Convergence of Game and Software Design Methodology?

I read in the nimrods blog about this article by Jacob Davenport, designer (or co- designer) of some great Icehouse games like Pikemen and Gnostica.

Few topics have garnered such praise and criticism as Agile / Extreme programming (XP) over the past 4 years. Its proponents argue that XP is a deliberate and disciplined approach to software development, while its opponents often claim that it is nothing more than a chaotic, shoot-from-the-hip approach that eschews formal requirements analysis and design. The Truth is, when applied to certain classes of projects, XP is a very attractive, effective approach that can be very successful.

Jacob's juxtaposition of XP on top of game design is interesting, though I wonder what point he is trying to make. I'm not a game designer, but the two seem so very different to me. Software methodologies exist for one primary reason - to reduce the defects (or, to put it a different way, increase the quality) of a software product. Usually this means bringing some formality to how we figure out what the problem is (requirements), how we plan to solve the problem (design), solving the problem (programming), and verifying that we solved it (testing).

Game design isn't so much about solving a problem as it is producing a system for entertainment - an act of creativity. I'm not sure that any sort of methodology would help me become a game designer. But I will concede that Jacob's ideas make a lot of sense around the area of game and rule refinement. He has some nice things to say about refactoring and testing that sound applicable to game design.