I don't know how widely known this is, or if they still subscribe to this policy, but Electronic Arts used to (and perhaps still do) have a kind of Zen mantra that they applied to all of their game designs. They call it "The X". The idea is that they come up with a single sentence that describes the Unique Selling Point for any given game. They do this for every game - every throwaway film license game, every iteration of Fifa or Madden, the lot. The X might be character-based, like "Be James Bond in his world of international espionage", or based around gameplay elements like "Battle the forces of Hell with the power of your Rock", or "Parkour across rooftops to overthrow the Authorities", or something. I made these examples up, but I daresay they're not a million miles away from what might have been actual X's for games. Depending on how imaginitive the marketing department are feeling, the X might even end up written unchanged on the back of the box.
The idea behind The X is to provide a common reference point for EA's vast teams of people with all their disparate skills and viewpoints. It's not just the dev team that are made aware of it, but the publishing and marketing folks as well, from the lowliest level scripter right up to the head honchos of the company. It helps to keep a team focussed towards a single goal, and it informs decisions - game design elements which support the X tend to get approved and worked on, whereas extraneous stuff that doesn't fit the X is discarded.
Now, someone who is not a fan of EA games, who heard about this and was feeling uncharitable might argue that The X contributes to the feeling of blandness that comes with many of EA's titles. Their games tend to be incredibly polished and slick, but a bit shallow. They can feel a bit like one-trick ponies, with a single gameplay "hook" that gets pushed in your face, and which may or may not be actually fun, and meanwhile everything else in the game can feel a bit cookie-cutter and tame. Worse still, EA occasionally act as a publisher for exquisite titles developed by other studios, but their normally brilliant marketing team fall flat when they have no X with which to sell the game. Oddworld: Stranger's Wrath fell foul of this, and the Oddworld Inhabitants team basically gave up making games shortly after, when the game failed to sell well.
Some games these days seem to wide and complex for an X. The likes of Fallout 3, or Grand Theft Auto IV would be considerably thinner experiences if the developers and publishers had chosen a single thematic or gameplay element and pushed it forward to the exclusion of everything else.
So, it's easy to be down on the concept of The X. It clearly came into being as a way to keep a behemoth of a company focussed when trying to get huge teams to crank out games, so how applicable could it be to apply this approach to tiny indie games?
As it turns out, it seems very applicable, but for different reasons. Primarily it works because indie games are incredibly restricted in terms of manpower, time and budget, and as such need to get right down to presenting the bare bones of what's cool about their game without wasting time on any of the fluff around the edges. Introversion's Defcon is one of my favourite examples of this, and feeds back perhaps to the earlier discussion about "purity" of design - the game is about global thermonuclear war (to quote Wargames, "the only way to win is not to play"), and the game's tagline of "Everybody Dies" sums up the game rather nicely. Although Braid is thematically quite deep and ambiguous, the gameplay has precisely one hook, which could be expressed something like "A Mario game where you can reverse time at will". The Experimental Gameplay Project take this idea even further: For every game, The X is whatever new toy or method of interaction the game demonstrates, and all that's added on top is just enough goals to give that X context and meaning, to turn it from a toy into a game. Experimental Gameplay, of course, spawned Tower of Goo ("Build the highest tower you can from squishy goo balls"), which then spawned the utterly excellent World of Goo. Contrast this with the foolish pitches you hear from wannabe indie devs who want to make "An MMO where you can go anywhere and do anything, and there are 10 character classes and you can customise all your equipment and it's set on Planet Zzryg'Yon...". These ideas have no focus, no single defining feature to underpin the development, and as such are far too ambitious and scrappy to ever see the light of day.
Our game will have an X. It might not be in the form of a specific sentence (although we might formulate one if we feel the need), but the paint physics is what will define this game, at least in terms of the gameplay. A puzzle/platformer is hardly a unique concept in and of itself, but that's okay because I know how to put those sorts of games together, what makes them tick. The platform/puzzler isn't the X. The X is in what kinds of things become possible when your main platform/puzzle element is paint, and I think that the kinds of scenarios we'll be able to present to players is what makes the R&D time developing the paint physics worthwhile.