So I attended a variety of workshops. With a heavy emphasis on indie game dev or "breaking into the game dev market"
Got to meet a bunch of different game developers from small indie (2-3 man operations) to even got to meet and talk with the President of Epic games(i.e. unreal/gears of war), also some IGF people there, and a whole huge variety of others. Oh yeah, Richard Garriot(aka Lord British from Ultima Online and _______ was there too.
They talked about everything from world design, story design, programming languages, and even minecraft.
Although there was a lot of topics covered. I will break it down into the highlights.
First things first, breaking into the game dev market.(working for others)
MAKE and finish a game and have a website where you can get it.
They even emphasized that if you download game maker 2000 and make a cloud watering flower games in 30 minutes. You are already ahead of 95% of the applicants most all of them receive.
Regardless of it being a sucky game or not, you have proven that you have seen and experienced a game from start to finish. Even the website could just be 3-5 pages. Even joe shmoe can make a website nowadays. If you don't have one game and a website, your less likely to be in the second stage of resume filtering.
If you can make more then 1 game, go beyond it, make one of different genres(i.e. shoot em up, a platformer, a rpg, a card game, a point and click, a 1 button game.
In addition, don't restrict yourself to 1 programming langauge, they emphsised, make 1 in gamemaker, make 1 in valve's source, make in 1 c++, another in java, and so forth.
2 programming langauges and 1 middleware, shows you are capable of learning and working in different environments. Very important.
Don't bother during interviews or resumes or CVs to mention your "great ideas about game design and development" unless you have some years and a few games under your belt, more than likely they don't want to hear it.
Make sure if you really want to work in game dev, that if you are applying for jobs outside of your area, that you emphasis that you will move yourself if hired. Many companies do not want to pay for moving intro or lower level people and will toss you to the bottom of the pile if you aren't already local. Unless of course you live in the areas that there are a lot of game dev studios. Canada, europe, mid west and west coast US.
they also emphasized the importance on having worked on at least 1 mod for an existing game(if you managed to get the 1 game on your website). Once again, it can be a sucky mod, but it proves you can learn and interface with different APIs and whatnot.
Also even a super simple mobile app in official marketplaces are a huge ++, even if the company doesn't do any mobile games(currently).
Custom CV letters and setting your goals realistically.
To be honest, if you have less than a few years experience, you are not going to get hired at valve or epic games. Unless you have proven that you can really really bring something massive and unique to the table. (i.e. portal dev team)
Get a few years at small studio, and write your CV targeted at them, boiler plates mass sent CV's are easy to spot.
World building. Basically it came down to, constantly experiment and try it out. If it doesn't work and its not fun, refine it. Don't spent too much time thinking about if its good or not, just keep making things. Moving forward.
Although heard elsewhere, Find a crappy game on the internet and try and make a clone of it, making it better than what it was, fixing what made it crappy with your own spin and style on it.
Richard Garriot was meh, enough said.
President of epic games, talked a little about how big of a PC gamer he is and not a 360 gamer
How he is personally disappointed in the whole "360/PC ports" especially of GoW. He mentioned that games sold on 360 don't do well on PC and vice versa and that they are going to avoid making ports in the future. It doesn't sell and all it does is piss off the gamers/fans.
In all of it, they mentioned Java once. They mentioned C# once or twice, C++ about 2-3 times. But surprisingly, most of the time, they basically seemed to imply the language itself doesn't matter much and you could write a game in whatever, especially middleware.
will add more later when I get home and find my notes...