if any) has a molecular structure that corresponds to that graph.

Philosophy Meets Chemistry Meets Math Meets Physics…

As I write out some of the issues, I realize how complicated all this may seem. And, yes, it is complicated. But in our meeting yesterday, it all went very quickly. Of course it helps that everyone there had seen similar issues before: this is the kind of thing that’s all over the foundations of what we do. But each case is different.

And somehow this case got a bit deeper and more philosophical than usual. “Let’s talk about naming stars”, someone said. Obviously there are nearby stars that we have explicit names for. And some other stars may have been identified in large-scale sky surveys, and given identifiers of some kind. But there are lots of stars in distant galaxies that will never have been named. So how should we represent them?

That led to talking about cities. Yes, there are definite, chartered cities that have officially been assigned names–and we probably have essentially all of these right now in the Wolfram Language, updated regularly. But what about some village that’s created for a single season by some nomadic people? How should we represent it? Well, it has a certain location, at least for a while. But is it even a definite single thing, or might it, say, devolve into two villages, or not a village at all?

One can argue almost endlessly about identity—and even existence—for many of these things. But ultimately it’s not the philosophy of such things that we’re interested in: we’re trying to build software that people will find useful. And so what matters in the end is what’s going to be useful.

Now of course that’s not a precise thing to know. But it’s like for language design in general: think of everything people might want to do, then see how to set up primitives that will let people do those things. Does one want some chemicals represented by entities? Yes, that’s useful. Does one want a way to represent arbitrary chemical structures by graphs? Yes, that’s useful.

But to see what to actually do, one has to understand quite deeply what’s really being represented in each case, and how everything is related. And that’s where the philosophy has to meet the chemistry, and the math, and the physics, and so on.

I’m happy to say that by the end of our hour-long meeting yesterday (informed by about 40 years of relevant experience I’ve had, and collectively 100+ years from people in the meeting), I think we’d come up with the essence of a really nice way to handle chemicals and chemical structures. It’s going to be a while before it’s all fully worked out and implemented in the Wolfram Language. But the ideas are going to help inform the way we compute and reason about chemistry for many years to come. And for me, figuring out things like this is an extremely satisfying way to spend my time. And I’m just glad that in my long-running effort to advance the Wolfram Language I get to do so much of it.

]]>
6
Stephen Wolfram <![CDATA[Oh My Gosh, It’s Covered in Rule 30s!]]> http://blog.internal.stephenwolfram.com/?p=13754 2017-06-22T21:59:20Z 2017-06-01T06:22:57Z blog-cambridge-station-thumbA British Train Station A week ago a new train station, named “Cambridge North”, opened in Cambridge, UK. Normally such an event would be far outside my sphere of awareness. (I think I last took a train to Cambridge in 1975.) But last week people started sending me pictures of the new train station, wondering if I [...]]]> blog-cambridge-station-thumb

A British Train Station

A week ago a new train station, named “Cambridge North”, opened in Cambridge, UK. Normally such an event would be far outside my sphere of awareness. (I think I last took a train to Cambridge in 1975.) But last week people started sending me pictures of the new train station, wondering if I could identify the pattern on it:

Cambridge North train station

And, yes, it does indeed look a lot like patterns I’ve spent years studying—that come from simple programs in the computational universe. My first—and still favorite—examples of simple programs are one-dimensional cellular automata like this:

One-dimensional cellular automata

The system evolves line by line from the top, determining the color of each cell according to the rule underneath. This particular cellular automata I called “rule 182”, because the bit pattern in the rule corresponds to the number 182 in binary. There are altogether 256 possible cellular automata like this, and this is what all of them do:

256 possible cellular automata

Many of them show fairly simple behavior. But the huge surprise I got when I first ran all these cellular automata in the early 1980s is that even though all the rules are very simple to state, some of them generate very complex behavior. The first in the list that does that—and still my favorite example—is rule 30:

Rule 30

If one runs it for 400 steps one gets this:

After 400 steps

And, yes, it’s remarkable that starting from one black cell at the top, and just repeatedly following a simple rule, it’s possible to get all this complexity. I think it’s actually an example of a hugely important phenomenon, that’s central to how complexity gets made in nature, as well as to how we can get a new level of technology. And in fact, I think it’s important enough that I spent more than a decade writing a 1200-page book (that just celebrated its 15th anniversary) based on it.

And for years I’ve actually had rule 30 on my business cards:

Business cards

But back to the Cambridge North train station. Its pattern is obviously not completely random. But if it was made by a rule, what kind of rule? Could it be a cellular automaton?

I zoomed in on a photograph of the pattern:

Enlarged pattern

Suddenly, something seemed awfully familiar: the triangles, the stripes, the L shapes. Wait a minute… it couldn’t actually be my favorite rule of all time, rule 30?

Clearly the pattern is tipped 45° from how I’d usually display a cellular automaton. And there are black triangles in the photograph, not white ones like in rule 30. But if one black-white inverts the rule (so it’s now rule 135), one gets this:

Black-white inversion of the pattern

And, yes, it’s the same kind of pattern as in the photograph! But if it’s rule 30 (or rule 135) what’s its initial condition? Rule 30 can actually be used as a cryptosystem—because it can be hard (maybe even NP complete) to reconstruct its initial condition.

But, OK, if it’s my favorite rule, I wondered if maybe it’s also my favorite initial condition—a single black cell. And, yes, it is! The train station pattern comes exactly from the (inverted) right-hand edge of my favorite rule 30 pattern!

Edge of rule 30

Here’s the Wolfram Language code. First run the cellular automaton, then rotate the pattern:

Rotate[ArrayPlot[CellularAutomaton[135, {{1},0},40],Mesh->True],-45 Degree]

It’s a little trickier to pull out precisely the section of the pattern that’s used. Here’s the code (the PlotRange is what determines the part of the pattern that’s shown):

Graphics[Rotate[First[ArrayPlot[CellularAutomaton[135, {{1},0},80],Mesh->True]],-45 Degree], PlotRange->{{83,104},{-12,60}}]

OK, so where is this pattern actually used at the train station? Everywhere!

Cambridge North collage

It’s made of perforated aluminum. You can actually look through it, reminiscent of an old latticed window. From inside, the pattern is left-right reversed—so if it’s rule 135 from outside, it’s rule 149 from inside. And at night, the pattern is black-white inverted, because there’s light coming from inside—so from the outside it’s “rule 135 by day, and rule 30 at night”.

What are some facts about the rule 30 pattern? It’s extremely hard to rigorously prove things about it (and that’s interesting in itself—and closely related to the fundamental phenomenon of computational irreducibility). But, for example—like, say, the digits of π—many aspects of it seem random. And, for instance, black and white squares appear to occur with equal frequency—meaning that at the train station the panels let in about 50% of the outside light.

If one looks at sequences of n cells, it seems that all 2n configurations will occur on average with equal frequency. But not everything is random. And so, for example, if one looks at 3×2 blocks of cells, only 24 of the 32 possible ones ever occur. (Maybe some people waiting for trains will figure out which blocks are missing…)

When we look at the pattern, our visual system particularly picks out the black triangles. And, yes, it seems as if triangles of any size can ultimately occur, albeit with frequency decreasing exponentially with size.

If one looks carefully at the right-hand edge of the rule 30 pattern, one can see that it repeats. However, the repetition period seems to increase exponentially as one goes in from the edge.

At the train station, there are lots of identical panels. But rule 30 is actually an inexhaustible source of new patterns. So what would happen if one just continued the evolution, and rendered it on successive panels? Here’s the result. It’s a pity about the hint of periodicity on the right-hand edge, and the big triangle on panel 5 (which might be a safety problem at the train station).

Successive panels

Fifteen more steps in from the edge, there’s no hint of that anymore:

Fifteen more steps

What about other initial conditions? If the initial conditions repeat, then so will the pattern. But otherwise, so far as one can tell, the pattern will look essentially the same as with a single-cell initial condition.

One can try other rules too. Here are a few from the same simplest 256-rule set as rule 30:

Simple 256-rule set

Moving deeper from the edge the results look a little different (for aficionados, rule 89 is a transformed version of rule 45, rule 182 of rule 90, and rule 193 of rule 110):

Moving deeper from the edge

And starting from random initial conditions, rather than a single black cell, things again look different:

Starting from random initial conditions

And here are a few more rules, started from random initial conditions:

A few more rules

Here’s a website (made in a couple of minutes with a tiny piece of Wolfram Language code) that lets you experiment (including with larger rule numbers, based on longer-range rules). (And if you want to explore more systematically, here’s a Wolfram Notebook to try.)

Cellular automaton panel explorer

It’s amazing what’s out there in the computational universe of possible programs. There’s an infinite range of possible patterns. But it’s cool that the Cambridge North train station uses my all-time favorite discovery in the computational universe—rule 30! And it looks great!

The Bigger Picture

There’s something curiously timeless about algorithmically generated forms. A dodecahedron from ancient Egypt still looks crisp and modern today. As do periodic tilings—or nested forms—even from centuries ago:

Periodic tilings and nested forms

But can one generate richer forms algorithmically? Before I discovered rule 30, I’d always assumed that any form generated from simple rules would always somehow end up being obviously simple. But rule 30 was a big shock to my intuition—and from it I realized that actually in the computational universe of all possible rules, it’s actually very easy to get rich and complex behavior, even from simple underlying rules.

And what’s more, the patterns that are generated often have remarkable visual interest. Here are a few produced by cellular automata (now with 3 possible colors for each cell, rather than 2):

Three-color cellular automata

There’s an amazing diversity of forms. And, yes, they’re often complicated. But because they’re based on simple underlying rules, they always have a certain logic to them: in a sense each of them tells a definite “algorithmic story”.

One thing that’s notable about forms we see in the computational universe is that they often look a lot like forms we see in nature. And I don’t think that’s a coincidence. Instead, I think what’s going on is that rules in the computational universe capture the essence of laws that govern lots of systems in nature—whether in physics, biology or wherever. And maybe there’s a certain familiarity or comfort associated with forms in the computational universe that comes from their similarity to forms we’re used to in nature.

But is what we get from the computational universe art? When we pick out something like rule 30 for a particular purpose, what we’re doing is conceptually a bit like photography: we’re not creating the underlying forms, but we are selecting the ones we choose to use.

In the computational universe, though, we can be more systematic. Given some aesthetic criterion, we can automatically search through perhaps even millions or billions of possible rules to find optimal ones: in a sense automatically “discovering art” in the computational universe.

We did an experiment on this for music back in 2007: WolframTones. And what’s remarkable is that even by sampling fairly small numbers of rules (cellular automata, as it happens), we’re able to produce all sorts of interesting short pieces of music—that often seem remarkably “creative” and “inventive”.

From a practical point of view, automatic discovery in the computational universe is important because it allows for mass customization. It makes it easy to be “original” (and “creative”)—and to find something different every time, or to fit constraints that have never been seen before (say, a pattern in a complicated geometric region).

The Cambridge North train station uses a particular rule from the computational universe to make what amounts to an ornamental pattern. But one can also use rules from the computational universe for other things in architecture. And one can even imagine a building in which everything—from overall massing down to details of moldings—is completely determined by something close to a single rule.

One might assume that such a building would somehow be minimalist and sterile. But the remarkable fact is that this doesn’t have to be true—and that instead there are plenty of rich, almost “organic” forms to be “mined” from the computational universe.

Ever since I started writing about one-dimensional cellular automata back in the early 1980s, there’s been all sorts of interesting art done with them. Lots of different rules have been used. Sometimes they’ve been what I called “class 4” rules that have a particularly organic look. But often it’s been other rules—and rule 30 has certainly made its share of appearances—whether it’s on floors, shirts, tea cosies, kinetic installations, or, recently, mass-customized scarves (with the knitting machine actually running the cellular automaton):

CA art

But today we’re celebrating a new and different manifestation of rule 30. Formed from permanent aluminum panels, in an ancient university town, a marvellous corner of the computational universe adorns one of the most practical of structures: a small train station. My compliments to the architects. May what they’ve made give generations of rail travelers a little glimpse of the wonders of the computational universe. And maybe perhaps a few, echoing the last words attributed to the traveler in the movie 2001: A Space Odyssey, exclaim “oh my gosh, it’s covered in rule 30s!”


(Thanks to Wolfram Summer School alum Alyssa Adams for sending us the photos of Cambridge North.)

This post has been updated to include an image of Cambridge North at night, courtesy of Quintin Doyle, Senior Architectural Designer, Atkins.

]]>
7
Stephen Wolfram <![CDATA[A New Kind of Science: A 15-Year View]]> http://blog.internal.stephenwolfram.com/?p=13536 2017-07-11T15:41:50Z 2017-05-16T13:43:21Z 15th-thumbStarting now, in celebration of its 15th anniversary, A New Kind of Science will be freely available in its entirety, with high-resolution images, on the web or for download. It’s now 15 years since I published my book A New Kind of Science—more than 25 since I started writing it, and more than 35 since [...]]]> 15th-thumb

Starting now, in celebration of its 15th anniversary, A New Kind of Science will be freely available in its entirety, with high-resolution images, on the web or for download.

A New Kind of Science

It’s now 15 years since I published my book A New Kind of Science—more than 25 since I started writing it, and more than 35 since I started working towards it. But with every passing year I feel I understand more about what the book is really about—and why it’s important. I wrote the book, as its title suggests, to contribute to the progress of science. But as the years have gone by, I’ve realized that the core of what’s in the book actually goes far beyond science—into many areas that will be increasingly important in defining our whole future.

So, viewed from a distance of 15 years, what is the book really about? At its core, it’s about something profoundly abstract: the theory of all possible theories, or the universe of all possible universes. But for me one of the achievements of the book is the realization that one can explore such fundamental things concretely—by doing actual experiments in the computational universe of possible programs. And in the end the book is full of what might at first seem like quite alien pictures made just by running very simple such programs.

Back in 1980, when I made my living as a theoretical physicist, if you’d asked me what I thought simple programs would do, I expect I would have said “not much”. I had been very interested in the kind of complexity one sees in nature, but I thought—like a typical reductionistic scientist—that the key to understanding it must lie in figuring out detailed features of the underlying component parts.

In retrospect I consider it incredibly lucky that all those years ago I happened to have the right interests and the right skills to actually try what is in a sense the most basic experiment in the computational universe: to systematically take a sequence of the simplest possible programs, and run them.

I could tell as soon as I did this that there were interesting things going on, but it took a couple more years before I began to really appreciate the force of what I’d seen. For me it all started with one picture:

Rule 30

Or, in modern form:

Rule 30, modern form

I call it rule 30. It’s my all-time favorite discovery, and today I carry it around everywhere on my business cards. What is it? It’s one of the simplest programs one can imagine. It operates on rows of black and white cells, starting from a single black cell, and then repeatedly applies the rules at the bottom. And the crucial point is that even though those rules are by any measure extremely simple, the pattern that emerges is not.

It’s a crucial—and utterly unexpected—feature of the computational universe: that even among the very simplest programs, it’s easy to get immensely complex behavior. It took me a solid decade to understand just how broad this phenomenon is. It doesn’t just happen in programs (“cellular automata”) like rule 30. It basically shows up whenever you start enumerating possible rules or possible programs whose behavior isn’t obviously trivial.

Similar phenomena had actually been seen for centuries in things like the digits of pi and the distribution of primes—but they were basically just viewed as curiosities, and not as signs of something profoundly important. It’s been nearly 35 years since I first saw what happens in rule 30, and with every passing year I feel I come to understand more clearly and deeply what its significance is.

Four centuries ago it was the discovery of the moons of Jupiter and their regularities that sowed the seeds for modern exact science, and for the modern scientific approach to thinking. Could my little rule 30 now be the seed for another such intellectual revolution, and a new way of thinking about everything?

In some ways I might personally prefer not to take responsibility for shepherding such ideas (“paradigm shifts” are hard and thankless work). And certainly for years I have just quietly used such ideas to develop technology and my own thinking. But as computation and AI become increasingly central to our world, I think it’s important that the implications of what’s out there in the computational universe be more widely understood.

Implications of the Computational Universe

Here’s the way I see it today. From observing the moons of Jupiter we came away with the idea that—if looked at right—the universe is an ordered and regular place, that we can ultimately understand. But now, in exploring the computational universe, we quickly come upon things like rule 30 where even the simplest rules seem to lead to irreducibly complex behavior.

One of the big ideas of A New Kind of Science is what I call the Principle of Computational Equivalence. The first step is to think of every process—whether it’s happening with black and white squares, or in physics, or inside our brains—as a computation that somehow transforms input to output. What the Principle of Computational Equivalence says is that above an extremely low threshold, all processes correspond to computations of equivalent sophistication.

It might not be true. It might be that something like rule 30 corresponds to a fundamentally simpler computation than the fluid dynamics of a hurricane, or the processes in my brain as I write this. But what the Principle of Computational Equivalence says is that in fact all these things are computationally equivalent.

It’s a very important statement, with many deep implications. For one thing, it implies what I call computational irreducibility. If something like rule 30 is doing a computation just as sophisticated as our brains or our mathematics, then there’s no way we can “outrun” it: to figure out what it will do, we have to do an irreducible amount of computation, effectively tracing each of its steps.

The mathematical tradition in exact science has emphasized the idea of predicting the behavior of systems by doing things like solving mathematical equations. But what computational irreducibility implies is that out in the computational universe that often won’t work, and instead the only way forward is just to explicitly run a computation to simulate the behavior of the system.

A Shift in Looking at the World

One of the things I did in A New Kind of Science was to show how simple programs can serve as models for the essential features of all sorts of physical, biological and other systems. Back when the book appeared, some people were skeptical about this. And indeed at that time there was a 300-year unbroken tradition that serious models in science should be based on mathematical equations.

But in the past 15 years something remarkable has happened. For now, when new models are created—whether of animal patterns or web browsing behavior—they are overwhelmingly more often based on programs than on mathematical equations.

Year by year, it’s been a slow, almost silent, process. But by this point, it’s a dramatic shift. Three centuries ago pure philosophical reasoning was supplanted by mathematical equations. Now in these few short years, equations have been largely supplanted by programs. For now, it’s mostly been something practical and pragmatic: the models work better, and are more useful.

But when it comes to understanding the foundations of what’s going on, one’s led not to things like mathematical theorems and calculus, but instead to ideas like the Principle of Computational Equivalence. Traditional mathematics-based ways of thinking have made concepts like force and momentum ubiquitous in the way we talk about the world. But now as we think in fundamentally computational terms we have to start talking in terms of concepts like undecidability and computational irreducibility.

Will some type of tumor always stop growing in some particular model? It might be undecidable. Is there a way to work out how a weather system will develop? It might be computationally irreducible.

These concepts are pretty important when it comes to understanding not only what can and cannot be modeled, but also what can and cannot be controlled in the world. Computational irreducibility in economics is going to limit what can be globally controlled. Computational irreducibility in biology is going to limit how generally effective therapies can be—and make highly personalized medicine a fundamental necessity.

And through ideas like the Principle of Computational Equivalence we can start to discuss just what it is that allows nature—seemingly so effortlessly—to generate so much that seems so complex to us. Or how even deterministic underlying rules can lead to computationally irreducible behavior that for all practical purposes can seem to show “free will”.

Cellular automata

Mining the Computational Universe

A central lesson of A New Kind of Science is that there’s a lot of incredible richness out there in the computational universe. And one reason that’s important is that it means that there’s a lot of incredible stuff out there for us to “mine” and harness for our purposes.

Want to automatically make an interesting custom piece of art? Just start looking at simple programs and automatically pick out one you like—as in our WolframTones music site from a decade ago. Want to find an optimal algorithm for something? Just search enough programs out there, and you’ll find one.

We’ve normally been used to creating things by building them up, step by step, with human effort—progressively creating architectural plans, or engineering drawings, or lines of code. But the discovery that there’s so much richness so easily accessible in the computational universe suggests a different approach: don’t try building anything; just define what you want, and then search for it in the computational universe.

Sometimes it’s really easy to find. Like let’s say you want to generate apparent randomness. Well, then just enumerate cellular automata (as I did in 1984), and very quickly you come upon rule 30—which turns out to be one of the very best known generators of apparent randomness (look down the center column of cell values, for examples). In other situations you might have to search 100,000 cases (as I did in finding the simplest axiom system for logic, or the simplest universal Turing machine), or you might have to search millions or even trillions of cases. But in the past 25 years, we’ve had incredible success in just discovering algorithms out there in the computational universe—and we rely on many of them in implementing the Wolfram Language.

At some level it’s quite sobering. One finds some tiny program out in the computational universe. One can tell it does what one wants. But when one looks at what it’s doing, one doesn’t have any real idea how it works. Maybe one can analyze some part—and be struck by how “clever” it is. But there just isn’t a way for us to understand the whole thing; it’s not something familiar from our usual patterns of thinking.

Of course, we’ve often had similar experiences before—when we use things from nature. We may notice that some particular substance is a useful drug or a great chemical catalyst, but we may have no idea why. But in doing engineering and in most of our modern efforts to build technology, the great emphasis has instead been on constructing things whose design and operation we can readily understand.

In the past we might have thought that was enough. But what our explorations of the computational universe show is that it’s not: selecting only things whose operation we can readily understand misses most of the immense power and richness that’s out there in the computational universe.

A World of Discovered Technology

What will the world look like when more of what we have is mined from the computational universe? Today the environment we build for ourselves is dominated by things like simple shapes and repetitive processes. But the more we use what’s out there in the computational universe, the less regular things will look. Sometimes they may look a bit “organic”, or like what we see in nature (since after all, nature follows similar kinds of rules). But sometimes they may look quite random, until perhaps suddenly and incomprehensibly they achieve something we recognize.

For several millennia we as a civilization have been on a path to understand more about what happens in our world—whether by using science to decode nature, or by creating our own environment through technology. But to use more of the richness of the computational universe we must at least to some extent forsake this path.

In the past, we somehow counted on the idea that between our brains and the tools we could create we would always have fundamentally greater computational power than the things around us—and as a result we would always be able to “understand” them. But what the Principle of Computational Equivalence says is that this isn’t true: out in the computational universe there are lots of things just as powerful as our brains or the tools we build. And as soon as we start using those things, we lose the “edge” we thought we had.

Today we still imagine we can identify discrete “bugs” in programs. But most of what’s powerful out there in the computational universe is rife with computational irreducibility—so the only real way to see what it does is just to run it and watch what happens.

We ourselves, as biological systems, are a great example of computation happening at a molecular scale—and we are no doubt rife with computational irreducibility (which is, at some fundamental level, why medicine is hard). I suppose it’s a tradeoff: we could limit our technology to consist only of things whose operation we understand. But then we would miss all that richness that’s out there in the computational universe. And we wouldn’t even be able to match the achievements of our own biology in the technology we create.

Machine Learning and the Neural Net Renaissance

There’s a common pattern I’ve noticed with intellectual fields. They go for decades and perhaps centuries with only incremental growth, and then suddenly, usually as a result of a methodological advance, there’s a burst of “hypergrowth” for perhaps 5 years, in which important new results arrive almost every week.

I was fortunate enough that my own very first field—particle physics—was in its period of hypergrowth right when I was involved in the late 1970s. And for myself, the 1990s felt like a kind of personal period of hypergrowth for what became A New Kind of Science—and indeed that’s why I couldn’t pull myself away from it for more than a decade.

But today, the obvious field in hypergrowth is machine learning, or, more specifically, neural nets. It’s funny for me to see this. I actually worked on neural nets back in 1981, before I started on cellular automata, and several years before I found rule 30. But I never managed to get neural nets to do anything very interesting—and actually I found them too messy and complicated for the fundamental questions I was concerned with.

And so I “simplified them”—and wound up with cellular automata. (I was also inspired by things like the Ising model in statistical physics, etc.) At the outset, I thought I might have simplified too far, and that my little cellular automata would never do anything interesting. But then I found things like rule 30. And I’ve been trying to understand its implications ever since.

In building Mathematica and the Wolfram Language, I’d always kept track of neural nets, and occasionally we’d use them in some small way for some algorithm or another. But about 5 years ago I suddenly started hearing amazing things: that somehow the idea of training neural nets to do sophisticated things was actually working. At first I wasn’t sure. But then we started building neural net capabilities in the Wolfram Language, and finally two years ago we released our ImageIdentify.com website—and now we’ve got our whole symbolic neural net system. And, yes, I’m impressed. There are lots of tasks that had traditionally been viewed as the unique domain of humans, but which now we can routinely do by computer.

But what’s actually going on in a neural net? It’s not really to do with the brain; that was just the inspiration (though in reality the brain probably works more or less the same way). A neural net is really a sequence of functions that operate on arrays of numbers, with each function typically taking quite a few inputs from around the array. It’s not so different from a cellular automaton. Except that in a cellular automaton, one’s usually dealing with, say, just 0s and 1s, not arbitrary numbers like 0.735. And instead of taking inputs from all over the place, in a cellular automaton each step takes inputs only from a very well-defined local region.

Now, to be fair, it’s pretty common to study “convolutional neural nets”, in which the patterns of inputs are very regular, just like in a cellular automaton. And it’s becoming clear that having precise (say 32-bit) numbers isn’t critical to the operation of neural nets; one can probably make do with just a few bits.

But a big feature of neural nets is that we know how to make them “learn”. In particular, they have enough features from traditional mathematics (like involving continuous numbers) that techniques like calculus can be applied to provide strategies to make them incrementally change their parameters to “fit their behavior” to whatever training examples they’re given.

It’s far from obvious how much computational effort, or how many training examples, will be needed. But the breakthrough of about five years ago was the discovery that for many important practical problems, what’s available with modern GPUs and modern web-collected training sets can be enough.

Pretty much nobody ends up explicitly setting or “engineering” the parameters in a neural net. Instead, what happens is that they’re found automatically. But unlike with simple programs like cellular automata, where one’s typically enumerating all possibilities, in current neural nets there’s an incremental process, essentially based on calculus, that manages to progressively improve the net—a little like the way biological evolution progressively improves the “fitness” of an organism.

It’s plenty remarkable what comes out from training a neural net in this way, and it’s plenty difficult to understand how the neural net does what it does. But in some sense the neural net isn’t venturing too far across the computational universe: it’s always basically keeping the same basic computational structure, and just changing its behavior by changing parameters.

But to me the success of today’s neural nets is a spectacular endorsement of the power of the computational universe, and another validation of the ideas of A New Kind of Science. Because it shows that out in the computational universe, away from the constraints of explicitly building systems whose detailed behavior one can foresee, there are immediately all sorts of rich and useful things to be found.

NKS Meets Modern Machine Learning

Is there a way to bring the full power of the computational universe—and the ideas of A New Kind of Science—to the kinds of things one does with neural nets? I suspect so. And in fact, as the details become clear, I wouldn’t be surprised if exploration of the computational universe saw its own period of hypergrowth: a “mining boom” of perhaps unprecedented proportions.

In current work on neural nets, there’s a definite tradeoff one sees. The more what’s going on inside the neural net is like a simple mathematical function with essentially arithmetic parameters, the easier it is to use ideas from calculus to train the network. But the more what’s going is like a discrete program, or like a computation whose whole structure can change, the more difficult it is to train the network.

It’s worth remembering, though, that the networks we’re routinely training now would have looked utterly impractical to train only a few years ago. It’s effectively just all those quadrillions of GPU operations that we can throw at the problem that makes training feasible. And I won’t be surprised if even quite pedestrian (say, local exhaustive search) techniques will fairly soon let one do significant training even in cases where no incremental numerical approach is possible. And perhaps even it will be possible to invent some major generalization of things like calculus that will operate in the full computational universe. (I have some suspicions, based on thinking about generalizing basic notions of geometry to cover things like cellular automaton rule spaces.)

What would this let one do? Likely it would let one find considerably simpler systems that could achieve particular computational goals. And maybe that would bring within reach some qualitatively new level of operations, perhaps beyond what we’re used to being possible with things like brains.

There’s a funny thing that’s going on with modeling these days. As neural nets become more successful, one begins to wonder: why bother to simulate what’s going on inside a system when one can just make a black-box model of its output using a neural net? Well, if we manage to get machine learning to reach deeper into the computational universe, we won’t have as much of this tradeoff any more—because we’ll be able to learn models of the mechanism as well as the output.

I’m pretty sure that bringing the full computational universe into the purview of machine learning will have spectacular consequences. But it’s worth realizing that computational universality—and the Principle of Computational Equivalence—make it less a matter of principle. Because they imply that even neural nets of the kinds we have now are universal, and are capable of emulating anything any other system can do. (In fact, this universality result was essentially what launched the whole modern idea of neural nets, back in 1943.)

And as a practical matter, the fact that current neural net primitives are being built into hardware and so on will make them a desirable foundation for actual technology systems, though, even if they’re far from optimal. But my guess is that there are tasks where for the foreseeable future access to the full computational universe will be necessary to make them even vaguely practical.

Finding AI

What will it take to make artificial intelligence? As a kid, I was very interested in figuring out how to make a computer know things, and be able to answer questions from what it knew. And when I studied neural nets in 1981, it was partly in the context of trying to understand how to build such a system. As it happens, I had just developed SMP, which was a forerunner of Mathematica (and ultimately the Wolfram Language)—and which was very much based on symbolic pattern matching (“if you see this, transform it to that”). At the time, though, I imagined that artificial intelligence was somehow a “higher level of computation”, and I didn’t know how to achieve it.

I returned to the problem every so often, and kept putting it off. But then when I was working on A New Kind of Science it struck me: if I’m to take the Principle of Computational Equivalence seriously, then there can’t be any fundamentally “higher level of computation”—so AI must be achievable just with the standard ideas of computation that I already know.

And it was this realization that got me started building Wolfram|Alpha. And, yes, what I found is that lots of those very “AI-oriented things”, like natural language understanding, could be done just with “ordinary computation”, without any magic new AI invention. Now, to be fair, part of what was happening was that we were using ideas and methods from A New Kind of Science: we weren’t just engineering everything; we were often searching the computational universe for rules and algorithms to use.

So what about “general AI”? Well, I think at this point that with the tools and understanding we have, we’re in a good position to automate essentially anything we can define. But definition is a more difficult and central issue than we might imagine.

The way I see things at this point is that there’s a lot of computation even near at hand in the computational universe. And it’s powerful computation. As powerful as anything that happens in our brains. But we don’t recognize it as “intelligence” unless it’s aligned with our human goals and purposes.

Ever since I was writing A New Kind of Science, I’ve been fond of quoting the aphorism “the weather has a mind of its own”. It sounds so animistic and pre-scientific. But what the Principle of Computational Equivalence says is that actually, according to the most modern science, it’s true: the fluid dynamics of the weather is the same in its computational sophistication as the electrical processes that go on in our brains.

But is it “intelligent”? When I talk to people about A New Kind of Science, and about AI, I’ll often get asked when I think we’ll achieve “consciousness” in a machine. Life, intelligence, consciousness: they are all concepts that we have a specific example of, here on Earth. But what are they in general? All life on Earth shares RNA and the structure of cell membranes. But surely that’s just because all life we know is part of one connected thread of history; it’s not that such details are fundamental to the very concept of life.

And so it is with intelligence. We have only one example we’re sure of: us humans. (We’re not even sure about animals.) But human intelligence as we experience it is deeply entangled with human civilization, human culture and ultimately also human physiology—even though none of these details are presumably relevant in the abstract definition of intelligence.

We might think about extraterrestrial intelligence. But what the Principle of Computational Equivalence implies is that actually there’s “alien intelligence” all around us. But somehow it’s just not quite aligned with human intelligence. We might look at rule 30, for example, and be able to see that it’s doing sophisticated computation, just like our brains. But somehow it just doesn’t seem to have any “point” to what it’s doing.

We imagine that in doing the things we humans do, we operate with certain goals or purposes. But rule 30, for example, just seems to be doing what it’s doing—just following some definite rule. In the end, though, one realizes we’re not so very different. After all, there are definite laws of nature that govern our brains. So anything we do is at some level just playing out those laws.

Any process can actually be described either in terms of mechanism (“the stone is moving according to Newton’s laws”), or in terms of goals (“the stone is moving so as to minimize potential energy”). The description in terms of mechanism is usually what’s most useful in connecting with science. But the description in terms of goals is usually what’s most useful in connecting with human intelligence.

And this is crucial in thinking about AI. We know we can have computational systems whose operations are as sophisticated as anything. But can we get them to do things that are aligned with human goals and purposes?

In a sense this is what I now view as the key problem of AI: it’s not about achieving underlying computational sophistication, but instead it’s about communicating what we want from this computation.

The Importance of Language

I’ve spent much of my life as a computer language designer—most importantly creating what is now the Wolfram Language. I’d always seen my role as a language designer being to imagine the possible computations people might want to do, then—like a reductionist scientist—trying to “drill down” to find good primitives from which all these computations could be built up. But somehow from A New Kind of Science, and from thinking about AI, I’ve come to think about it a little differently.

Now what I more see myself as doing is making a bridge between our patterns of human thinking, and what the computational universe is capable of. There are all sorts of amazing things that can in principle be done by computation. But what the language does is to provide a way for us humans to express what we want done, or want to achieve—and then to get this actually executed, as automatically as possible.

Language design has to start from what we know and are familiar with. In the Wolfram Language, we name the built-in primitives with English words, leveraging the meanings that those words have acquired. But the Wolfram Language is not like natural language. It’s something more structured, and more powerful. It’s based on the words and concepts that we’re familiar with through the shared corpus of human knowledge. But it gives us a way to build up arbitrarily sophisticated programs that in effect express arbitrarily complex goals.

Yes, the computational universe is capable of remarkable things. But they’re not necessarily things that we humans can describe or relate to. But in building the Wolfram Language my goal is to do the best I can in capturing everything we humans want—and being able to express it in executable computational terms.

When we look at the computational universe, it’s hard not to be struck by the limitations of what we know how to describe or think about. Modern neural nets provide an interesting example. For the ImageIdentify function of the Wolfram Language we’ve trained a neural net to identify thousands of kinds of things in the world. And to cater to our human purposes, what the network ultimately does is to describe what it sees in terms of concepts that we can name with words—tables, chairs, elephants, etc.

But internally what the network is doing is to identify a series of features of any object in the world. Is it green? Is it round? And so on. And what happens as the neural network is trained is that it identifies features it finds useful for distinguishing different kinds of things in the world. But the point is that almost none of these features are ones to which we happen to have assigned words in human language.

Out in the computational universe it’s possible to find what may be incredibly useful ways to describe things. But they’re alien to us humans. They’re not something we know how to express, based on the corpus of knowledge our civilization has developed.

Now of course new concepts are being added to the corpus of human knowledge all the time. Back a century ago, if someone saw a nested pattern they wouldn’t have any way to describe it. But now we’d just say “it’s a fractal”. But the problem is that in the computational universe there’s an infinite collection of “potentially useful concepts”—with which we can never hope to ultimately keep up.

The Analogy in Mathematics

When I wrote A New Kind of Science I viewed it in no small part as an effort to break away from the use of mathematics—at least as a foundation for science. But one of the things I realized is that the ideas in the book also have a lot of implications for pure mathematics itself.

What is mathematics? Well, it’s a study of certain abstract kinds of systems, based on things like numbers and geometry. In a sense it’s exploring a small corner of the computational universe of all possible abstract systems. But still, plenty has been done in mathematics: indeed, the 3 million or so published theorems of mathematics represent perhaps the largest single coherent intellectual structure that our species has built.

Ever since Euclid, people have at least notionally imagined that mathematics starts from certain axioms (say, a+b=b+a, a+0=a, etc.), then builds up derivations of theorems. Why is math hard? The answer is fundamentally rooted in the phenomenon of computational irreducibility—which here is manifest in the fact that there’s no general way to shortcut the series of steps needed to derive a theorem. In other words, it can be arbitrarily hard to get a result in mathematics. But worse than that—as Gödel’s Theorem showed—there can be mathematical statements where there just aren’t any finite ways to prove or disprove them from the axioms. And in such cases, the statements just have to be considered “undecidable”.

And in a sense what’s remarkable about math is that one can usefully do it at all. Because it could be that most mathematical results one cares about would be undecidable. So why doesn’t that happen?

Well, if one considers arbitrary abstract systems it happens a lot. Take a typical cellular automaton—or a Turing machine—and ask whether it’s true that the system, say, always settles down to periodic behavior regardless of its initial state. Even something as simple as that will often be undecidable.

So why doesn’t this happen in mathematics? Maybe there’s something special about the particular axioms used in mathematics. And certainly if one thinks they’re the ones that uniquely describe science and the world there might be a reason for that. But one of the whole points of the book is that actually there’s a whole computational universe of possible rules that can be useful for doing science and describing the world.

And in fact I don’t think there’s anything abstractly special about the particular axioms that have traditionally been used in mathematics: I think they’re just accidents of history.

What about the theorems that people investigate in mathematics? Again, I think there’s a strong historical character to them. For all but the most trivial areas of mathematics, there’s a whole sea of undecidability out there. But somehow mathematics picks the islands where theorems can actually be proved—often particularly priding itself on places close to the sea of undecidability where the proof can only be done with great effort.

I’ve been interested in the whole network of published theorems in mathematics (it’s a thing to curate, like wars in history, or properties of chemicals). And one of the things I’m curious about is whether something there’s an inexorable sequence to the mathematics that’s done, or whether, in a sense, random parts are being picked.

And here, I think, there’s a considerable analogy to the kind of thing we were discussing before with language. What is a proof? Basically it’s a way of explaining to someone why something is true. I’ve made all sorts of automated proofs in which there are hundreds of steps, each perfectly verifiable by computer. But—like the innards of a neural net—what’s going on looks alien and not understandable by a human.

For a human to understand, there have to be familiar “conceptual waypoints”. It’s pretty much like with words in languages. If some particular part of a proof has a name (“Smith’s Theorem”), and has a known meaning, then it’s useful to us. But if it’s just a lump of undifferentiated computation, it won’t be meaningful to us.

In pretty much any axiom system, there’s an infinite set of possible theorems. But which ones are “interesting”? That’s really a human question. And basically it’s going to end up being ones with “stories”. In the book I show that for the simple case of basic logic, the theorems that have historically been considered interesting enough to be given names happen to be precisely the ones that are in some sense minimal.

But my guess is that for richer axiom systems pretty much anything that’s going to be considered “interesting” is going to have to be reached from things that are already considered interesting. It’s like building up words or concepts: you don’t get to introduce new ones unless you can directly relate them to existing ones.

In recent years I’ve wondered quite a bit about how inexorable or not progress is in a field like mathematics. Is there just one historical path that can be taken, say from arithmetic to algebra to the higher reaches of modern mathematics? Or are there an infinite diversity of possible paths, with completely different histories for mathematics?

The answer is going to depend on—in a sense—the “structure of metamathematical space”: just what is the network of true theorems that avoid the sea of undecidability? Maybe it’ll be different for different fields of mathematics, and some will be more “inexorable” (so it feels like the math is being “discovered”) than others (where it seems more like the math is arbitrary, and “invented”).

But to me one of the most interesting things is how close—when viewed in these kinds of terms—questions about the nature and character of mathematics end up being to questions about the nature and character of intelligence and AI. And it’s this kind of commonality that makes me realize just how powerful and general the ideas in A New Kind of Science actually are.

When Is There a Science?

There are some areas of science—like physics and astronomy—where the traditional mathematical approach has done quite well. But there are others—like biology, social science and linguistics—where it’s had a lot less to say. And one of the things I’ve long believed is that what’s needed to make progress in these areas is to generalize the kinds of models one’s using, to consider a broader range of what’s out there in the computational universe.

And indeed in the past 15 or so years there’s been increasing success in doing this. And there are lots of biological and social systems, for example, where models have now been constructed using simple programs.

But unlike with mathematical models which can potentially be “solved”, these computational models often show computational irreducibility, and are typically used by doing explicit simulations. This can be perfectly successful for making particular predictions, or for applying the models in technology. But a bit like for the automated proofs of mathematical theorems one might still ask, “is this really science?”.

Yes, one can simulate what a system does, but does one “understand” it? Well, the problem is that computational irreducibility implies that in some fundamental sense one can’t always “understand” things. There might be no useful “story” that can be told; there may be no “conceptual waypoints”—only lots of detailed computation.

Imagine that one’s trying to make a science of how the brain understands language—one of the big goals of linguistics. Well, perhaps we’ll get an adequate model of the precise rules which determine the firing of neurons or some other low-level representation of the brain. And then we look at the patterns generated in understanding some whole collection of sentences.

Well, what if those patterns look like the behavior of rule 30? Or, closer at hand, the innards of some recurrent neural network? Can we “tell a story” about what’s happening? To do so would basically require that we create some kind of higher-level symbolic representation: something where we effectively have words for core elements of what’s going on.

But computational irreducibility implies that there may ultimately be no way to create such a thing. Yes, it will always be possible to find patches of computational reducibility, where some things can be said. But there won’t be a complete story that can be told. And one might say there won’t be a useful reductionistic piece of science to be done. But that’s just one of the things that happens when one’s dealing with (as the title says) a new kind of science.

Controlling the AIs

People have gotten very worried about AI in recent years. They wonder what’s going to happen when AIs “get much smarter” than us humans. Well, the Principle of Computational Equivalence has one piece of good news: at some fundamental level, AIs will never be “smarter”—they’ll just be able to do computations that are ultimately equivalent to what our brains do, or, for that matter, what all sorts of simple programs do.

As a practical matter, of course, AIs will be able to process larger amounts of data more quickly than actual brains. And no doubt we’ll choose to have them run many aspects of the world for us—from medical devices, to central banks to transportation systems, and much more.

So then it’s important to figure how we’ll tell them what to do. As soon as we’re making serious use of what’s out there in the computational universe, we’re not going to be able to give a line-by-line description of what the AIs are going to do. Rather, we’re going to have to define goals for the AIs, then let them figure out how best to achieve those goals.

In a sense we’ve already been doing something like this for years in the Wolfram Language. There’s some high-level function that describes something you want to do (“lay out a graph”, “classify data”, etc.). Then it’s up to the language to automatically figure out the best way to do it.

And in the end the real challenge is to find a way to describe goals. Yes, you want to search for cellular automata that will make a “nice carpet pattern”, or a “good edge detector”. But what exactly do those things mean? What you need is a language that a human can use to say as precisely as possible what they mean.

It’s really the same problem as I’ve been talking about a lot here. One has to have a way for humans to be able to talk about things they care about. There’s infinite detail out there in the computational universe. But through our civilization and our shared cultural history we’ve come to identify certain concepts that are important to us. And when we describe our goals, it’s in terms of these concepts.

Three hundred years ago people like Leibniz were interested in finding a precise symbolic way to represent the content of human thoughts and human discourse. He was far too early. But now I think we’re finally in a position to actually make this work. In fact, we’ve already gotten a long way with the Wolfram Language in being able to describe real things in the world. And I’m hoping it’ll be possible to construct a fairly complete “symbolic discourse language” that lets us talk about the things we care about.

Right now we write legal contracts in “legalese” as a way to make them slightly more precise than ordinary natural language. But with a symbolic discourse language we’ll be able to write true “smart contracts” that describe in high-level terms what we want to have happen—and then machines will automatically be able to verify or execute the contract.

But what about the AIs? Well, we need to tell them what we generally want them to do. We need to have a contract with them. Or maybe we need to have a constitution for them. And it’ll be written in some kind of symbolic discourse language, that both allows us humans to express what we want, and is executable by the AIs.

There’s lots to say about what should be in an AI Constitution, and how the construction of such things might map onto the political and cultural landscape of the world. But one of the obvious questions is: can the constitution be simple, like Asimov’s Laws of Robotics?

And here what we know from A New Kind of Science tells us the answer: it can’t be. In a sense the constitution is an attempt to sculpt what can happen in the world and what can’t. But computational irreducibility says that there will be an unbounded collection of cases to consider.

For me it’s interesting to see how theoretical ideas like computational irreducibility end up impinging on these very practical—and central—societal issues. Yes, it all started with questions about things like the theory of all possible theories. But in the end it turns into issues that everyone in society is going to end up being concerned about.

There’s an Endless Frontier

Will we reach the end of science? Will we—or our AIs—eventually invent everything there is to be invented?

For mathematics, it’s easy to see that there’s an infinite number of possible theorems one can construct. For science, there’s an infinite number of possible detailed questions to ask. And there’s also an infinite array of possible inventions one can construct.

But the real question is: will there always be interesting new things out there?

Well, computational irreducibility says there will always be new things that need an irreducible amount of computational work to reach from what’s already there. So in a sense there’ll always be “surprises”, that aren’t immediately evident from what’s come before.

But will it just be like an endless array of different weirdly shaped rocks? Or will there be fundamental new features that appear, that we humans consider interesting?

It’s back to the very same issue we’ve encountered several times before: for us humans to find things “interesting” we have to have a conceptual framework that we can use to think about them. Yes, we can identify a “persistent structure” in a cellular automaton. Then maybe we can start talking about “collisions between structures”. But when we just see a whole mess of stuff going on, it’s not going to be “interesting” to us unless we have some higher-level symbolic way to talk about it.

In a sense, then, the rate of “interesting discovery” isn’t going to be limited by our ability to go out into the computational universe and find things. Instead, it’s going to be limited by our ability as humans to build a conceptual framework for what we’re finding.

It’s a bit like what happened in the whole development of what became A New Kind of Science. People had seen related phenomena for centuries if not millennia (distribution of primes, digits of pi, etc.). But without a conceptual framework they just didn’t seem “interesting”, and nothing was built around them. And indeed as I understand more about what’s out there in the computational universe—and even about things I saw long ago there—I gradually build up a conceptual framework that lets me go further.

By the way, it’s worth realizing that inventions work a little differently from discoveries. One can see something new happen in the computational universe, and that might be a discovery. But an invention is about figuring out how something can be achieved in the computational universe.

And—like in patent law—it isn’t really an invention if you just say “look, this does that”. You have to somehow understand a purpose that it’s achieving.

In the past, the focus of the process of invention has tended to be on actually getting something to work (“find the lightbulb filament that works”, etc.). But in the computational universe, the focus shifts to the question of what you want the invention to do. Because once you’ve described the goal, finding a way to achieve it is something that can be automated.

That’s not to say that it will always be easy. In fact, computational irreducibility implies that it can be arbitrarily difficult. Let’s say you know the precise rules by which some chemicals can interact. Can you find a chemical synthesis pathway that will let you get to some particular chemical structure? There may be a way, but computational irreducibility implies that there may be no way to find out how long the pathway may be. And if you haven’t found a pathway you may never be sure if it’s because there isn’t one, or just because you didn’t reach it yet.

The Fundamental Theory of Physics

If one thinks about reaching the edge of science, one cannot help but wonder about the fundamental theory of physics. Given everything we’ve seen in the computational universe, is it conceivable that our physical universe could just correspond to one of those programs out there in the computational universe?

Of course, we won’t really know until or unless we find it. But in the years since A New Kind of Science appeared, I’ve become ever more optimistic about the possibilities.

Needless to say, it would be a big change for physics. Today there are basically two major frameworks for thinking about fundamental physics: general relativity and quantum field theory. General relativity is a bit more than 100 years old; quantum field theory maybe 90. And both have achieved spectacular things. But neither has succeeded in delivering us a complete fundamental theory of physics. And if nothing else, I think after all this time, it’s worth trying something new.

But there’s another thing: from actually exploring the computational universe, we have a huge amount of new intuition about what’s possible, even in very simple models. We might have thought that the kind of richness we know exists in physics would require some very elaborate underlying model. But what’s become clear is that that kind of richness can perfectly well emerge even from a very simple underlying model.

What might the underlying model be like? I’m not going to discuss this in great detail here, but suffice it to say that I think the most important thing about the model is that it should have as little as possible built in. We shouldn’t have the hubris to think we know how the universe is constructed; we should just take a general type of model that’s as unstructured as possible, and do what we typically do in the computational universe: just search for a program that does what we want.

My favorite formulation for a model that’s as unstructured as possible is a network: just a collection of nodes with connections between them. It’s perfectly possible to formulate such a model as an algebraic-like structure, and probably many other kinds of things. But we can think of it as a network. And in the way I’ve imagined setting it up, it’s a network that’s somehow “underneath” space and time: every aspect of space and time as we know it must emerge from the actual behavior of the network.

Over the past decade or so there’s been increasing interest in things like loop quantum gravity and spin networks. They’re related to what I’ve been doing in the same way that they also involve networks. And maybe there’s some deeper relationship. But in their usual formulation, they’re much more mathematically elaborate.

From the point of view of the traditional methods of physics, this might seem like a good idea. But with the intuition we have from studying the computational universe—and using it for science and technology—it seems completely unnecessary. Yes, we don’t yet know the fundamental theory of physics. But it seems sensible to start with the simplest hypothesis. And that’s definitely something like a simple network of the kind I’ve studied.

At the outset, it’ll look pretty alien to people (including myself) trained in traditional theoretical physics. But some of what emerges isn’t so alien. A big result I found nearly 20 years ago (that still hasn’t been widely understood) is that when you look at a large enough network of the kind I studied you can show that its averaged behavior follows Einstein’s equations for gravity. In other words, without putting any fancy physics into the underlying model, it ends up automatically emerging. I think it’s pretty exciting.

People ask a lot about quantum mechanics. Yes, my underlying model doesn’t build in quantum mechanics (just as it doesn’t build in general relativity). Now, it’s a little difficult to pin down exactly what the essence of “being quantum mechanical” actually is. But there are some very suggestive signs that my simple networks actually end up showing what amounts to quantum behavior—just like in the physics we know.

OK, so how should one set about actually finding the fundamental theory of physics if it’s out there in the computational universe of possible programs? Well, the obvious thing is to just start searching for it, starting with the simplest programs.

I’ve been doing this—more sporadically than I would like—for the past 15 years or so. And my main discovery so far is that it’s actually quite easy to find programs that aren’t obviously not our universe. There are plenty of programs where space or time are obviously completely different from the way they are in our universe, or there’s some other pathology. But it turns out it’s not so difficult to find candidate universes that aren’t obviously not our universe.

But we’re immediately bitten by computational irreducibility. We can simulate the candidate universe for billions of steps. But we don’t know what it’s going to do—and whether it’s going to grow up to be like our universe, or completely different.

It’s pretty unlikely that in looking at that tiny fragment of the very beginning of a universe we’re going to ever be able to see anything familiar, like a photon. And it’s not at all obvious that we’ll be able to construct any kind of descriptive theory, or effective physics. But in a sense the problem is bizarrely similar to the one we have even in systems like neural networks: there’s computation going on there, but can we identify “conceptual waypoints” from which we can build up a theory that we might understand?

It’s not at all clear our universe has to be understandable at that level, and it’s quite possible that for a very long time we’ll be left in the strange situation of thinking we might have “found our universe” out in the computational universe, but not being sure.

Of course, we might be lucky, and it might be possible to deduce an effective physics, and see that some little program that we found ends up reproducing our whole universe. It would be a remarkable moment for science. But it would immediately raise a host of new questions—like why this universe, and not another?

Box of a Trillion Souls

Right now us humans exist as biological systems. But in the future it’s certainly going to be technologically possible to reproduce all the processes in our brains in some purely digital—computational—form. So insofar as those processes represent “us”, we’re going to be able to be “virtualized” on pretty much any computational substrate. And in this case we might imagine that the whole future of a civilization could wind up in effect as a “box of a trillion souls”.

Inside that box there would be all kinds of computations going on, representing the thoughts and experiences of all those disembodied souls. Those computations would reflect the rich history of our civilization, and all the things that have happened to us. But at some level they wouldn’t be anything special.

It’s perhaps a bit disappointing, but the Principle of Computational Equivalence tells us that ultimately these computations will be no more sophisticated than the ones that go on in all sorts of other systems—even ones with simple rules, and no elaborate history of civilization. Yes, the details will reflect all that history. But in a sense without knowing what to look for—or what to care about—one won’t be able to tell that there’s anything special about it.

OK, but what about for the “souls” themselves? Will one be able to understand their behavior by seeing that they achieve certain purposes? Well, in our current biological existence, we have all sorts of constraints and features that give us goals and purposes. But in a virtualized “uploaded” form, most of these just go away.

I’ve thought quite a bit about how “human” purposes might evolve in such a situation, recognizing, of course, that in virtualized form there’s little difference between human and AI. The disappointing vision is that perhaps the future of our civilization consists in disembodied souls in effect “playing videogames” for the rest of eternity.

But what I’ve slowly realized is that it’s actually quite unrealistic to project our view of goals and purposes from our experience today into that future situation. Imagine talking to someone from a thousand years ago and trying to explain that people in the future would be walking on treadmills every day, or continually sending photographs to their friends. The point is that such activities don’t make sense until the cultural framework around them has developed.

It’s the same story yet again as with trying to characterize what’s interesting or what’s explainable. It relies on the development of a whole network of conceptual waypoints.

Can we imagine what the mathematics of 100 years from now will be like? It depends on concepts we don’t yet know. So similarly if we try to imagine human motivation in the future, it’s going to rely on concepts we don’t know. Our best description from today’s viewpoint might be that those disembodied souls are just “playing videogames”. But to them there might be a whole subtle motivation structure that they could only explain by rewinding all sorts of steps in history and cultural development.

By the way, if we know the fundamental theory of physics then in a sense we can make the virtualization complete, at least in principle: we can just run a simulation of the universe for those disembodied souls. Of course, if that’s what’s happening, then there’s no particular reason it has to be a simulation of our particular universe. It could as well be any universe from out in the computational universe.

Now, as I’ve mentioned, even in any given universe one will never in a sense run out of things to do, or discover. But I suppose I myself at least find it amusing to imagine that at some point those disembodied souls might get bored with just being in a simulated version of our physical universe—and might decide it’s more fun (whatever that means to them) to go out and explore the broader computational universe. Which would mean that in a sense the future of humanity would be an infinite voyage of discovery in the context of none other than A New Kind of Science!

The Economics of the Computational Universe

Long before we have to think about disembodied human souls, we’ll have to confront the issue of what humans should be doing in a world where more and more can be done automatically by AIs. Now in a sense this issue is nothing new: it’s just an extension of the long-running story of technology and automation. But somehow this time it feels different.

And I think the reason is in a sense just that there’s so much out there in the computational universe, that’s so easy to get to. Yes, we can build a machine that automates some particular task. We can even have a general-purpose computer that can be programmed to do a full range of different tasks. But even though these kinds of automation extend what we can do, it still feels like there’s effort that we have to put into them.

But the picture now is different—because in effect what we’re saying is that if we can just define the goal we want to achieve, then everything else will be automatic. All sorts of computation, and, yes, “thinking”, may have to be done, but the idea is that it’s just going to happen, without human effort.

At first, something seems wrong. How could we get all that benefit, without putting in more effort? It’s a bit like asking how nature could manage to make all the complexity it does—even though when we build artifacts, even with great effort, they end up far less complex. The answer, I think, is it’s mining the computational universe. And it’s exactly the same thing for us: by mining the computational universe, we can achieve essentially an unbounded level of automation.

If we look at the important resources in today’s world, many of them still depend on actual materials. And often these materials are literally mined from the Earth. Of course, there are accidents of geography and geology that determine by whom and where that mining can be done. And in the end there’s a limit (if often very large) to the amount of material that’ll ever be available.

But when it comes to the computational universe, there’s in a sense an inexhaustible supply of material—and it’s accessible to anyone. Yes, there are technical issues about how to “do the mining”, and there’s a whole stack of technology associated with doing it well. But the ultimate resource of the computational universe is a global and infinite one. There’s no scarcity and no reason to be “expensive”. One just has to understand that it’s there, and take advantage of it.

The Path to Computational Thinking

Probably the greatest intellectual shift of the past century has been the one towards the computational way of thinking about things. I’ve often said that if one picks almost any field “X”, from archaeology to zoology, then by now there either is, or soon will be, a field called “computational X”—and it’s going to be the future of the field.

I myself have been deeply involved in trying to enable such computational fields, in particular through the development of the Wolfram Language. But I’ve also been interested in what is essentially the meta problem: how should one teach abstract computational thinking, for example to kids? The Wolfram Language is certainly important as a practical tool. But what about the conceptual, theoretical foundations?

Well, that’s where A New Kind of Science comes in. Because at its core it’s discussing the pure abstract phenomenon of computation, independent of its applications to particular fields or tasks. It’s a bit like with elementary mathematics: there are things to teach and understand just to introduce the ideas of mathematical thinking, independent of their specific applications. And so it is too with the core of A New Kind of Science. There are things to learn about the computational universe that give intuition and introduce patterns of computational thinking—quite independent of detailed applications.

One can think of it as a kind of “pre computer science” , or “pre computational X”. Before one gets into discussing the specifics of particular computational processes, one can just study the simple but pure things one finds in the computational universe.

And, yes, even before kids learn to do arithmetic, it’s perfectly possible for them to fill out something like a cellular automaton coloring book—or to execute for themselves or on a computer a whole range of different simple programs. What does it teach? Well, it certainly teaches the idea that there can be definite rules or algorithms for things—and that if one follows them one can create useful and interesting results. And, yes, it helps that systems like cellular automata make obvious visual patterns, that for example one can even find in nature (say on mollusc shells).

As the world becomes more computational—and more things are done by AIs and by mining the computational universe—there’s going to an extremely high value not only in understanding computational thinking, but also in having the kind of intuition that develops from exploring the computational universe and that is, in a sense, the foundation for A New Kind of Science.

What’s Left to Figure Out?

My goal over the decade that I spent writing A New Kind of Science was, as much as possible, to answer all the first round of “obvious questions” about the computational universe. And looking back 15 years later I think that worked out pretty well. Indeed, today, when I wonder about something to do with the computational universe, I find it’s incredibly likely that somewhere in the main text or notes of the book I already said something about it.

But one of the biggest things that’s changed over the past 15 years is that I’ve gradually begun to understand more of the implications of what the book describes. There are lots of specific ideas and discoveries in the book. But in the longer term I think what’s most significant is how they serve as foundations, both practical and conceptual, for a whole range of new things that one can now understand and explore.

But even in terms of the basic science of the computational universe, there are certainly specific results one would still like to get. For example, it would be great to get more evidence for or against the Principle of Computational Equivalence, and its domain of applicability.

Like most general principles in science, the whole epistemological status of the Principles of Computational Equivalence is somewhat complicated. Is it like a mathematical theorem that can be proved? Is it like a law of nature that might (or might not) be true about the universe? Or is it like a definition, say of the very concept of computation? Well, much like, say, the Second Law of Thermodynamics or Evolution by Natural Selection, it’s a combination of these.

But one thing that’s significant is that it’s possible to get concrete evidence for (or against) the Principle of Computational Equivalence. The principle says that even systems with very simple rules should be capable of arbitrarily sophisticated computation—so that in particular they should be able to act as universal computers.

And indeed one of the results of the book is that this is true for one of the simplest possible cellular automata (rule 110). Five years after the book was published I decided to put up a prize for evidence about another case: the simplest conceivably universal Turing machine. And I was very pleased that in just a few months the prize was won, the Turing machine was proved universal, and there was another piece of evidence for the Principle of Computational Equivalence.

There’s a lot to do in developing the applications of A New Kind of Science. There are models to be made of all sorts of systems. There’s technology to be found. Art to be created. There’s also a lot to do in understanding the implications.

But it’s important not to forget the pure investigation of the computational universe. In the analogy of mathematics, there are applications to be pursued. But there’s also a “pure mathematics” that’s worth pursuing in its own right. And so it is with the computational universe: there’s a huge amount to explore just at an abstract level. And indeed (as the title of the book implies) there’s enough to define a whole new kind of science: a pure science of the computational universe. And it’s the opening of that new kind of science that I think is the core achievement of A New Kind of Science—and the one of which I am most proud.


For the 10th anniversary of A New Kind of Science, I wrote three posts:

The complete high-resolution A New Kind of Science is now available on the web. There are also a limited number of print copies of the book still available (all individually coded!).

]]>
0
Stephen Wolfram <![CDATA[Machine Learning for Middle Schoolers]]> http://blog.internal.stephenwolfram.com/?p=13158 2017-05-22T19:00:21Z 2017-05-11T19:50:25Z Machine Learning for Middle Schoolers(An Elementary Introduction to the Wolfram Language is available in print, as an ebook, and free on the web—as well as in Wolfram Programming Lab in the Wolfram Open Cloud. There’s also now a free online hands-on course based on it.) A year ago I published a book entitled An Elementary Introduction to the Wolfram [...]]]> Machine Learning for Middle Schoolers

(An Elementary Introduction to the Wolfram Language is available in print, as an ebook, and free on the web—as well as in Wolfram Programming Lab in the Wolfram Open Cloud. There’s also now a free online hands-on course based on it.)

An Elementary Introduction to the Wolfram LanguageA year ago I published a book entitled An Elementary Introduction to the Wolfram Language—as part of my effort to teach computational thinking to the next generation. I just published the second edition of the book—with (among other things) a significantly extended section on modern machine learning.

I originally expected my book’s readers would be high schoolers and up. But it’s actually also found a significant audience among middle schoolers (11- to 14-year-olds). So the question now is: can one teach the core concepts of modern machine learning even to middle schoolers? Well, the interesting thing is that—thanks to the whole technology stack we’ve now got in the Wolfram Language—the answer seems to be “yes”!

Here’s what I did in the book:


After this main text, the book has Exercises, Q&A and Tech Notes.

Exercises, Q&A, Tech Notes

The Backstory

What was my thinking behind this machine learning section? Well, first, it has to fit into the flow of the book—using only concepts that have already been introduced, and, when possible, reinforcing them. So it can talk about images, and real-world data, and graphs, and text—but not functional programming or external data resources.

Chapter list

With modern machine learning, it’s easy to show “wow” examples—like our imageidentify.com website from 2015 (based on the Wolfram Language ImageIdentify function). But my goal in the book was also to communicate a bit of the background and intuition of how machine learning works, and where it can be used.

I start off by explaining that machine learning is different from traditional “programming”, because it’s based on learning from examples, rather than on explicitly specifying computational steps. The first thing I discuss is something that doesn’t really need all the fanciness of modern neural-net machine learning: it’s recognizing what languages text fragments are from:

LanguageIdentify[{"thank you", "merci", "dar las gracias", "感謝",    "благодарить"}]

Kids (and other people) can sort of imagine (or discuss in a classroom) how something like this might work—looking words up in dictionaries, etc. And I think it’s useful to give a first example that doesn’t seem like “pure magic”. (In reality, LanguageIdentify uses a combination of traditional lookup, and modern machine learning techniques.)

But then I give a much more “magic” example—of ImageIdentify:

ImageIdentify[]

I don’t immediately try to explain how it works, but instead go on to something different: sentiment analysis. Kids have lots of fun trying out sentiment analysis. But the real point here is that it shows the idea of making a “classifier”: there are an infinite number of possible inputs, but only (in this case) 3 possible outputs:

Classify["Sentiment", "I'm so excited to be programming"]

Having seen this, we’re ready to give a little more indication of how something like this works. And what I do is to show the function Classify classifying handwritten digits into 0s and 1s. I’m not saying what’s going on inside, but people can get the idea that Classify is given a bunch of examples, and then it’s using those to classify a particular input as being 0 or 1:

Classify[]

OK, but how does it do this? In reality one’s dealing with ideas about attractors—and inputs that lie in the basins of attraction for particular outputs. But in a first approximation, one can say that inputs that are “nearer to”, say, the 0 examples are taken to be 0s, and inputs that are nearer to the 1 examples are taken to be 1s.

People don’t usually have much difficulty with that explanation—unless they start to think too hard about what “nearest” might really mean in this context. But rather than concentrating on that, what I do in the book is just to talk about the case of numbers, where it’s really easy to see what “nearest” means:

Nearest[{10, 20, 30, 40, 50, 60, 70, 80}, 22]

Nearest isn’t the most exciting function to play with: one potentially puts a lot of things in, and then just one “nearest thing” comes out. Still, Nearest is nice because its functionality is pretty easy to understand (and one can have reasonable guesses about algorithms it could use).

Having seen Nearest for numbers, I show Nearest for colors. In the book, I’ve already talked about how colors are represented by red-green-blue triples of numbers, so this isn’t such a stretch—but seeing Nearest operate on colors begins to make it a little more plausible that it could operate on things like images too.

Nearest[]

Next I show the case of words. In the book, I’ve already done quite a bit with strings and words. In the main text I don’t talk about the precise definition of “nearness” for words, but again, kids easily get the basic idea. (In a Tech Note, I do talk about EditDistance, another good algorithmic operation that people can think about and try out.)

Nearest[WordList[], "good", 10]

OK, so how does one get from here to something like ImageIdentify? The approach I used is to talk next about OCR and TextRecognize. This doesn’t seem as “magic” as ImageIdentify (and lots of people know about “OCR’ing documents”), but it’s a good place to get a further idea of what ImageIdentify is doing.

Turning a piece of text into an image, and then back into the same text again, doesn’t seem that impressive or useful. But it gets more interesting if one blurs the text out (and, yes, blurring an image is something I talked about earlier in the book):

Table[Blur[Rasterize[Style["hello", 20]], r], {r, 0, 4}]

Given the blurred image, the question is: can one still recognize the text? At this stage in the book I haven’t talked about /@ (Map) or % (last output) yet, so I have to write the code out a bit more verbosely. But the result is:

TextRecognize /@ %

And, yes, when the image isn’t too blurred, TextRecognize can recognize the text, but when the text gets too blurred, it stops being able to. I like this example, because it shows something impressive—but not “magic”—happening. And I think it’s useful to show both where machine learning-based functions succeed, and where they fail. By the way, the result here is different from the one in the book—because the text font is different, and those details matter when one’s on the edge of what can be recognized. (If one was doing this in a class, for example, one might try some different fonts and sizes, and discuss why some survive more blurring than others.)

TextRecognize shows how one can effectively do something like ImageIdentify, but with just 26 letterforms (well, actually, TextRecognize handles many more glyphs than that). But now in the book I show ImageIdentify again, blurring like we did with letters:

Table[Blur[], r], {r, 0, 22, 2}]

ImageIdentify /@ %

It’s fun to see what it does, but it’s also helpful. Because it gives a sense of the “attractor” around the “cheetah” concept: stay fairly close and the cheetah can still be recognized; go too far away and it can’t. (A slightly tricky issue is that we’re continually producing new, better neural nets for ImageIdentify—so even between when the book was finished and today there’ve been some new nets—and it so happens they give different results for the not-a-cheetah cases. Presumably the new results are “better”, though it’s not clear what that means, given that we don’t have an official right-answer “blurred cheetah” category, and who’s to say whether the blurriest image is more like a whortleberry or a person.)

I won’t go through my whole discussion of machine learning from the book here. Suffice it to say that after discussing explicitly trained functions like TextRecognize and ImageIdentify, I start discussing “unsupervised learning”, and things like clustering in feature space. I think our new FeatureSpacePlot is particularly helpful.

It’s fairly clear what it means to arrange colors:

FeatureSpacePlot[RandomColor[100]]

But then one can “do the same thing” with images of letters. (In the book the code is a little longer, because I haven’t talked about /@ yet.)

FeatureSpacePlot[Rasterize /@ Alphabet[]]

And what’s nice about this is that—as well as being useful in its own right—it also reinforces the idea of how something like TextRecognize might work by finding the “nearest letter” to whatever input it’s given.

My final example in the section uses photographs. FeatureSpacePlot does a nice job of separating images of different kinds of things—again giving an idea of how ImageIdentify might work:

FeatureSpacePlot[{}]

Obviously in just 10 pages in an elementary book I’m not able to give a complete exposition of modern machine learning. But I was pleased to see how many of the core concepts I was able to touch on.

Of course, the fact that this was possible at all depends critically on our whole Wolfram Language technology stack. Whether it’s the very fact that we have machine learning in the language, or the fact that we can seamlessly work with images or text or whatever, or the whole (28-year-old!) Wolfram Notebook system that lets us put all these pieces together—all these pieces are critical to making it possible to bring modern machine learning to people like middle schoolers.

And what I really like is that what one gets to do isn’t toy stuff: one can take what I’m discussing in the book, and immediately apply it in real-world situations. At some level the fact that this works is a reflection of the whole automation emphasis of the Wolfram Language: there’s very sophisticated stuff going on inside, but it’s automated at all levels, so one doesn’t need to be an expert and understand the details to be able to use it—or to get a good intuition about what can work and what can’t.

Going Further

OK, so how would one go further in teaching machine learning?

One early thing might be to start talking about probabilities. ImageIdentify has various possible choices of identifications, but what probabilities does it assign to them?
ImageIdentify[, All, 10, "Probability"]

This can lead to a useful discussion about prior probabilities, and about issues like trading off specificity for certainty.

But the big thing to talk about is training. (After all, “machine learning trainer” will surely be a big future career for some of today’s middle schoolers…) And the good news is that in the Wolfram Language environment, it’s possible to make training work with only a modest amount of data.

Let’s get some examples of images of characters from Guardians of the Galaxy by searching the web (we’re using an external search API, so you unfortunately can’t do exactly this on the Open Cloud):

data = AssociationMap[ WebImageSearch[#, "Thumbnails"] &, {"Star-Lord", "Gamora", "Groot", "Rocket Raccoon"}]

Now we can use these images as training material to create a classifier:

classifier=Classify[data]

And, sure enough, it can identify Rocket:

classifier[]

And, yes, it thinks a real raccoon is him too:

classifier[]

How does it do it? Well, let’s look at FeatureSpacePlot:

FeatureSpacePlot[Flatten[Values[data]]]

Some of this looks good—but some looks confusing. Because it’s arranging some of the images not according to who they’re of, but just according to their background colors. And here we begin to see some of the subtlety of machine learning. The actual classifier we built works only because in the training examples for each character there were ones with different backgrounds—so it can figure out that background isn’t the only distinguishing feature.

Actually, there’s another critical thing as well: Classify isn’t starting from scratch in classifying the images. Because it’s already been pre-trained to pick out “good features” that help distinguish real-world images. In fact, it’s actually using everything it learned from the creation of ImageIdentify—and the tens of millions of images it saw in connection with that—to know up front what features it should pay attention to.

It’s a bit weird to see, but internally Classify is characterizing each image as a list of numbers, each associated with a different “feature”:

FeatureExtract[{}]

One can do an extreme version of this in which one insists that each image is reduced to just two numbers—and that’s essentially how FeatureSpacePlot determines where to position an image:

DimensionReduce[{}]

Under the Hood

OK, but what’s going on under the hood? Well, it’s complicated. But in the Wolfram Language it’s easy to see—and getting a look at it helps in terms of getting an intuition about how neural nets really work. So, for example, here’s the low-level Wolfram Language symbolic representation of the neural net that powers ImageIdentify:

net = NetModel["Wolfram ImageIdentify Net for WL 11.1"]

And there’s actually even more: just click and keep drilling down:

net = NetModel["Wolfram ImageIdentify Net for WL 11.1"]

And yes, this is hard to understand—certainly for middle schoolers, and even for professionals. But if we take this whole neural net object, and apply it to a picture of a tiger, it’ll do what ImageIdentify does, and tell us it’s a tiger:

net[]

But here’s a neat thing, made possible by a whole stack of functionality in the Wolfram Language: we can actually go “inside” the neural net, to get a sense of what’s happening. As an example, let’s just take the first 3 “layers” of the network, apply them to the tiger, and visualize what comes out:

Image /@ Take[net, 3][]

Basically what’s happening is that the network has made lots of copies of the original image, and then processed each of them to pick out a different aspect of the image. (What’s going on actually seems to be remarkably similar to the first few levels of visual processing in the brain.)

What if we go deeper into the network? Here’s what happens at layer 10. The images are more abstracted, and presumably pick out higher-level features:

Image /@ Take[Take[net, 10][],20]

Go to level 20, and the network is “thinking about” lots of little images:

ImageAssemble[Partition[Image /@ Take[net, 20][],30]]

But by level 28, it’s beginning to “come to some conclusions”, with only a few of its possible channels of activity “lighting up”:

ImageAdjust[ImageAssemble[Partition[Image /@ Take[net, 28][],50]]]

Finally, by level 31, all that’s left is an array of numbers, with a few peaks visible:

ListLinePlot[Take[net, 31][]]

And applying the very last layer of the network (a “softmax” layer) only a couple of peaks are left:

ListLinePlot[net[,None], PlotRange -> All]

And the highest one is exactly the one that corresponds to the concept of “tiger”:

net[,"TopProbabilities"]

I’m not imagining that middle schoolers will follow all these details (and no, nobody should be learning neural net layer types like they learn parts of the water cycle). But I think it’s really useful to see “inside” ImageIdentify, and get even a rough sense of how it works. To someone like me it still seems a little like magic that it all comes together as it does. But what’s great is that now with our latest Wolfram Language tools one can easily look inside, and start getting an intuition about what’s going on.

The Process of Training

The idea of the Wolfram Language Classify function is to do machine learning at the highest possible level—as automatically as possible, and building on as much pre-training as possible. But if one wants to get a more complete feeling for what machine learning is like, it’s useful to see what happens if one instead tries to just train a neural net from scratch.

There is an immediate practical issue though: to get a neural net, starting from scratch, to actually do anything useful, one typically has to give it a very large amount of training data—which is hard to collect and wrangle. But the good news here is that with the recent release of the Wolfram Data Repository we have a growing collection of ready-to-use training sets immediately available for use in the Wolfram Language.

Like here’s the classic MNIST handwritten digit training set, with its 60,000 training examples:

ResourceData["MNIST"]

One thing one can do with a training set like this is just feed a random sample of it into Classify. And sure enough this gives one a classifier function that’s essentially a simple version of TextRecognize for handwritten digits:

c = Classify[RandomSample[ResourceData["MNIST"], 1000]]

And even with just 1000 training examples, it does pretty well:

c[{}]

And, yes, we can use FeatureSpacePlot to see how the different digits tend to separate in feature space:

FeatureSpacePlot[First /@ RandomSample[ResourceData["MNIST"], 250]]

But, OK, what if we want to actually train a neural net from scratch, with none of the fancy automation of Classify? Well, first we have to set up a raw neural net. And conveniently, the Wolfram Language has a bunch of classic neural nets built in. Here one’s called LeNet:

lenet = NetModel["LeNet"]

It’s much simpler than the ImageIdentify net, but it’s still pretty complicated. But we don’t have to understand what’s inside it to start training it. Instead, in the Wolfram Language, we can just use NetTrain (which, needless to say, automatically applies all the latest GPU tricks and so on):

net = NetTrain[lenet, RandomSample[ResourceData["MNIST"], 1000]]

It’s pretty neat to watch the training happening, and to see the orange line of the neural net’s error rate for fitting the examples keep going down. After about 20 seconds, NetTrain decides it’s gone far enough, and generates a finally trained net—which works pretty well:

Net[{}]

If you stop the training early, it won’t do quite so well:

net = NetTrain[lenet, RandomSample[ResourceData["MNIST"], 1000], MaxTrainingRounds -> 1]

net[{}]

In the professional world of machine learning, there’s a whole art and science of figuring out the best parameters for training. But with what we’ve got now in the Wolfram Language, nothing is stopping a middle schooler from doing their own experiments, visualizing and analyzing the results, and getting as good an intuition as anyone.

What Are Neural Nets Made Of?

OK, so if we want to really get down to the lowest level, we have to talk about what neural nets are made of. I’m not sure how much of this is middle-school stuff—but as soon as one knows about graphs of functions, one can already explain quite a bit. Because, you see, the “layers” in a neural net are actually just functions, that take numbers in, and put numbers out.

Take layer 2 of LeNet. It’s essentially just a simple Ramp function, which we can immediately plot (and, yes, it looks like a ramp):

Plot[Ramp[x], {x, -1, 1}]

Neural nets don’t typically just deal with individual numbers, though. They deal with arrays (or “tensors”) of numbers—represented in the Wolfram Language as nested lists. And each layer takes an array of numbers in, and puts an array of numbers out. Here’s a typical single layer:

layer = NetInitialize[LinearLayer[4, "Input" -> 2]]

This particular layer is set up to take 2 numbers as input, and put 4 numbers out:

layer[{2, 3}]

It might seem to be doing something quite “random”, and actually it is. Because the actual function the layer is implementing is determined by yet another array of numbers, or “weights”—which NetInitialize here just sets randomly. Here’s what it set them to in this particular case:

NetExtract[layer, "Weights"]

Why is any of this useful? Well, the crucial point is that what NetTrain does is to progressively tweak the weights in each layer of a neural network to try to get the overall behavior of the net to match the training examples you gave.

There are two immediate issues, though. First, the structure of the network has to be such that it’s possible to get the behavior you want by using some appropriate set of weights. And second, there has to be some way to progressively tweak weights so as to get to appropriate values.

Well, it turns out a single LinearLayer like the one above can’t do anything interesting. Here’s a contour plot of (the first element of) its output, as a function of its two inputs. And as the name LinearLayer might suggest, we always get something flat and linear out:

ContourPlot[First[layer[{x, y}]], {x, -1, 1}, {y, -1, 1}]

But here’s the big discovery that makes neural nets useful: if we chain together several layers, it’s easy to get something much more complicated. (And, yes, in the Wolfram Language outputs from one layer get knitted into inputs to the next layer in a nice, automatic way.) Here’s an example with 4 layers—two linear layers and two ramps:

net = NetInitialize[   NetChain[{LinearLayer[10], Ramp, LinearLayer[1], Ramp},     "Input" -> 2]]

And now when we plot the function, it’s more complicated:

ContourPlot[net[{x, y}], {x, -1, 1}, {y, -1, 1}]

We can actually also look at an even simpler case—of a neural net with 3 layers, and just one number as final output. (For technical reasons, it’s nice to still have 2 inputs, though we’ll always set one of those inputs to the constant value of 1.)

net = NetInitialize[   NetChain[{LinearLayer[3], Ramp, LinearLayer[1]}, "Input" -> 2]]

Here’s what this particular network does as a function of its input:

Plot[net[{x, 1}], {x, -2, 2}]

Inside the network, there’s an array of 3 numbers being generated—and it turns out that “3” causes there to be at most 3 (+1) distinct linear parts in the function. Increase the 3 to 100, and things can get more complicated:

net = NetInitialize[   NetChain[{LinearLayer[100], Ramp, LinearLayer[1]}, "Input" -> 2]]
Plot[net[{x, 1}], {x, -2, 2}]

Now, the point is that this is in a sense a “random function”, determined by the particular random weights picked by NetInitialize. If we run NetInitialize a bunch of times, we’ll get a bunch of different results:

Table[With[{net =      NetInitialize[      NetChain[{LinearLayer[100], Ramp, LinearLayer[1]},        "Input" -> 2]]}, Plot[net[{x, 1}], {x, -2, 2}]], 8]

But the big question is: can we find an instance of this “random function” that’s useful for whatever we’re trying to do? Or, more particularly, can we find a random function that reproduces particular training examples?

Let’s imagine that our training examples give the values of the function at the dots in this plot (by the way, the setup here is more like machine learning in the style of Predict than Classify):

ListLinePlot[Table[Mod[n^2, 5], {n, 15}], Mesh -> All]

Here’s an instance of our network again:

net = NetInitialize[   NetChain[{LinearLayer[100], Ramp, LinearLayer[1]}, "Input" -> 2]]

And here’s a plot of what it initially does over the range of the training examples (and, yes, it’s obviously completely wrong):

Plot[net[{n, 1}], {n, 1, 15}]

Well, let’s just try training our network on our training data using NetTrain:

net = NetTrain[net, data = Table[{n, 1} -> {Mod[n^2, 5]}, {n, 15}]]

After about 20 seconds of training on my computer, there’s some vague sign that we’re beginning to reproduce at least some aspects of the original training data. But it’s at best slow going—and it’s not clear what’s eventually going to happen.

Plot[net[{n, 1}], {n, 1, 15}]

It’s a frontier question in neural net research just what structure of net will work best in any particular case (yes, we’re working on this question). But here let’s just try a slightly more complicated network:

net = NetInitialize[   NetChain[{LinearLayer[100], Tanh, LinearLayer[10], Ramp,      LinearLayer[1]}, "Input" -> 2]]

Random instances of this network don’t give very different results from our last network (though the presence of that Tanh layer makes the functions a bit smoother):

Tanh layer

But now let’s do some training (data was defined above):

net = NetTrain[net, data]

And here’s the result—which is surprisingly decent:

Plot[net[{n, 1}], {n, 1, 15}]

In fact, if we compare it to our original training data we see that the training values lie right on the function that the neural net produced:

Show[Plot[net[{n, 1}], {n, 1, 15}],   ListPlot[Table[Mod[n^2, 5], {n, 1, 15}], PlotStyle -> Red]]

Here’s what happened during the training process. The neural net effectively “tried out” a bunch of different possibilities, finally settling on the result here:

Machine learning animation

In what sense is the result “correct”? Well, it fits the training examples, and that’s really all we can ask. Because that’s all the input we gave. How it “interpolates” between the training examples is really its own business.  We’d like it to learn to “generalize” from the data it’s given—but it can’t really deduce much about the whole distribution of the data from the few points it’s being given here, so the kind of smooth interpolation it’s doing is as good as anything.

Outside the range of the training values, the neural net does what seem to be fairly random things—but again, there’s no “right answer” so one can’t really fault it:

Plot[net[{n, 1}], {n, -5, 25}]

But the fact that with the arbitrariness and messiness of our original neural net, we were able to successfully train it at all is quite remarkable. Neural nets of pretty much the type we’re talking about here had actually been studied for more than 60 years—but until the modern “deep learning revolution” nobody knew that it was going to be practical to train them for real problems.

But now—particularly with everything we have now in the Wolfram Language—it’s easy for anyone to do this.

So Much to Explore

Modern machine learning is very new—so even many of the obvious experiments haven’t been tried yet. But with our whole Wolfram Language setup there’s a lot that even middle schoolers can do. For example (and I admit I’m curious about this as I write this post): one can ask just how much something like the tiny neural net we were studying can learn.

Here’s a plot of the lengths of the first 60 Roman numerals:

ListLinePlot[Table[StringLength[RomanNumeral[n]], {n, 60}]]

After a small amount of training, here’s what the network managed to reproduce:

NetTrain[net,    Table[{n, 1} -> {StringLength[RomanNumeral[n]]}, {n, 60}]];
Plot[%[{n, 1}], {n, 1, 60}]

And one might think that maybe this is the best it’ll ever do. But I was curious if it could eventually do better—and so I just let it train for 2 minutes on my computer. And here’s the considerably better result that came out:

NetTrain[net,    Table[{n, 1} -> {StringLength[RomanNumeral[n]]}, {n, 60}],    MaxTrainingRounds -> Quantity[2, "Minutes"]];

Plot[%[{n, 1}], {n, 1, 60}]

I think I can see why this particular thing works the way it does.  But seeing it suggests all sorts of new questions to pursue. But to me the most exciting point is the overarching one of just how wide open this territory is—and how easy it is now to explore it.

Yes, there are plenty of technical details—some fundamental, some superficial. But transcending all of these, there’s intuition to be developed. And that’s something that can perfectly well start with the middle schoolers…

]]>
1
Stephen Wolfram <![CDATA[Launching the Wolfram Data Repository: Data Publishing that Really Works]]> http://blog.internal.stephenwolfram.com/?p=12901 2017-09-05T17:15:13Z 2017-04-20T16:04:20Z launching-the-wolfram-data-repositoryAfter a Decade, It’s Finally Here! I’m pleased to announce that as of today, the Wolfram Data Repository is officially launched! It’s been a long road. I actually initiated the project a decade ago—but it’s only now, with all sorts of innovations in the Wolfram Language and its symbolic ways of representing data, as well [...]]]> launching-the-wolfram-data-repository

After a Decade, It’s Finally Here!

I’m pleased to announce that as of today, the Wolfram Data Repository is officially launched! It’s been a long road. I actually initiated the project a decade ago—but it’s only now, with all sorts of innovations in the Wolfram Language and its symbolic ways of representing data, as well as with the arrival of the Wolfram Cloud, that all the pieces are finally in place to make a true computable data repository that works the way I think it should.

Wolfram Data Respository

It’s happened to me a zillion times: I’m reading a paper or something, and I come across an interesting table or plot. And I think to myself: “I’d really like to get the data behind that, to try some things out”. But how can I get the data?

If I’m lucky there’ll be a link somewhere in the paper. But it’s usually a frustrating experience to follow it. Because even if there’s data there (and often there actually isn’t), it’s almost never in a form where one can readily use it. It’s usually quite raw—and often hard to decode, and perhaps even intertwined with text. And even if I can see the data I want, I almost always find myself threading my way through footnotes to figure out what’s going on with it. And in the end I usually just decide it’s too much trouble to actually pull out the data I want.

And I suppose one might think that this is just par for the course in working with data. But in modern times, we have a great counterexample: the Wolfram Language. It’s been one of my goals with the Wolfram Language to build into it as much data as possible—and make all of that data immediately usable and computable. And I have to say that it’s worked out great. Whether you need the mass of Jupiter, or the masses of all known exoplanets, or Alan Turing’s date of birth—or a trillion much more obscure things—you just ask for them in the language, and you’ll get them in a form where you can immediately compute with them.

Here’s the mass of Jupiter (and, yes, one can use “Wolfram|Alpha-style” natural language to ask for it):

Jupiter(planet)["Mass"]

Dividing it by the mass of the Earth immediately works:

Entity["Planet", "Jupiter"]["Mass"]/Entity["Planet", "Earth"]["Mass"]

Here’s a histogram of the masses of known exoplanets, divided by the mass of Jupiter:

Histogram[  EntityClass["Exoplanet", All]["Mass"]/   Entity["Planet", "Jupiter"]["Mass"]]

And here, for good measure, is Alan Turing’s date of birth, in an immediately computable form:

Alan Turing(person)["BirthDate"]

Of course, it’s taken many years and lots of work to make everything this smooth, and to get to the point where all those thousands of different kinds of data are fully integrated into the Wolfram Language—and Wolfram|Alpha.

But what about other data—say data from some new study or experiment? It’s easy to upload it someplace in some raw form. But the challenge is to make the data actually useful.

And that’s where the new Wolfram Data Repository comes in. Its idea is to leverage everything we’ve done with the Wolfram Language—and Wolfram|Alpha, and the Wolfram Cloud—to make it as easy as possible to make data as broadly usable and computable as possible.

There are many parts to this. But let me state our basic goal. I want it to be the case that if someone is dealing with data they understand well, then they should be able to prepare that data for the Wolfram Data Repository in as little as 30 minutes—and then have that data be something that other people can readily use and compute with.

It’s important to set expectations. Making data fully computable—to the standard of what’s built into the Wolfram Language—is extremely hard. But there’s a lower standard that still makes data extremely useful for many purposes. And what’s important about the Wolfram Data Repository (and the technology around it) is it now makes that standard easy to achieve—with the result that it’s now practical to publish data in a form that can really be used by many people.

An Example

Each item published in the Wolfram Data Repository gets its own webpage. Here, for example, is the page for a public dataset about meteorite landings:

Meteorite Landings

At the top is some general information about the dataset. But then there’s a piece of a Wolfram Notebook illustrating how to use the dataset in the Wolfram Language. And by looking at this notebook, one can start to see some of the real power of the Wolfram Data Repository.

One thing to notice is that it’s very easy to get the data. All you do is ask for ResourceData["Meteorite Landings"]. And whether you’re using the Wolfram Language on a desktop or in the cloud, this will give you a nice symbolic representation of data about 45716 meteorite landings (and, yes, the data is carefully cached so this is as fast as possible, etc.):

And then the important thing is that you can immediately start to do whatever computation you want on that dataset. As an example, this takes the "Coordinates" element from all rows, then takes a random sample of 1000 results, and geo plots them:

GeoListPlot[RandomSample[ResourceData["Meteorite Landings"][All, "Coordinates"],1000]]

Many things have to come together for this to work. First, the data has to be reliably accessible—as it is in the Wolfram Cloud. Second, one has to be able to tell where the coordinates are—which is easy if one can see the dataset in a Wolfram Notebook. And finally, the coordinates have to be in a form in which they can immediately be computed with.

This last point is critical. Just storing the textual form of a coordinate—as one might in something like a spreadsheet—isn’t good enough. One has to have it in a computable form. And needless to say, the Wolfram Language has such a form for geo coordinates: the symbolic construct GeoPosition[{lat, lon}].

There are other things one can immediately see from the meteorites dataset too. Like notice there’s a "Mass" column. And because we’re using the Wolfram Language, masses don’t have to just be numbers; they can be symbolic Quantity objects that correctly include their units. There’s also a "Year" column in the data, and again, each year is represented by an actual, computable, symbolic DateObject construct.

There are lots of different kinds of possible data, and one needs a sophisticated data ontology to handle them. But that’s exactly what we’ve built for the Wolfram Language, and for Wolfram|Alpha, and it’s now been very thoroughly tested. It involves 10,000 kinds of units, and tens of millions of “core entities”, like cities and chemicals and so on. We call it the Wolfram Data Framework (WDF)—and it’s one of the things that makes the Wolfram Data Repository possible.

What’s in the Wolfram Data Repository So Far?

Today is the initial launch of the Wolfram Data Repository, and to get ready for this launch we’ve been adding sample content to the repository for several months. Some of what we’ve added are “obvious” famous datasets. Some are datasets that we found for some reason interesting, or curious. And some are datasets that we created ourselves—and in some cases that I created myself, for example, in the course of writing my book A New Kind of Science.

WDR Home Page Categories

There’s plenty already in the Wolfram Data Repository that’ll immediately be useful in a variety of applications. But in a sense what’s there now is just an example of what can be there—and the kinds of things we hope and expect will be contributed by many other people and organizations.

The fact that the Wolfram Data Repository is built on top of our Wolfram Language technology stack immediately gives it great generality—and means that it can handle data of any kind. It’s not just tables of numerical data as one might have in a spreadsheet or simple database. It’s data of any type and structure, in any possible combination or arrangement.

Home Page Types

There are time series:

Take[ResourceData["US Federal Outlays by Agency"], 3]

There are training sets for machine learning:

RandomSample[ResourceData["MNIST"], 30]

There’s gridded data:

ResourceData["GMM-3 Mars Gravity Map"]

There’s the text of many books:

WordCloud[ResourceData["On the Origin of Species"]]

There’s geospatial data:

ResourceData["Global Air Navigation Aids"]

GeoHistogram[ResourceData"Global Air Navigation Aids"][All, "Geoposition"], 50, GeoRange-> United States]

Many of the data resources currently in the Wolfram Data Repository are quite tabular in nature. But unlike traditional spreadsheets or tables in databases, they’re not restricted to having just one level of rows and columns—because they’re represented using symbolic Wolfram Language Dataset constructs, which can handle arbitrarily ragged structures, of any depth.

ResourceData["Sample Data: Solar System Planets and Moons"]

But what about data that normally lives in relational or graph databases? Well, there’s a construct called EntityStore that was recently added to the Wolfram Language. We’ve actually been using something like it for years inside Wolfram|Alpha. But what EntityStore now does is to let you set up arbitrary networks of entities, properties and values, right in the Wolfram Language. It typically takes more curation than setting up something like a Dataset—but the result is a very convenient representation of knowledge, on which all the same functions can be used as with built-in Wolfram Language knowledge.

Here’s a data resource that’s an entity store:

ResourceData ["Museum of Modern Art Holdings and Artists"]

This adds the entity stores to the list of entity stores to be used automatically:

PrependTo [$EntityStores, %];

Now here are 5 random entities of type "MoMAArtist" from the entity store:

RandomEntity ["MoMAArtist", 5]

For each artist, one can extract a dataset of values:

Entity[MoMAArtist], Otto Mühl[Dataset]

This queries the entity store to find artists with the most recent birth dates:

EntityList[EntityClass["MoMAArtist", "BirthDate" -> TakeLargest[5]]]

How It Works

The Wolfram Data Repository is built on top of a new, very general thing in the Wolfram Language called the “resource system”. (Yes, expect all sorts of other repository and marketplace-like things to be rolling out shortly.)

The resource system has “resource objects”, that are stored in the cloud (using CloudObject), then automatically downloaded and cached on the desktop if necessary (using LocalObject). Each ResourceObject contains both primary content and metadata. For the Wolfram Data Repository, the primary content is data, which you can access using ResourceData.

ResourceData

The Wolfram Data Repository that we’re launching today is a public resource, that lives in the public Wolfram Cloud. But we’re also going to be rolling out private Wolfram Data Repositories, that can be run in Enterprise Private Clouds—and indeed inside our own company we’ve already set up several private data repositories, that contain internal data for our company.

There’s no limit in principle on the size of the data that can be stored in the Wolfram Data Repository. But for now, the “plumbing” is optimized for data that’s at most about a few gigabytes in size—and indeed the existing examples in the Wolfram Data Repository make it clear that an awful lot of useful data never even gets bigger than a few megabytes in size.

The Wolfram Data Repository is primarily intended for the case of definitive data that’s not continually changing. For data that’s constantly flowing in—say from IoT devices—we released last year the Wolfram Data Drop. Both Data Repository and Data Drop are deeply integrated into the Wolfram Language, and through our resource system, there’ll be some variants and combinations coming in the future.

Delivering Data to the World

Our goal with the Wolfram Data Repository is to provide a central place for data from all sorts of organizations to live—in such a way that it can readily be found and used.

Each entry in the Wolfram Data Repository has an associated webpage, which describes the data it contains, and gives examples that can immediately be run in the Wolfram Cloud (or downloaded to the desktop).

Open in Cloud

On the webpage for each repository entry (and in the ResourceObject that represents it), there’s also metadata, for indexing and searching—including standard Dublin Core bibliographic data. To make it easier to refer to the Wolfram Data Repository entries, every entry also has a unique DOI.

The way we’re managing the Wolfram Data Repository, every entry also has a unique readable registered name, that’s used both for the URL of its webpage, and for the specification of the ResourceObject that represents the entry.

It’s extremely easy to use data from the Wolfram Data Repository inside a Wolfram Notebook, or indeed in any Wolfram Language program. The data is ultimately stored in the Wolfram Cloud. But you can always download it—for example right from the webpage for any repository entry.

The richest and most useful form in which to get the data is the Wolfram Language or the Wolfram Data Framework (WDF)—either in ASCII or in binary. But we’re also setting it up so you can download in other formats, like JSON (and in suitable cases CSV, TXT, PNG, etc.) just by pressing a button.

Data Downloads

Of course, even formats like JSON don’t have native ways to represent entities, or quantities with units, or dates, or geo positions—or all those other things that WDF and the Wolfram Data Repository deal with. So if you really want to handle data in its full form, it’s much better to work directly in the Wolfram Language. But then with the Wolfram Language you can always process some slice of the data into some simpler form that does makes sense to export in a lower-level format.

How Contributions Work

The Wolfram Data Repository as we’re releasing it today is a platform for publishing data to the world. And to get it started, we’ve put in about 500 sample entries. But starting today we’re accepting contributions from anyone. We’re going to review and vet contributions much like we’ve done for the past decade for the Wolfram Demonstrations Project. And we’re going to emphasize contributions and data that we feel are of general interest.

But the technology of the Wolfram Data Repository—and the resource system that underlies it—is quite general, and allows people not just to publish data freely to the world, but also to share data in a more controlled fashion. The way it works is that people prepare their data just like they would for submission to the public Wolfram Data Repository. But then instead of actually submitting it, they just deploy it to their own Wolfram Cloud accounts, giving access to whomever they want.

And in fact, the general workflow is that even when people are submitting to the public Wolfram Data Repository, we’re going to expect them to have first deployed their data to their own Wolfram Cloud accounts. And as soon as they do that, they’ll get webpages and everything—just like in the public Wolfram Data Repository.

OK, so how does one create a repository entry? You can either do it programmatically using Wolfram Language code, or do it more interactively using Wolfram Notebooks. Let’s talk about the notebook way first.

You start by getting a template notebook. You can either do this through the menu item File > New > Data Resource, or you can use CreateNotebook["DataResource"]. Either way, you’ll get something that looks like this:

Data Resource Construction Notebook

Basically it’s then a question of “filling out the form”. A very important section is the one that actually provides the content for the resource:

Resource Content

Yes, it’s Wolfram Language code—and what’s nice is that it’s flexible enough to allow for basically any content you want. You can either just enter the content directly in the notebook, or you can have the notebook refer to a local file, or to a cloud object you have.

An important part of the Construction Notebook (at least if you want to have a nice webpage for your data) is the section that lets you give examples. When the examples are actually put up on the webpage, they’ll reference the data resource you’re creating. But when you’re filling in the Construction Notebook the resource hasn’t been created yet. The symbolic character of the Wolfram Language comes to the rescue, though. Because it lets you reference the content of the data resource symbolically as $$Data in the inputs that’ll be displayed, but lets you set $$Data to actual data when you’re working in the Construction Notebook to build up the examples.

Heart Rate Data

Alright, so once you’ve filled out the Construction Notebook, what do you do? There are two initial choices: set up the resource locally on your computer, or set it up in the cloud:

Private Deploy

And then, if you’re ready, you can actually submit your resource for publication in the public Wolfram Data Repository (yes, you need to get a Publisher ID, so your resource can be associated with your organization rather than just with your personal account):

Submit to the Wolfram Data Repository Page

It’s often convenient to set up resources in notebooks. But like everything else in our technology stack, there’s a programmatic Wolfram Language way to do it too—and sometimes this is what will be best.

Remember that everything that is going to be in the Wolfram Data Repository is ultimately a ResourceObject. And a ResourceObject—like everything else in the Wolfram Language—is just a symbolic expression, which happens to contain an association that gives the content and metadata of the resource object.

Well, once you’ve created an appropriate ResourceObject, you can just deploy it to the cloud using CloudDeploy. And when you do this, a private webpage associated with your cloud account will automatically be created. That webpage will in turn correspond to a CloudObject. And by setting the permissions of that cloud object, you can determine who will be able to look at the webpage, and who will be able to get the data that’s associated with it.

When you’ve got a ResourceObject, you can submit it to the public Wolfram Data Repository just by using ResourceSubmit.

By the way, all this stuff works not just for the main Wolfram Data Repository in the public Wolfram Cloud, but also for data repositories in private clouds. The administrator of an Enterprise Private Cloud can decide how they want to vet data resources that are submitted (and how they want to manage things like name collisions)—though often they may choose just to publish any resource that’s submitted.

The procedure we’ve designed for vetting and editing resources for the public Wolfram Data Repository is quite elaborate—though in any given case we expect it to run quickly. It involves doing automated tests on the incoming data and examples—and then ensuring that these continue working as changes are made, for example in subsequent versions of the Wolfram Language. Administrators of private clouds definitely don’t have to use this procedure—but we’ll be making our tools available if they want to.

Making a Data-Backed Publication

OK, so let’s say there’s a data resource in the Wolfram Data Repository. How can it actually be used to create a data-backed publication? The most obvious answer is just for the publication to include a link to the webpage for the data resource in the Wolfram Data Repository. And once people go to the page, it immediately shows them how to access the data in the Wolfram Language, use it in the Wolfram Open Cloud, download it, or whatever.

But what about an actual visualization or whatever that appears in the paper? How can people know how to make it? One possibility is that the visualization can just be included among the examples on the webpage for the data resource. But there’s also a more direct way, which uses Source Links in the Wolfram Cloud.

Here’s how it works. You create a Wolfram Notebook that takes data from the Wolfram Data Repository and creates the visualization:

Creating Visualizations with a Wolfram Notebook

Then you deploy this visualization to the Wolfram Cloud—either using Wolfram Language functions like CloudDeploy and EmbedCode, or using menu items. But when you do the deployment, you say to include a source link (SourceLink->Automatic in the Wolfram Language). And this means that when you get an embeddable graphic, it comes with a source link that takes you back to the notebook that made the graphic:

Sourcelink to the Notebook

So if someone is reading along and they get to that graphic, they can just follow its source link to see how it was made, and to see how it accesses data from the Wolfram Data Repository. With the Wolfram Data Repository you can do data-backed publishing; with source links you can also do full notebook-backed publishing.

The Big Win

Now that we’ve talked a bit about how the Wolfram Data Repository works, let’s talk again about why it’s important—and why having data in it is so valuable.

The #1 reason is simple: it makes data immediately useful, and computable.

There’s nice, easy access to the data (just use ResourceData["..."]). But the really important—and unique—thing is that data in the Wolfram Data Repository is stored in a uniform, symbolic way, as WDF, leveraging everything we’ve done with data over the course of so many years in the Wolfram Language and Wolfram|Alpha.

Why is it good to have data in WDF? First, because in WDF the meaning of everything is explicit: whether it’s an entity, or quantity, or geo position, or whatever, it’s a symbolic element that’s been carefully designed and documented. (And it’s not just a disembodied collection of numbers or strings.) And there’s another important thing: data in WDF is already in precisely the form it’s needed for one to be able to immediately visualize, analyze or otherwise compute with it using any of the many thousands of functions that are built into the Wolfram Language.

Wolfram Notebooks are also an important part of the picture—because they make it easy to show how to work with the data, and give immediately runnable examples. Also critical is the fact that the Wolfram Language is so succinct and easy to read—because that’s what makes it possible to give standalone examples that people can readily understand, modify and incorporate into their own work.

In many cases using the Wolfram Data Repository will consist of identifying some data resource (say through a link from a document), then using the Wolfram Language in Wolfram Notebooks to explore the data in it. But the Wolfram Data Repository is fully integrated into the Wolfram Language, so it can be used wherever the language is used. Which means the data from the Wolfram Data Repository can be used not just in the cloud or on the desktop, but also in servers and so on. And, for example, it can also be used in APIs or scheduled tasks, using the exact same ResourceData functions as ever.

The most common way the Wolfram Data Repository will be used is one resource at a time. But what’s really great about the uniformity and standardization that WDF provides is that it allows different data resources to be used together: those dates or geo positions mean the same thing even in different data resources, so they can immediately be put together in the same analysis, visualization, or whatever.

The Wolfram Data Repository builds on the whole technology stack that we’ve been assembling for the past three decades. In some ways it’s just a sophisticated piece of infrastructure that makes a lot of things easier to do. But I can already tell that its implications go far beyond that—and that it’s going to have a qualitative effect on the extent to which people can successfully share and reuse a wide range of kinds of data.

The Process of Data Curation

It’s a big win to have data in the Wolfram Data Repository. But what’s involved in getting it there? There’s almost always a certain amount of data curation required.

Let’s take a look again at the meteorite landings dataset I showed earlier in this post. It started from a collection of data made available in a nicely organized way by NASA. (Quite often one has to scrape webpages or PDFs; this is a case where the data happens to be set up to be downloadable in a variety of convenient formats.)

Meteorite Landing Data from NASA

As is fairly typical, the basic elements of the data here are numbers and strings. So the first thing to do is to figure out how to map these to meaningful symbolic constructs in WDF. For example, the “mass” column is labeled as being “(g)”, i.e. in grams—so each element in it should get converted to Quantity[value,"Grams"]. It’s a little trickier, though, because for some rows—corresponding to some meteorites—the value is just blank, presumably because it isn’t known.

So how should that be represented? Well, because the Wolfram Language is symbolic it’s pretty easy. And in fact there’s a standard symbolic construct Missing[...] for indicating missing data, which is handled consistently in analysis and visualization functions.

As we start to look further into the dataset, we see all sorts of other things. There’s a column labeled “year”. OK, we can convert that into DateObject[{value}]—though we need to be careful about any BC dates (how would they appear in the raw data?).

Next there are columns “reclat” and “reclong”, as well as a column called “GeoLocation” that seems to combine these, but with numbers quoted a different precision. A little bit of searching suggests that we should just take reclat and reclong as the latitude and longitude of the meteorite—then convert these into the symbolic form GeoPosition[{lat,lon}].

To do this in practice, we’d start by just importing all the data:

Import ["https://data.nasa.gov/api/views/gh4g-9sfh/rows.csv?accessType=\ DOWNLOAD", "CSV"]

OK, let’s extract a sample row:

data [[2]]

Already there’s something unexpected: the date isn’t just the year, but instead it’s a precise time. So this needs to be converted:

Interpreter ["DateTime"][data[[2, 7]]]

Now we’ve got to reset this to correspond only to a date at a granularity of a year:

DateObject [{DateValue[Interpreter["DateTime"][data[[2, 7]]],     "Year"]}, "Year"]

Here is the geo position:

GeoPosition [{data[[2, -3]], data[[2, -2]]}]" title="GeoPosition[{data[[2, -3]], data[[2, -2]]}]

And we can keep going, gradually building up code that can be applied to each row of the imported data. In practice there are often little things that go wrong. There’s something missing in some row. There’s an extra piece of text (a “footnote”) somewhere. There’s something in the data that got misinterpreted as a delimiter when the data was provided for download. Each one of these needs to be handled—preferably with as much automation as possible.

But in the end we have a big list of rows, each of which needs to be assembled into an association, then all combined to make a Dataset object that can be checked to see if it’s good to go into the Wolfram Data Repository.

The Data Curation Hierarchy

The example above is fairly typical of basic curation that can be done in less than 30 minutes by any decently skilled user of the Wolfram Language. (A person who’s absorbed my book An Elementary Introduction to the Wolfram Language should, for example, be able to do it.)

It’s a fairly simple example—where notably the original form of the data was fairly clean. But even in this case it’s worth understanding what hasn’t been done. For example, look at the column labeled "Classification" in the final dataset. It’s got a bunch of strings in it. And, yes, we can do something like make a word cloud of these strings:

WordCloud [  Normal[ResourceData["Meteorite Landings"][All, "Classification"]]]

But to really make these values computable, we’d have to do more work. We’d have to figure out some kind of symbolic representation for meteorite classification, then we’d have to do curation (and undoubtedly ask some meteorite experts) to fit everything nicely into that representation. The advantage of doing this is that we could then ask questions about those values (“what meteorites are above L3?”), and expect to compute answers. But there’s plenty we can already do with this data resource without that.

My experience in general has been that there’s a definite hierarchy of effort and payoff in getting data to be computable at different levels—starting with the data just existing in digital form, and ending with the data being cleanly computable enough that it can be fully integrated in the core Wolfram Language, and used for repeated, systematic computations.

Data Hierarchy Levels

 

Let’s talk about this hierarchy a bit.

The zeroth thing, of course, is that the data has to exist. And the next thing is that it has to be in digital form. If it started on handwritten index cards, for example, it had better have been entered into a document or spreadsheet or something.

But then the next issue is: how are people supposed to get access to that document or spreadsheet? Well, a good answer is that it should be in some kind of accessible cloud—perhaps referenced with a definite URI. And for a lot of data repositories that exist out there, just making the data accessible like this is the end of the story.

But one has to go a lot further to make the data actually useful. The next step is typically to make sure that the data is arranged in some definite structure. It might be a set of rows and columns, or it might be something more elaborate, and, say, hierarchical. But the point is to have a definite, known structure.

In the Wolfram Language, it’s typically trivial to take data that’s stored in any reasonable format, and use Import to get it into the Wolfram Language, arranged in some appropriate way. (As I’ll talk about later, it might be a Dataset, it might be an EntityStore, it might just be a list of Image objects, or it might be all sorts of other things.)

But, OK, now things start getting more difficult. We need to be able to recognize, say, that such-and-such a column has entries representing countries, or pairs of dates, or animal species, or whatever. SemanticImport uses machine learning and does a decent job of automatically importing many kinds of data. But there are often things that have to be fixed. How exactly is missing data represented? Are there extra annotations that get in the way of automatic interpretation? This is where one starts needing experts, who really understand the data.

But let’s say one’s got through this stage. Well, then in my experience, the best thing to do is to start visualizing the data. And very often one will immediately see things that are horribly wrong. Some particular quantity was represented in several inconsistent ways in the data. Maybe there was some obvious transcription or other error. And so on. But with luck it’s fairly easy to transform the data to handle the obvious issues—though to actually get it right almost always requires someone who is an expert on the data.

What comes out of this process is typically very useful for many purposes—and it’s the level of curation that we’re expecting for things submitted to the Wolfram Data Repository.

It’ll be possible to do all sorts of analysis and visualization and other things with data in this form.

But if one wants, for example, to actually integrate the data into Wolfram|Alpha, there’s considerably more that has to be done. For a start, everything that can realistically be represented symbolically has to be represented symbolically. It’s not good enough to have random strings giving values of things—because one can’t ask systematic questions about those. And this typically requires inventing systematic ways to represent new kinds of concepts in the world—like the "Classification" for meteorites.

Wolfram|Alpha works by taking natural language input. So the next issue is: when there’s something in the data that can be referred to, how do people refer to it in natural language? Often there’ll be a whole collection of names for something, with all sorts of variations. One has to algorithmically capture all of the possibilities.

Next, one has to think about what kinds of questions will be asked about the data. In Wolfram|Alpha, the fact that the questions get asked in natural language forces a certain kind of simplicity on them. But it makes one also need to figure out just what the linguistics of the questions can be (and typically this is much more complicated than the linguistics for entities or other definite things). And then—and this is often a very difficult part—one has to figure out what people want to compute, and how they want to compute it.

At least in the world of Wolfram|Alpha, it turns out to be quite rare for people just to ask for raw pieces of data. They want answers to questions—that have to be computed with models, or methods, or algorithms, from the underlying data. For meteorites, they might want to know not the raw information about when a meteorite fell, but compute the weathering of the meteorite, based on when it fell, what climate it’s in, what it’s made of, and so on. And to have data successfully be integrated into Wolfram|Alpha, those kinds of computations all need to be there.

For full Wolfram|Alpha there’s even more. Not only does one have to be able to give a single answer, one has to be able to generate a whole report, that includes related answers, and presents them in a well-organized way.

It’s ultimately a lot of work. There are very few domains that have been added to Wolfram|Alpha with less than a few skilled person-months of work. And there are plenty of domains that have taken person-years or tens of person-years. And to get the right answers, there always has to be a domain expert involved.

Getting data integrated into Wolfram|Alpha is a significant achievement. But there’s further one can go—and indeed to integrate data into the Wolfram Language one has to go further. In Wolfram|Alpha people are asking one-off questions—and the goal is to do as well as possible on individual questions. But if there’s data in the Wolfram Language, people won’t just ask one-off questions with it: they’ll also do large-scale systematic computations. And this demands a much greater level of consistency and completeness—which in my experience rarely takes less than person-years per domain to achieve.

But OK. So where does this leave the Wolfram Data Repository? Well, the good news is that all that work we’ve put into Wolfram|Alpha and the Wolfram Language can be leveraged for the Wolfram Data Repository. It would take huge amounts of work to achieve what’s needed to actually integrate data into Wolfram|Alpha or the Wolfram Language. But given all the technology we have, it takes very modest amounts of work to make data already very useful. And that’s what the Wolfram Data Repository is about.

The Data Publishing Ecosystem

With the Wolfram Data Repository (and Wolfram Notebooks) there’s finally a great way to do true data-backed publishing—and to ensure that data can be made available in an immediately useful and computable way.

For at least a decade there’s been lots of interest in sharing data in areas like research and government. And there’ve been all sorts of data repositories created—often with good software engineering—with the result that instead of data just sitting on someone’s local computer, it’s now pretty common for it to be uploaded to a central server or cloud location.

But the problem has been that the data in these repositories is almost always in a quite raw form—and not set up to be generally meaningful and computable. And in the past—except in very specific domains—there’s been no really good way to do this, at least in any generality. But the point of the Wolfram Data Repository is to use all the development we’ve done on the Wolfram Language and WDF to finally be able to provide a framework for having data in an immediately computable form.

The effect is dramatic. One goes from a situation where people are routinely getting frustrated trying to make use of data to one in which data is immediately and readily usable. Often there’s been lots of investment and years of painstaking work put into accumulating some particular set of data. And it’s often sad to see how little the data actually gets used—even though it’s in principle accessible to anyone. But I’m hoping that the Wolfram Data Repository will provide a way to change this—by allowing data not just to be accessible, but also computable, and easy for anyone to immediately and routinely use as part of their work.

There’s great value to having data be computable—but there’s also some cost to making it so. Of course, if one’s just collecting the data now, and particularly if it’s coming from automated sources, like networks of sensors, then one can just set it up to be in nice, computable WDF right from the start (say by using the data semantics layer of the Wolfram Data Drop). But at least for a while there’s going to still be a lot of data that’s in the form of things like spreadsheets and traditional databases—-that don’t even have the technology to support the kinds of structures one would need to directly represent WDF and computable data.

So that means that there’ll inevitably have to be some effort put into curating the data to make it computable. Of course, with everything that’s now in the Wolfram Language, the level of tools available for curation has become extremely high. But to do curation properly, there’s always some level of human effort—and some expert input—that’s required. And a key question in understanding the post-Wolfram-Data-Repository data publishing ecosystem is who is actually going to do this work.

In a first approximation, it could be the original producers of the data—or it could be professional or other “curation specialists”—or some combination. There are advantages and disadvantages to all of these possibilities. But I suspect that at least for things like research data it’ll be most efficient to start with the original producers of the data.

The situation now with data curation is a little similar to the historical situation with document production. Back when I was first doing science (yes, in the 1970s) people handwrote papers, then gave them to professional typists to type. Once typed, papers would be submitted to publishers, who would then get professional copyeditors to copyedit them, and typesetters to typeset them for printing. It was all quite time consuming and expensive. But over the course of the 1980s, authors began to learn to type their own papers on a computer—and then started just uploading them directly to servers, in effect putting them immediately in publishable form.

It’s not a perfect analogy, but in both data curation and document editing there are issues of structure and formatting—and then there are issues that require actual understanding of the content. (Sometimes there are also more global “policy” issues too.) And for producing computable data, as for producing documents, almost always the most efficient thing will be to start with authors “typing their own papers”—or in the case of data, putting their data into WDF themselves.

Of course, to do this requires learning at least a little about computable data, and about how to do curation. And to assist with this we’re working with various groups to develop materials and provide training about such things. Part of what has to be communicated is about mechanics: how to move data, convert formats, and so on. But part of it is also about principles—and about how to make the best judgement calls in setting up data that’s computable.

We’re planning to organize “curate-a-thons” where people who know the Wolfram Language and have experience with WDF data curation can pair up with people who understand particular datasets—and hopefully quickly get all sorts of data that they may have accumulated over decades into computable form—and into the Wolfram Data Repository.

In the end I’m confident that a very wide range of people (not just techies, but also humanities people and so on) will be able to become proficient at data curation with the Wolfram Language. But I expect there’ll always be a certain mixture of “type it yourself” and “have someone type it for you” approaches to data curation. Some people will make their data computable themselves—or will have someone right there in their lab or whatever who does. And some people will instead rely on outside providers to do it.

Who will these providers be? There’ll be individuals or companies set up much like the ones who provide editing and publishing services today. And to support this we’re planning a “Certified Data Curator” program to help define consistent standards for people who will work with the originators of a wide range of different kinds of data putting it into computable form.

But in additional to individuals or specific “curation companies”, there are at least two other kinds of entities that have the potential to be major facilitators of making data computable.

The first is research libraries. The role of libraries at many universities is somewhat in flux these days. But something potentially very important for them to do is to provide a central place for organizing—and making computable—data from the university and beyond. And in many ways this is just a modern analog of traditional library activities like archiving and cataloging.

It might involve the library actually having a private cloud version of the Wolfram Data Repository—and it might involve the library having its own staff to do curation. Or it might just involve the library providing advice. But I’ve found there’s quite a bit of enthusiasm in the library community for this kind of direction (and it’s perhaps an interesting sign that at our company people involved in data curation have often originally been trained in library science).

In addition to libraries, another type of organization that should be involved in making data computable is publishing companies. Some might say that publishing companies have had it a bit easy in the last couple of decades. Back in the day, every paper they published involved all sorts of production work, taking it from manuscript to final typeset version. But for years now, authors have been delivering their papers in digital forms that publishers don’t have to do much work on.

With data, though, there’s again something for publishers to do, and again a place for them to potentially add great value. Authors can pretty much put raw data into public repositories for themselves. But what would make publishers visibly add value is for them to process (or “edit”) the data—putting in the work to make it computable. The investment and processes will be quite similar to what was involved on the text side in the past—it’s just that now instead of learning about phototypesetting systems, publishers should be learning about WDF and the Wolfram Language.

It’s worth saying that as of today all data that we accept into the Wolfram Data Repository is being made freely available. But we’re anticipating in the near future we’ll also incorporate a marketplace in which data can be bought and sold (and even potentially have meaningful DRM, at least if it’s restricted to being used in the Wolfram Language). It’ll also be possible to have a private cloud version of the Wolfram Data Repository—in which whatever organization that runs it can set up whatever rules it wants about contributions, subscriptions and access.

One feature of traditional paper publishing is the sense of permanence it provides: once even just a few hundred printed copies of a paper are on shelves in university libraries around the world, it’s reasonable to assume that the paper is going to be preserved forever. With digital material, preservation is more complicated.

If someone just deploys a data resource to their Wolfram Cloud account, then it can be available to the world—but only so long as the account is maintained. The Wolfram Data Repository, though, is intended to be something much more permanent. Once we’ve accepted a piece of data for the repository, our goal is to ensure that it’ll continue to be available, come what may. It’s an interesting question how best to achieve that, given all sorts of possible future scenarios in the world. But now that the Wolfram Data Repository is finally launched, we’re going to be working with several well-known organizations to make sure that its content is as securely maintained as possible.

Data-Backed Journals

The Wolfram Data Repository—and private versions of it—is basically a powerful, enabling technology for making data available in computable form. And sometimes all one wants to do is to make the data available.

But at least in academic publishing, the main point usually isn’t the data. There’s usually a “story to be told”—and the data is just backup for that story. Of course, having that data backing is really important—and potentially quite transformative. Because when one has the data, in computable form, it’s realistic for people to work with it themselves, reproducing or checking the research, and directly building on it themselves.

But, OK, how does the Wolfram Data Repository relate to traditional academic publishing? For our official Wolfram Data Repository we’re going to have definite standards for what we accept—and we’re going to concentrate on data that we think is of general interest or use. We have a whole process for checking the structure of data, and applying software quality assurance methods, as well as expert review, to it.

And, yes, each entry in the Wolfram Data Repository gets a DOI, just like a journal article. But for our official Wolfram Data Repository we’re focused on data—and not the story around it. We don’t see it as our role to check the methods by which the data was obtained, or to decide whether conclusions drawn from it are valid or not.

But given the Wolfram Data Repository, there are lots of new opportunities for data-backed academic journals that do in effect “tell stories”, but now have the infrastructure to back them up with data that can readily be used.

I’m looking forward, for example, to finally making the journal Complex Systems that I founded 30 years ago a true data-backed journal. And there are many existing journals where it makes sense to use versions of the Wolfram Data Repository (often in a private cloud) to deliver computable data associated with journal articles.

But what’s also interesting is that now that one can take computable data for granted, there’s a whole new generation of “Journal of Data-Backed ____” journals that become possible—that not only use data from the Wolfram Data Repository, but also actually present their results as Wolfram Notebooks that can immediately be rerun and extended (and can also, for example, contain interactive elements).

The Corporate Version

I’ve been talking about the Wolfram Data Repository in the context of things like academic journals. But it’s also important in corporate settings. Because it gives a very clean way to have data shared across an organization (or shared with customers, etc.).

Typically in a corporate setting one’s talking about private cloud versions. And of course these can have their own rules about how contributions work, and who can access what. And the data can not only be immediately used in Wolfram Notebooks, but also in automatically generated reports, or instant APIs.

It’s been interesting to see—during the time we’ve been testing the Wolfram Data Repository—just how many applications we’ve found for it within our own company.

There’s information that used to be on webpages, but is now in our private Wolfram Data Repository, and is now immediately usable for computation. There’s information that used to be in databases, and which required serious programming to access, but is now immediately accessible through the Wolfram Language. And there are all sorts of even quite small lists and so on that used to exist only in textual form, but are now computable data in our data repository.

It’s always been my goal to have a truly “computable company”—and putting in place our private Wolfram Data Repository is an important step in achieving this.

My Very Own Data

In addition to public and corporate uses, there are also great uses of Wolfram Data Repository technology for individuals—and particularly for individual researchers. In my own case, I’ve got huge amounts of data that I’ve collected or generated over the course of my life. I happen to be pretty organized at keeping things—but it’s still usually something of an adventure to remember enough to “bring back to life” data I haven’t dealt with in a decade or more. And in practice I make much less use of older data than I should—even though in many cases it took me immense effort to collect or generate the data in the first place.

But now it’s a different story. Because all I have to do is to upload data once and for all to the Wolfram Data Repository, and then it’s easy for me to get and use the data whenever I want to. Some data (like medical or financial records) I want just for myself, so I use a private cloud version of the Wolfram Data Repository. But other data I’ve been getting uploaded into the public Wolfram Data Repository.

Here’s an example. It comes from a page in my book A New Kind of Science:

Page 833 from A New Kind of Science

The page says that by searching about 8 trillion possible systems in the computational universe I found 199 that satisfy some particular criterion. And in the book I show examples of some of these. But where’s the data?

Well, because I’m fairly organized about such things, I can go into my file system, and find the actual Wolfram Notebook from 2001 that generated the picture in the book. And that leads me to a file that contains the raw data—which then takes a very short time to turn into a data resource for the Wolfram Data Repository:

Three-Color Cellular Automaton Rules that Double Their Input

We’ve been systematically mining data from my research going back into the 1980s—even from Mathematica Version 1 notebooks from 1988 (which, yes, still work today). Sometimes the experience is a little less inspiring. Like to find a list of people referenced in the index of A New Kind of Science, together with their countries and dates, the best approach seemed to be to scrape the online book website:

ResourceData["People Mentioned in Stephen Wolfram\[CloseCurlyQuote]s \ \[OpenCurlyDoubleQuote]A New Kind of Science\[CloseCurlyDoubleQuote]"]

And to get a list of the books I used while working on A New Kind of Science required going into an ancient FileMaker database. But now all the data—nicely merged with Open Library information deduced from ISBNs—is in a clean WDF form in the Wolfram Data Repository. So I can do such things as immediately make a word cloud of the titles of the books:

WordCloud[  StringRiffle[   Normal[ResourceData["Books in Stephen Wolfram's Library"][All,      "Title"]]]]

What It Means

Many things have had to come together to make today’s launch of the Wolfram Data Repository possible. In the modern software world it’s easy to build something that takes blobs of data and puts them someplace in the cloud for people to access. But what’s vastly more difficult is to have the data actually be immediately useful—and making that possible is what’s required the whole development of our Wolfram Language and Wolfram Cloud technology stack, which are now the basis for the Wolfram Data Repository.

But now that the Wolfram Data Repository exists—and private versions of it can be set up—there are lots of new opportunities. For the research community, the most obvious is finally being able to do genuine data-backed publication, where one can routinely make underlying data from pieces of research available in a way that people can actually use. There are variants of this in education—making data easy to access and use for educational exercises and projects.

In the corporate world, it’s about making data conveniently available across an organization. And for individuals, it’s about maintaining data in such a way that it can be readily used for computation, and built on.

But in the end, I see the Wolfram Data Repository as a key enabling technology for defining how one can work with data in the future—and I’m excited that after all this time it’s finally now launched and available to everyone.


To comment, please visit the copy of this post at the Wolfram Blog »

]]>
0