I enjoy the post mortems in the gd mag. I have thought about doing something similar on our website. My thoughts are to let the developer decide on what to concentrate on. If they feel giving a general post mortem is sufficient then run with it. If they think focusing on a particular code is the best way to go, then that's the answer. Only the developer can tell us and that's why I read it... to get the developer's perspective.

I have been reading (and liking) the GDMag postmortems for almost ten years now.

Even though I think that it can never hurt to invest more thought in lessons learned and how to convey them best to the community, it does not feel as if past postmortem articles would have been generally lacking such information.

I like the not-so-formal requirements and style for Game Developer Magazine Postmortems. Developers themselves should decide on whether or not to focus on a particular aspect of the development process.

My perspective may be different from the one you're seeking, as I'm only an enthusiast as opposed to an industry professional, but I enjoy both the general and more specialized postmortems. The former type tend to focus on approaches to project management, it's true, but the hurdles that each project faces are often very different and so I always learn something new.

question: would it be too dense a magazine for a non-developer--that is to say, a games enthusiast--to read?

i mean, if there are long bits as on gamasutra (which i still love) about such and such programming technique, not spelled out in layman's terms, then i won't check it out, but i've always been curious.

I think a more focused Post Mortem would be a great idea. The details are where it gets really interesting. Every project has a thousand stories. The Spiderman 2, Tony Hawk Downhill Skate Wii and the Resident Evil 4 post mortems have been the most interesting and useful.

Parkbench - I'd say that some, but by no means all of the content is interesting to non-professionals - there's at least one technical article per month, for starters.

Check out the sample digital issue for a better idea:

http://gamedeveloper.texterity.com/gamedeveloper/sample/

I'd like to see more specialized PMs simply because so many of the PMs we see in the magazine cover the same ground again and again. Beyond the cliches of iteration, scoping ,etc, the rights are usually things the team gambled on successfully while the wrongs are the things the team gambled on unsuccessfully. If we can narrow the focus, maybe we can glean some deeper lessons.