Theory

[icycalm]

http://insomnia.ac/commentary/on_emerge ... _miracles/

I had been nurturing the delusion that after the devastating arguments I brought forth in this article no one would so much as fucking dare to attempt defend the absurd notion of emergence in videogames -- at least not in my presence. And then I just got this little email, which also happens to be the first response I have received on this article. (Amusingly enough, no one has even linked it yet. Compare this to what happened with the hobag article -- yet more proof that the more valuable insights you give to people, the fewer of them will actually bother with them.)

Hello there, I'm Dustin. I stumbled upon Insomnia (ironically at 4:00 AM) this morning. I perused your site and read some of your articles in their entirety. Let me say that I enjoy Braid, Spelunky, Cave Story, and other games you deem worthy only of "artfags." Alright, you caught me. I want your cock. Bad. I'll cradle the balls and stroke the shaft while humming Cave Story's soundtrack. Obviously.

Get the fuck over yourself.

Making homophobic remarks towards people that disagree with you is juvenile [I know. It's one of the methods I employ in an effort to keep the child in me alive. --icy]. Posturing yourself as an enlightened philsopher, spewing walls of pseudo-intellectual psychobabble text does not make you smart. You are not some long-lost disciple of Baudrillard, as hard as you try. Guess what? Interest in Philosophyy and video games aren't mutually exclusive. I'm a Lewisian philosopher [lol --icy] through and through, and any intro-level Critical Thinking student [lolololol --icy] could see that all of your articles are nothing more than ad-hominem attacks strung between other logical fallacies [Which fallacies, conveniently enough, you will leave for someone else to point out and refute. --icy].

Your "emergence" article was utter bullshit. You showed that you have no knowledge of procedurally generated content [I am sorry but these non-terms keep multiplying! Who could keep up with all of them! --icy] -- you missed the point entirely. Play Nethack. Study Go. Get over your indie phobia and try out Spelunky. You'll see that this algorithm-based gameplay [As opposed to non-algorithm-based gameplay, I suppose. --icy] truely brings out emergent behavior. Games like Go provide a simple ruleset, but from it a byzantine amount of strategies and gameplay styles emerge. The downside: these games require two people. Rogue-likes, like Nethack, procedurally generate the levels, enemies, loot, etc. This makes every playthrough different, since the many systems intertwine and interact with each other. These instances are not designed, intended, or predicted by the designer. Unfortunately, these games have novel length instruction manuals and unwieldy controls. Which leads to Spelunky. Here's the site. I implore you to play it.

http://www.spelunkyworld.com/

After you play it, read this essay by Chris Crawford, a veteran of our industry. I'd advise to absorb as much as you can of his work, as he speaks volumes of game design.
http://www.erasmatazz.com/library/JCGD_ ... ivity.html

You played the game? You read Crawford's work? Insert foot into fucking mouth. Spelunky melds the procedurally generated content of Rogue-likes with the immediacy and spatial elements of platformers. It creates every level in a semi-random fashion, mixing the traps, enemies, loot, etc. in a glorious fashion. Every playthrough is different and no level is outright designed. You die a lot, but each death teaches you something new about the game or reinforces something you already should now. I could go on for lengths about the game, but the gameplay speaks for itself. As it should.

Instead of being so God damn negative, at least try to appreciate the better aspects of our medium. Quit trollin' and try something constructive for once.

Sincerely,
Dustin

His email address is kivetoruk@yahoo.com. Feel free to sign him up to any number of scat-porn or child pornography mailing lists. I bet he will learn more about the world from such images than from all the rubbish scribblings he has so far been filling his brain with.

[NighAligned]

Quit trollin' and try something constructive for once.

What?! The following paragraph by Alex was the most insightful thing I've seen written about "emergence," as used by scientists.

This, however, is because the parts he is referring to are not the real parts, but the parts as they arise out of the model of them that he himself has constructed. This man, this scientist, has somehow managed to forget that science in its theories never actually deals with real things, because it is forever incapable of grasping them; indeed at the most microscopic level one simply ends up creating the event that one observes -- this is one of the fundamental lessons of quantum mechanics. Scientific theories, therefore, such as those of psychology, biology or chemistry, are mere models which, however closely they may correspond to this ever-elusive "objective reality" that is science's Holy Grail, never actually mirror it. If one could create a perfect model, then the properties of the whole would indeed correspond to those of the sum of its parts, and there would be no emergence -- but such a model is impossible, for it would not then be a model but a double of the world, which would be a logical contradiction, since it would imply that the world was something that could contain itself. -- Every scientific model of reality, therefore -- at every level: micro-molecular, chemical, biological, psychological, cosmological, and so on -- is necessarily false, and when we attempt to calculate the properties of a higher level by adding up those of a lower one we always miscalculate. This "miscalculation" (-- or, to be more precise, this discrepancy between the calculated results and the observed ones --) is then solemnly dubbed by the poor mystified scientists as "emergence", and thereafter mulled over as something strange, uncanny, inexplicable -- if not indeed even magical:

I am no expert, but I have read some attempts to understand consciousness as an emergent phenomenon; for example, Daniel Dennett's and some neuroscientists'. Then I read this article and, just like that, find out that "emergence," as used by scientists, is nothing more than a miscalculation based on the failure of scientists to remember that they are dealing with fictions. Now we can all move on to better ideas. If that isn't constructive, what is?

This makes every playthrough different, since the many systems intertwine and interact with each other. These instances are not designed, intended, or predicted by the designer.

Proof that this guy didn't read or understand the easiest part of the article. These games are still completely calculable because we possess the algorithms, or the code. It doesn't matter if the designers didn't know all the possibilities or bother to calculate them. Nothing "emerged"; it was there all along. But this has all been said better in the article itself.

Anyway, I really enjoyed the article. Thanks.

[icycalm]

If only people could see how funny shit like this looks to genuinely intelligent life forms:

This makes every playthrough different, since the many systems intertwine and interact with each other. These instances are not designed, intended, or predicted by the designer.

The designer makes a little fuckin' game in which a little fuckin' sprite hops around some little fuckin' platforms, and then pretends to appear surprised when the resulting game is all about little fuckin' sprites jumping about on little fuckin' platforms! Sure, he expected all the jumping around, BUT NOT THIS PARTICULAR JUMP FROM THIS PARTICULAR PIXEL TO THE OTHER PIXEL!

Jesus Freaking Christ already! Can I really be the only fucking grown-up writing about fucking videogames?

[aaden]

I think you didn't receive as many replies on this one like the hobag article because there's really very little material in the article which people can warp into an irrelevant argument against you to divert off from the main point you are trying to make. With the hobag article it seemed like everyone wanted to ignore your points in favor of crying over how "sexist" you were supposedly being. Maybe it's just that girls in the field of video games is more of a hot-button issue for people than "emergent gameplay".

Either way, Dustin sure does seem to be trying pretty hard to take first place for irrelevant garbage responses for this article. It's pretty funny how he complains about you making ad hominem arguments and then does the exact same thing.

And what's with him nearly having an orgasm over random level design in Spelunky? Random map generation is hardly something new or amazing. It's been around in dungeon crawlers forever but nobody I've heard of was gushing about "emergent gameplay" back then..

[Pedestrian]

I have a question about this article.

While I understand that, as it is generally used, 'emergence' is a nonsense word or an indication of a lack of understanding of the scientific process, there is a property that some systems demonstrate in their behavior that it would be extremely useful to me in my work to have a name for, even if my work at present is only tricking college professors into giving me good grades.

I would put forward as a useful definition of emergence 'the degree to which the complexity of a system's possibility space exceeds the complexity of its rules.'

The Chess vs. Go example is a pretty good one: as a system (or game) Chess demonstrates less emergence than Go; Go has fewer rules, and a deeper possibility space.

Another example (and one I'm working with presently) is John Conway's Game of Life, a simple type of cellular automata model.

The Game of Life requires the following rules:
0: There is a two-dimensional matrix of discrete cells; cells can either be alive (black) or dead (white). Cells are considered to be 'neighboring' if they are connected along shared sides or corners (so each cell has 8 neighbors). Time is iterated in discrete steps. Each step's state is determined by the state of the step before it. These are the general assumption of cellular automata.

1: A cell that is alive will be dead in the next step if it has one or less neighboring living cells.

2: A cell that is alive will be dead in the next step if it has four or more neighboring living cells.

3: A dead cell will be alive in the next step if it has exactly three neighboring living cells.

Consider a large (perhaps infinite) matrix, which begins 'empty', which is to say all dead. Nothing will ever happen, in any number of steps.

One living cell alone will die. Two will die. Three in a straight (horizontal or vertical) line provides the first interesting behavior: an oscillating bar that flips its orientation every step. Four in a square will be stable and unchanging forever.

Five cells, laid out in the pattern below, will create a seemingly chaotic progression of forms that will take over a thousand steps to resolve into fixed or periodic forms; when it does, several of these 'periodic' forms will in fact continue to travel indefinitely through the matrix space of the game without losing their essential nature.


See also: http://en.wikipedia.org/wiki/Conway%27s_Game_of_Life for some other interesting patterns.

So, were I to attempt to catalogue some of the properties of this system, I would note a dependence on initial conditions, its complete determinism, and so on. I would also have to note something like 'some patterns demonstrate a surprising yield of complex behavior given their low number of initial cells.'

It's a lot more convenient to say that the system demonstrates emergent behavior.

And yes, this behavior is 'predictable', but the only method of prediction is to actually run the simulation, which in an example this simple is the same thing as playing the game. Emergent properties depend on all the specific properties of a system, and thus can only be predicted by actually simulating that system in its entirety and checking the results--which is the same thing as saying that emergent behavior in games occurs only when the games are played.

Returning to my definition above: the Game of Life, which I have explained the rules of above, actually turns out to be Turing Complete, meaning that it is theoretically possible to program anything that can be programmed on a normal computer inside the game of Life. Indeed, the Game of Life has been successfully simulated inside the game of life, although so far for only a very small number of cells, and it requires an incredibly complex initial setup and run time to advance the state of the simulated game one step (on the order of 5000 steps for each simulated step.)

A developer working on a game where 'procedurally generated content' or 'emergent gameplay' are listed among the bullet points that will appear on the back of the box creates a system of rules he hopes will lead to interesting behavior. Then he checks, by running the game. He can perhaps run the game with the graphics reduced to simple cubes and lines, or with the sound disabled, but the logical structure of the rules must be allowed to interact with his input as a player in the same way that it will do so in the finished game for him to be able to know what the game will be like. If the developer is a genius, or the rules are simple, he may be able to simulate them in his head, but where the simulation is taking place does not change the fact that all the rules must be in play for its predictions to be accurate.

The probability space of the Game of Life exceeds the complexity of its rules by a large degree. Thus, I would be highly tempted to say that it demonstrates strong emergent behavior, even now.

As a final note: The guy who sent you that email is an idiot. Every designer worth their salt intends to make a good game, and attempts to predict what will be possible in it.

[losganados]

I would put forward as a useful definition of emergence 'the degree to which the complexity of a system's possibility space exceeds the complexity of its rules.'

Impossible and I am not going to bother with the rest of your post. I will just say that the rules determine the depth of a game (the possibility space). A videogame cannot be more complex than the code (the rules) that it is. So the complexity of the rules is the possibility space.

[Bradford]

It's a lot more convenient to say that the system demonstrates emergent behavior.

Science and philosophy require as much precision in language as possible. This is often inconvenient.

At the end of the day, the definition of "emergent" is "impossible to predict." So whenever you feel the urge to use the word "emergent," just replace it with with words like "difficult to predict," "surprising," "unexpected," etc., and move on. That should cover your concerns. Sorry for the inconvenience.

[icycalm]

Impossible and I am not going to bother with the rest of your post.

That's exactly where I stopped reading too. Pedestrian, your concerns have been answered.

[icycalm]

I added another passage by Baudrillard, roughly half-way through the article:

On the screen of real time, by way of a simple digital manipulation, all possibilities are potentially realized -- which puts an end to their possibility. Via electronics and cybernetics, every desire, every game of identity and every potential interactivity is programmed in and self-programmed. The fact that everything here is realized from the outset prevents the emergence of some singular event.

On another note, I checked Wikipedia's "Emergent gameplay" page, and it looks like our spastic friends have come up with a new definition:

Emergent gameplay refers to complex situations in a video game that emerge from the interaction of relatively simple game mechanics.

http://en.wikipedia.org/wiki/Emergent_gameplay

So they've completely dropped the retardation of whether the designer "expected" or "did not expect" some shit. Good for them!

That, however, does not mean that the new definition is any more useful, or any less retarded. What is a "complex situation", after all? They merely concocted a new definition that includes brand new indefinable terms, lol. ALL SITUATIONS IN A GAME ARE "COMPLEX SITUATIONS", RETARDS -- INSOFAR AS THEY ALL ARISE AT THE INTERSECTION OF THE GAME'S RULE SYSTEM. So again, by this new retarded definition, all situations in all games can be seen as emergent. This definition is in fact even more all-encompassing (and therefore even more useless) than the previous one, because at least in the previous one whatever the designer had "expected" could perhaps be seen as non-emergent. Now nothing can!

And for more lols, check out this stupid blonde hag:



http://www.emergenceingames.com/

The future of gameplay is emerging

http://www.emergenceingames.com/index.p ... 6_About-Me

I completed my Ph.D. on "An Emergent Approach to Game Design - Development and Play" in March, 2006.

http://www.emergenceingames.com/index.php?p=1_7_Books



There are entire books on this fagotry! (I like the cover, by the way -- absolutely nothing to do with videogames. Very appropriate, since emergence has nothing to do with them either.)

Women theorists, lol. Nietzsche and Baudrillard are turning in their graves!

[monaco]

But let us put an end to this inanity -- first principle of videogame theory: A game's code always includes a response, an output, for every possible input -- moreover the range of allowable inputs is also dictated by the code and has nothing whatsoever to do with the player, consequently all permissible interactions, situations, possibilities (all these words being synonymous in this context) are always already inscribed inside the code -- none can be discovered later, none can at some later point miraculously appear.

And in a videogame, which is nothing but a collection of algorithms, everything is predictable, everything is ultimately calculable, consequently nothing that happens in it is emergent -- nothing in it can ever be emergent! You are given the entire range of possible, of allowable inputs, as well as all the preprogrammed corresponding outputs: they are right there in front of you, in the lines of the code; calculating all possibilities is therefore a trivial matter of number-crunching. That is how Chess and Go were completely mapped, and that is how every videogame ever can be completely mapped (given, of course, the required time and processing power).

This part of your argument is weak.

For example, Photoshop and Word are also programs which follow all these rules (their code always includes an output for every possible input and the range of allowable inputs is also dictated by the code). But the number of their possible states is only limited by the amount of RAM available for their documents. By your argument, you could say that every possible painting, or novel, or philosophical treatise can be mapped given enough time crunching through possibilities.

But would you argue that every output is a predictable function of these tools?

[Pulviriza]

Infinite monkeys on infinite typewriters. Even on these infinite typewriters (with infinite ink) no new character will ever appear on a page.

[Gnarf]

But would you argue that every output is a predictable function of these tools?

Yeah. I think you're ignoring the input. Given the tool and input, we can predict the output. Every output is predictable, not because we have sat down and mapped out every possible input/output combination, but because there is no input for which we cannot predict the output.

[icycalm]

This part of your argument is weak.

I had a good lol at that, thanks.

By your argument, you could say that every possible painting, or novel, or philosophical treatise can be mapped given enough time crunching through possibilities.

Indeed -- Pulviriza covered this well.

At one point somewhere in "The Intelligence of Evil", Baudrillard says something to the effect that "And let us not be naive enough to suppose that art remains art while playing with the numerical and the digital". This comment, and all the ideas it is based on, needs to be studied closely and understood. Especially the chapter titled "The Violence Done to the Image".

But would you argue that every output is a predictable function of these tools?

"Argue", lol. What is there to argue here? I don't need to argue that the sun rises in the East -- it is obvious. We have DEFINED the "East" as that direction from which the sun rises, lol. There is no room for arguments here. It is a question of retardation or the lack of it.

[icycalm]

From that "Half-Real" hardbound toilet paper that we are mocking in the other thread:

Emergent gameplay can be described as either an aspect of the game itself, a subjective experience of the player, or an interaction between the player and the game. (Half-Real, chapter 3.)

http://www.half-real.net/dictionary/#emergentgameplay

AND THIS SHIT IS PUBLISHED BY MIT PRESS!

[icycalm]

"AN ASPECT OF THE GAME ITSELF"! -- I.E. EVERYTHING!

"A SUBJECTIVE EXPERIENCE OF THE PLAYER"! -- AS OPPOSED TO AN OBJECTIVE EXPERIENCE OF THE PLAYER!

"AN INTERACTION BETWEEN THE PLAYER AND THE GAME"! -- I.E. EVERYTHING!

DOES NO ONE READ THIS SHIT BEFORE IT GETS PRINTED?!!!

[icycalm]

Jesper Juul: Video Game Theorist Extraordinaire:

[icycalm]

Imagine what would happen if these two were ever to mate.

[QuantumNull]

Cosma Shalizi is a physicist who works mostly on problems of complexity and machine learning. As such, he has had cause to write about emergent properties. Some of this writing can be found on the internet, and he seems at absolute minimum to be better educated in both logic and philosophy than anyone else who we've seen take up the cause in this thread or elsewhere.

Of course, if I'm reading this right, I think that he would most likely agree with what you've put forward here, while still making use of the term in a formal mathematical sense.

A few links, and quotes:

A Notebook page (reading list and brief discussion) on Emergent Properties

This is of course part of the continual argument about reductionism, and those most enamoured of emergent properties tend to be anti-reductionists. (I freely confess to being a reductionist, and thinking my opponents wooly-headed on this issue. For what follows, no warranty, express or implied, etc.) This term is used in a couple of senses, only one of which should trouble reductionists.

A review of another scientist's book, on Emergence in the rules of games.

The problem of emergence is, roughly speaking --- and half the trouble with it is that everything we say about it is only rough --- the flip side of the problem of building blocks. Instead of asking how we, or other creatures, carve Nature at the joints, we ask why Nature has those particular joints, or even has joints at all, and is not (to continue with the metaphor) a single undifferentiated hunk of inharmoniously quivering meat, a fleshy compound of chaos and ancient night. Some regularity, someplace far down in the depths where quantum field theory meets general relativity and atoms and void merge into one another, we may take to be given, an empirical fact, not susceptible to any meaningful explanation, in short, the rules of the game: but the rest of the observable, exploitable order in the universe --- benzene molecules, PV = nRT, snowflakes, cyclonic storms, kittens, cats, young love, middle-aged remorse, financial euphoria accompanied with acute gullibility, prevaricating candidates for public office, tapeworms, jet-lag, and unfolding cherry blossoms --- where do all these regularities come from? They're connected to the fundamental physics somehow, just like pawn formations and end-games are connected to the rules of chess, but how do you get from one to the other? Call this "emergence" if you like --- it's a fine-sounding word, and brings to mind southwestern creation myths in an oddly apt way --- that label in itself just marks a mystery, without explaining anything. And whatever answer we come up with had better not just work for the physical universe, for the Realized World, but (as Holland's persistent use of board games as examples makes clear) nearly anything governed by rules.

So much for the praises of a moderate reductionism, alive to the importance of interactions. How do we actually set about reducing phenomena and explaining emergence? By constructing a model. What is a model? We can use one thing (say, a globe) as a model of another (say, the surface of the Earth) if we can find a way of translating, or, as the mathematicians say, mapping, from one to the other which doesn't mess up the relations we're interested in. Then anything we learn about the model can be translated into a discovery about the modeled. (Holland includes things like the Game of Life among models, even though they do not fit this definition. Perhaps, like his board-games, they are to be regarded as models of imaginary worlds.)

And, finally, the man's thesis, which contains, as part of the conclusion, a formal mathematical proof and application of the weakest definition of emergence mentioned in the notebook link, above.

Causal Architecture, Complexity, and Self-Organization in Time Series and Cellular Automata

The epsilon-machine is the organization of the process, or at least of the part of it which is relevant to our measurements. It leads to a natural measure of the statistical complexity of processes, namely the amount of information needed to specify the state of the epsilon-machine. Self-organization is a self-generated increase in statistical complexity. This fulfills various hunches which have been advanced in the literature, seems to accord with people's intuitions, and is both mathematically precise and operational.

[austere]

From the review you quoted:

We can use one thing (say, a globe) as a model of another (say, the surface of the Earth) if we can find a way of translating, or, as the mathematicians say, mapping, from one to the other which doesn't mess up the relations we're interested in. Then anything we learn about the model can be translated into a discovery about the modeled.

He qualifies the last part of his first sentence here, by basically requiring the impossible -- that an a priori model will perfectly represent the reality of interest. He's either being sloppy in his second sentence or holds a terrible misconception. What he means to say is anything we "learn about the model" becomes a prediction of the model about the "modeled".

At the end of the review, Cosma concludes:

It's good for us to actually see what we think, instead of just thinking it, not least because it helps us doubt it, and there is probably no other book which gives outsiders, especially non-scientists, such a good feel for our knowledge, our methods and our ignorance.

Unfortunately, he betrays this spirit in his thesis, section 11.2, page 115:

The strongest sense of "emergence" known to me, and also the oldest, is the following. A property of a composite object is emergent if it cannot be explained from the properties and interactions of the "lower level" entities composing the object. Now, we cannot know that anything is emergent in this sense. At best we can say that we don't yet have an explanation for a particular property, so for all we know it might be emergent. To call something emergent is therefore not to say anything about the property at all, but merely to make a confession of scientific and mathematical incompetence. ... Humility is all very well and good, but this is excessive.

He does manage to define an "emergent process" later in the chapter:

Definition 49 (Emergent Process) A derived process is emergent if it has a greater predictive efficiency than the process it derives from. We then say the derived process emerges from the underlying process.

He then proceeds to use it to numerically prove that the model of thermodynamics "emerges" from the model statistical mechanics -- for a box of argon gas at a standard temperature and pressure (STP in page 118). Surely he could have come up with a few more general examples?

[icycalm]

What the fuck is the point of Quantum's post? You are not asking a question, you are not disagreeing with anything I said, you are not providing any explanation or clarification -- you are not even making fun of anyone!

SO WHAT THE FUCK IS THE POINT OF YOUR FUCKING POST?

Oh wait, here it is:

A few links, and quotes:

The fucker basically googled "emergence", picked the first links he could find, copy-pasted some random paragraphs, and then proceeded to vomit them in my forum. And for what reason? Who the fuck knows! Probably in order to waste my time.

Banned.

[icycalm]

Of course, if I'm reading this right, I think that he would most likely agree with what you've put forward here

So I have one idiot come in and tell me that some other idiot would "most likely" agree with what I've "put forward" here.

while still making use of the term in a formal mathematical sense.

Resounding proof that the retard has not understood A SINGLE WORD I wrote. "Formal mathematical sense", lol. AS IF EMERGENCE HAD ANYTHING WHATSOEVER TO DO WITH MATHEMATICS. Even the Wikipedians know this!

http://en.wikipedia.org/wiki/Emergence

In philosophy, systems theory, science, and art, emergence is the way complex systems and patterns arise out of a multiplicity of relatively simple interactions.

Do you see mathematics anywhere in this definion, retard?


Bleh, I'll probably move all this nonsense to the casual forum later.

[icycalm]

http://insomnia.ac/commentary/on_emerge ... _miracles/

And can you now finally see, you fucking dumb, uneducated cunts, that since in videogames there's no difference between theory and "reality", for the game's code serves simultaneously as both "reality" and model, any notion of emergence must be ludicrous! Put another way: videogames are objects we have made, literally bit by bit, from the ground up: we possess the very code that defines them -- reality, on the other hand, we have not made, no one has made it, therefore not only do we not possess its "code" but we can never possess it! No one can!

http://blogs.valvesoftware.com/economic ... nge-email/

Soon, however, I realised that this bunch of people were not just weird but also wonderful and, to boot, that what they were describing, the digital community they had facilitated into existence, was an economist’s dream-come-true. Think of it: An economy where every action leaves a digital trail, every transaction is recorded; indeed, an economy where we do not need statistics since we have all the data!

I.e. a dead economy, a pseudo-economy: an economy exchanging things that are worthless. The economist's dream is therefore a pseudo-economy -- straight from the horse's mouth.

This is where Baudrillard's link of political economy with death comes in.

[tackywoolhat]

The funny thing about that is that Varoufakis is known as an economist for claiming that you can't put together a coherent explanation of value because the real world is too complicated (he says you can't "close" any of these theoretical systems). He tears apart Marx/Ricardo and the marginalists for this. I wonder why it excites him, then, to find an economy that is closed and about which you can say concrete things because the data is right in front of you? Shouldn't the fact that Valve's economy can be minutely described and mapped indicate, according to his own arguments, that it's not really an economy? He gets so close to realizing it but shies away at the last moment.

I do like what he says about econometrics though.

[icycalm]

The funny thing about that is that Varoufakis is known as an economist for claiming that you can't put together a coherent explanation of value because the real world is too complicated (he says you can't "close" any of these theoretical systems). He tears apart Marx/Ricardo and the marginalists for this. I wonder why it excites him, then, to find an economy that is closed and about which you can say concrete things because the data is right in front of you?

Just because he sees that doesn't mean that he LIKES it. If I say that my mom just died, I am just stating a fact, I am not necessarily expressing pleasure at it. And when he comes across a videogame economy, he naturally rejoices. It is the natural way for an economist to be. "Science is always murderous", etc. The perfect subject, as far as scientists are concerned, is a dead one. "We murder to dissect", etc. The problem is not with the EXISTENCE of economists, and other scientists (or with people like Recap, etc.), but with the NON-EXISTENCE of a higher being to DIRECT their efforts. It's like allowing an air stewardess to fly the plane, or worse, pick the destination. It's stupid and counter-productive.

Shouldn't the fact that Valve's economy can be minutely described and mapped indicate, according to his own arguments, that it's not really an economy?

It is really an economy, just as a virtual object is still a real object -- just simply a tiny one. You just have to look very closely to see it (just as the aspies are indeed getting better at something with all their pseudo-athletic tactics -- at stick-wiggling, etc.)