Came across this while googling "philosophy of games". It really has nothing to do with that, but I thought it well worth linking anyway. The whole thing is extremely correct and extremely well-written, with the exception of a few lines towards the end:
Metanet Software is devoted to producing high quality video games. Our definition of a good video game is very simple: a good game is one which is fun to play.
A game which tries to make people think it is fun will never be as successful as a game which actually is fun, because in general, most people know when they're having fun. We just tend to work that way.
This is just simpleminded nonsense. No one has yet managed to come up with a "theory of fun", without which all the above simply makes no sense. But the rest is awesome.
Marketing should simply take the finished product and market it: make it appealing to consumers. Changing the design of the product in order to make it appeal to consumers is basically cheating: the purpose of the product is it's function (in this case, to act as a form of entertainment media), and unless part of that function encompasses consumer appeal (which is the case with some products, such as cola, where the product serves no other real useful purpose) then offloading the marketer's job onto the engineers is a misguided practise.
These beliefs are not based on idealism or any concept of "artistic integrity", but simply from a practical evaluation of video games: there is a statistical correlation between games considered great, and games whose design stemmed only from engineering and aesthetic concerns, and which didn't consider secondary issues such as consumer appeal.
Reading this article after all those pseudo-philosophical abstracts was such a huge relief I almost feel like kissing whoever wrote it. So clear, concise and correct, it is a pleasure to read again and again.
