Translation studies, some notes on a semi-formal design approach

At this point in my career(?) I’ve made several games that have focused on the idea of translating games. From one game form into another game form, as with Indie Bungle 2 (indie darlings → Breakout), Sibilant Snakelikes (popular games → Snake), and in an instance or two in BREAKSOUT and Get X Avoid Y. From a concept to a game form, as with SNAKISMS (philosophical isms → Snake) and again perhaps some of the PONGS and BREAKSOUT games. From another medium into a game form, as with Let’s Play: The Shining (movie → Atari-esque game) and The Artist Is Present (performance art → Sierra-esque game). There are other examples and it’s clearly been a preoccupation over the years. Indeed, now I’m working on v r 4 which translates between game engines (Unity → Bitsy) and Chogue (with Jonathan Lessard, which sort of translates from chess and Rogue into a hybrid of the two).

Lately I’ve been finding this formal approach of translation particularly helpful as a way of attempting to frame more scholarly/serious/rigorous accounts of game design and development as creative processes, in conjunction with my work on the process documentation approach we’re now calling MDMA (Method for Design Materialisation and Analysis, ha ha). One of the most difficult things about trying to actually talk about design is that it’s so ephemeral much of the time. Even with the best will in the world and the determination to pause and reflect on your design work in the moment as you make decisions, it can be hard to think of how to even framewhat you’re doing, and thus hard to get words out. The most important thing in that context is to actually know what you’re trying to make, for which you can refer to design documents, artist statements, or similar. But even then it can be tricky to make the connections between some specific design decision and the high level statement of purpose.

Translation as a kind of formal method of game design feels very interesting to me because it gets at those questions of moment-to-moment design. In particular, because your focus becomes the act of translating from one system to another system, one language to another language, you’re forced to think deeply about how those systems are constructed, what their important features are, how they work to express their ideas, etc. This is especially true when, as is the case in pretty much everything I’ve done, the target system (the game form you’re translating into) is simpler/cruder/smaller than the source system (the game/film/literary/artistic/etc. form you’re translating from). And it’s not just that you’re therefore making decisions about how to simplify or parody or awkwardly spell out in text, but rather that you’re forced to ask fundamental questions about what the really important ideas are in the source system, and what the really fundamental expressive affordances are of the target system.

By way of a small example of this, consider v r 4 which I’m actively working on now (and so, bear in mind, I haven’t solved all these problems of translation!). v r 4involves translating my earlier 3D Unity games v r 1, v r 2, and v r 3 into the very simple and charming 2D engine Bitsy. The previous v r games were, to some extent, about 3D and about Unity, so the translation process needs to take this into account.

When I began work on the game, I found myself thinking about the challenge almost exclusively in visual representation terms - “how can I draw the 3D worlds of those games into the low-resolution, tile-based ‘rooms’ of Bitsy?” To some extent, this would reproduce the games. You would be able to walk amongst the plinths of v r 2, say, and they’d just be 2D representations of the plinths inside a bird’s eye view of the buildings they exist in. Even this immediately leads to semi-interesting/fun questions: Bitsy has a square screen, but the buildings are long, so do they have to be split in two across two rooms? Is that problematic in terms of not being able to see the full space? Can we equate the occlusion involved in a 3D engine, which means you can’t see all the plinths at once anyway, with the occlusion of a split space? Etc.

I could probably come up with visual representations of the games, but thinking about the games purely visually more or less entirely misses the point of those source systems, which are perhaps more ‘ontological’ than anything else: they concern the nature of the materials of Unity and virtual spaces. They about asking what it means for a virtual object to really exist (e.g. the invisible objects in v r 2) or what the status of a ‘natural’ element like water is in the context of its highly technical and variable implementation (in v r 3). So if I’m going to make a translation, the Bitsy version has to at least try to get at those fundamental questions rather than skate over the visual surface. (And in fact this relates back to needing to have an artist/research statement or manifesto associated with your work - it’s really thatwhich must be translated, along with visual and other elements.)

Thus in v r 4 I’m preoccupied with, for instance, the question of what it means to have invisible objects in plinths in a Bitsy world. I’ve had to run small experiments with the affordance of the engine to find out whether you can hide an item tile (an object you can pick up under normal circumstances) inside a wall tile (a solid object that could serve as a plinth): you can’t as far as I can tell, though last night I had dreams of perhaps leveraging alpha values to deal with this, but Bitsy’s colour-picked is slightly broken right now so I can’t check. If you can’t hide one element inside another, how do you represent the core idea of v r 2 of having elements that both are and are not present? My best thought about this is that perhaps I create wall tiles that represent different objects that will be inside the plinths while I make the game, so they would all be totally visible and distinguishable. Then at the end of the game, before release, I paint in the tiles so that they all become white squares and thus plinths - effectively building the plinth around the object. Now, that would be indistinguishable from having just made a white square in the first place, but there’s a sense in which the pixel form of the object is “still there” (those pixels remain solid), it’s just all the other pixels are now solid too. I’m kind of pleased with this processual answer to the question for now.

Then there’s the question of what objects would go in the plinths, since I’m not using Unity and it probably doesn’t make sense to try to reproduce Unity’s categories of objects… and on and on.

My key point here, I think, is that this process of translation surfaces highly explicit and specific opportunities to think and write deeply about the underlying nature of both the source object (whether a game, a film, a performance, or something else) and the target game mechanics, engine, etc. It provides a very well-defined formal constraint at a high level (translation), but leaves all the real creative work open, so that despite being formal it doesn’t feel (to me) like an overly restrictive or deadening pursuit.

I think I’ve now greatly outstayed my welcome. Back to the translation mines for me! Ciao for now.