As always, GDC was good this year. The venue shift from Moscone West/North to North/South was interesting, but worked pretty smoothly. Overcrowding didn't seem to be much of an issue, and traffic flow certainly wasn't the nightmare that it could have been, so clearly someone did their job correctly behind the scenes. I imagine it takes a lot of careful work to make sure that everyone doesn't cluster around three interesting sessions all next door to each other. At least, I assume that happened by design, although I guess that the organizers might just be really, really lucky.
Again as always, networking is the most valuable take-away from GDC. The ability to schmooze with really cool people is second to none here. The AI guys tend to band together, between the AI summit, the AI-related sessions during the main conference, and of course all the glory of the AI programmer's dinner. Over the past three GDCs for me it has been a lot of fun to catch back up with old contacts, compare notes on what's been going on all year, and generally speculate about interesting ways to model the problem of fitting a bunch of geeky AI guys into a small restaurant. I really feel like I'm growing into a position within the industry and within the AI community especially, and considering some of the people who float around in that circle, it's a pretty big honor.
As for the other take-aways from GDC... well, expect my article coverage to appear soon right here on GameDev's GDC coverage page. Don't miss it.
Also, I have officially unloaded 13 Epoch demo kits this year, which is roughly twice as many as last year. I've been picking a lot of people's brains about what they want to see in future language technology, and it's been amazingly encouraging to hear people essentially describe my own personal roadmap for Epoch. So I think there's a lot of room for success there. It certainly will be a major challenge, but if it weren't ridiculously hard, it wouldn't be nearly as much fun [grin]