I was invited to speak at CodeStock 2009 this year. There were some great people here, met some new people I've only previously known on Twitter. I wasn't able to be here for most of the Friday sessions, but did go to one on Friday afternoon and several on Saturday.
Probably the most entertaining presentation I saw was Chris G. Williams on XNA Programming. Mostly because I never really looked into XNA at all. There does look like a good bit of potential there. Chris mentioned that most of the hard code gaming developers were using C/C++ instead of XNA, but that there was still great potential out there for Windows XNA development (games on the Zune, on Microsoft Surface, etc). He had some really cool demos for some of his XNA games. One in particular was a classic RPG style game that he was working on - really cool stuff is happening in that area. Kudos to Chris for his work in this area.
It kind of makes me think that although I think I do a good bit of "sexy" code development (thread safety, performance work, etc) it is certainly apparent that people are more interested in the new "sexy" code tools and programming frameworks. Performance is still a huge concern, but I think many people kind of ignore it - the idea is that you can throw hardware at that problem. Seems silly to me - you eventually have a point of diminishing returns there with the addition of new functionality and features. Maybe developers do think more about it that I am giving credit for, but it just seems like there is not as much of a concern there (unless you work on Facebook or Twitter :)