The everyday blog of Richard Bartle.

RSS feeds: v0.91; v1.0 (RDF); v2.0; Atom.

Previous entry. Next entry.

1:07pm on Friday, 4th September, 2009:

What I Want to Hear about MMOs


My DiGRA talk was on the subject of what I don't want to hear about MMOs. It ended with a pointer towards what I do want to hear, but unfortunately I let my talk go on for so long that I didn't have much time for an explanation of what I meant or to answer questions. So here's a fuller explanation of what I was getting at.

There are no mathematical foundations to Game Studies. Anyone looking at Game Studies as a budding academic discipline is entitled to regard it as mere flim-flam if it doesn't have such foundations. It could be a completely ungrounded topic consisting of a web of circularities and self-references with no anchors to tie it down. If it doesn't have any underpinnings expressible in subject-neutral terms (ie. some mathematical formalism) then it's a perfectly legitimate question to ask why it doesn't. If Game Studies wants to be taken seriously, it needs to be given serious treatment. We're seeing movement in this direction from the direction of philosophy (indeed, Ian Bogost's talk immediately after mine did just that), which provides a framework that situates Game Studies in a wider context. However, games are fundamentally about the conversion of the potentialities offered by a rule set into sessions of instantiated play (causally dependent on the rules but interpretable as narrative), and this looks awfully similar to the situation with computer programs. There, we have a description of a rule set which, when executed, results in a running program. The mathematics for describing this already exists in terms of various formal logics — a means by which procedural knowledge can be described declaratively, thereby enabling it to be reasoned about as a first-class object. So why have no logicians taken an interest in providing an axiomatic sematics even for something as amenable to it as Chess? (Caveat: maybe they have, but I couldn't find anything on the net about it and none of the logicians I asked knew of any work in that area, even though they said it would probably only take them half an afternoon to create such a formal definition).

This isn't just about gaining the respect of other disciplines, though: Game Studies needs such formalisms itself. There's a reason that scientific and philosophical subjects use logics upon which to build higher structures — it helps! Game Studies has "Game Theory", which is used to describe player choice and decision-making for certain types of game, and this is useful within that context; it doesn't tell you what a game is, though, or when two examples of gameplay are the same. If people are approaching games from different directions (typically games in relationship to some "home" discipline versus games as the object of study themselves) then they need a way to ascertain that what they are talking about is or isn't the same thing. A mathematical underpinning allows them to do that. They don't necessarily need to understand the maths themselves, they just need to understand the level below the one they use. This is like saying you don't need to know how the internal combustion engine works in order to drive a car, you just need to know the controls that work it. Sure, you may be a better driver if you do know how it works, but there could be better drivers than you who don't. So long as someone knows how it works, so that it can feed into the design of the car, then that's all the driver needs. If no-one knows how it works, we could be using engines with only one cylinder that only go a short way before spluttering to a halt. With Game Studies, OK, so it seems as if we're going well, but we don't know that until we look underneath the bonnet and see if we can figure out what the engine does.

Here's an example of what I want to be able to do but currently can't.

In the Naxxramas instance of World of Warcraft there is a section where there are slimes moving across the path you need to run across. These slimes don't move all that fast, and they're in about 4 lines with regular spaces between the lines and the slimes. If you get hit by a slime, your character is deaded. So, the idea is to time your run across a line, wait in the gap between lines, then time your run across the next one and so on until you reach the other side.

This sub-game is known universally as "frogger". I think it may even say "frogger" on the map. The reason it's called this is because the mechanic is that of an old arcade game, Frogger. You can play it online here, if you don't know it.

Frogger puts you in the role of a frog trying to cross a road. Traffic comes in flows, and you need to jump betweem the vehicles in the various lanes to get to the other side. WoW has the player character instead of frogs and slimes instead of traffic, but strip away the dressing and it's basically the same game.

What I want to be able to do is show mathematically that it's the same game. I want to be able to write a description in a formal logic of the gameplay of Frogger which is the same as the one for the Naxxramas sub-game.

I also want to be able to show where they're not the same, because although it's very obvious that the two are identical at some level of abstraction, actually they are quite dissimilar at others. In Frogger, traffic comes from both directions, the gaps between vehicles is not always the same, there is only one section where you can wait in safety (between the road and a river) and you can't move diagonally. I'm useless at Frogger but find the Naxx version trivially easy. The two are clearly different — yet at some level, they're identical. I want to be able to write both down and do equivalence proofs.

Because logicians have been reluctant to visit games, frustrated designers have started to try develop their own means of describing games and gameplay (Raph Koster's grammar of gameplay is perhaps the best known). This is an uphill struggle, though, without a training in logic. I have a slight advantage in that I was actually taught denotional and operational semantics in 1979, so I can see what's possible; however, temporal logics (which would be necessary for most digital games) didn't come along until afterwards, so I'm not fully equipped. Also, I don't actually like working at this level — it's tantamount to programming in binary. I want to use the tools it delivers, I don't want to have to make the tools myself.

Anyway, that's what I was getting at. This explains why my answers to the questions that were asked after my talk may have seemed a little abrupt or evasive...

Latest entries.

Archived entries.

About this blog.

Copyright © 2009 Richard Bartle (richard@mud.co.uk).