Depth vs breadth is an issue with the study of every medium, and with games particularly, there is a staggering amount of potential breadth. Coverage of detail

Depth vs breadth is an issue with the study of every medium, and with games particularly, there is a staggering amount of potential breadth. Coverage of detail results in "top 10 tips" articles, coverage of the broad strokes results in an oversimplification. My current approach to game design revolves around a three-stage process: find well-justified principles, set concrete success benchmarks based off the principles, and routinely do studies of specific elements to test the benchmarks. This is something that can be applied fractally, from the top level down into every last asset, and isn't really specific to games, but it lets me cross the boundaries between media, ask and answer detailed questions like "what form of music is most appropriate for this scene?" A question like that isn't adequately answered by personal taste, and really benefits from a principled framework starting from basic philosophy-of-truth questions about the concepts. And this goes for stuff with the software and UX, too: requirements around stability, performance, ease of use and so forth appear in the success benchmarks and enforce a rough shape for the featureset. When the game "doesn't work" despite meeting technical standards, it's usually because there's a conceptual issue where the ideas the game explores are presented incoherently, where features are added because "this feature appeared in another game and it was fun there" and the contradictions implied by the juxtaposition are ignored. When the game goes in that direction, it makes players throw up their hands and say, "I don't know what this game is about." It's also harder to make such a game, because you end up laboring to paper over those contradictions with complicated special casing.