More like: It's a tradition in Game Developer Magazine to solicit Postmortems from game developers. The articles on Gamasutra are from old GDM issues. I have a feeling that some of the games listed wouldn't have postmortems at all if GDM did not provide this free publicity.
It's interesting that the game industry is one of the few places where it is accepted (and even desirable) to publicly list the things that went wrong when developing a product.
Shrink-wrap software games are unique in that way: By the time a list of all the bugs, glitches and development snafus is published, the primary revenue for the product has already been earned.
When any other piece of software would be worried about its v2.0 reputation, a game will be sitting forlorn on the discount remainder rack, discarded by its creator.
My (very limited) experience in the game development industry was that it was half way between academia and traditional software companies. I think what drives this is that the problems are much harder, and there are so many different types of problems, that developers from different companies are more willing to share ideas and experiences with each other simply as a survival mechanism.
Many people consider "writing a postmortem of the product" part of the game development process, even though not all those do. In any case, I don't think it's a GDM-only phenomenom.
Sure, but publishing them (and the particular format they use) is something Game Developer has advanced more than anyone else in games. It's become their hallmark.
It's interesting that the game industry is one of the few places where it is accepted (and even desirable) to publicly list the things that went wrong when developing a product.