First off, it was a rather disappointing experience. There were five slots for seminars, and most included only two choices for what to see. The choices, on the whole, were nothing I hadn't seen before.
I ended up bailing early, because the Origin presentation on "Managing Online presence for multi-player persistent something-or-other" has been given a dozen times at a dozen other conferences I'd attended, and the only other alternative available was how to implement realistic physics for open-loop dynamic modeling (translation: how to make Lara Croft's ponytail bounce). All in all, I'd gotten much better for half the price at the IGDN conference in Plano last year. From now on, I'm gonna hold off on GDC road-trips unless there's something I've really got to see.
Awright, on to the substance. . .
1999 game development is gonna be remembered in two ways:
(Wall Street Journal version) The beginning of the end of the honeymoon period for game development.
(Michael Swaine version) The year that the game development industry started to get bitch-slapped into submission.
While watching the seminars, I noticed something. Of the three non-technical seminars I attended, two were by people who's companies had recently gone bust, and the third was by a company that's currently in such a financial free-fall that they almost certainly will be bust this time next year.
The fact that the companies had died was not surprising. After all, companies go bust all the time. What was amazing was that nobody seems to mind.
For a moment, think of a civil engineering consulting business, like my wife's. Imagine if someone entrusted my wife to design a site-plan for a million-dollar building. Instead of being diligent about designing the site, she just stamped whatever crossed her desk and told the city engineers and managers what they wanted to hear. When the building was 75% complete, somebody noticed that the building was being constructed in a flood-plain and could not be legally occupied. Wanna know what would happen? First off, my wife would be sued by the developer, construction company, and city. Nobody in the state would ever hire her again. When the lawsuits were settled after we declared bankruptcy, she would probably have to take a job outside civil engineering, because no city, construction company, or developer would submit a project to her.
Contrast this to what happened to the unnamed game development company (let's call them StartupCo). A large publisher entrusted a developer with a large amount of funding to build a top-shelf game. This developer hired a staff of engineers and artists, made a skeletal design, and started to write a game with a minimum of plan and no project management. After a year, the parent company noticed that there is very little to see and not even a date as to when they will see something completed. The funding was cut off, scuttling StartupCo and leaving the publisher with nothing to show for the investment but a ZIP-disk of half-finished code. What happened to the developer, you ask? Well, another development company (let's call them StupidCo) immediately hired him to manage their top-shelf project. Six months later, the developer gives a seminar on "how I learned a valuable lesson, and it only cost my publisher a million dollars." at the GDC.
This has led me to make another "commandment". This one's gonna have to go near the top of the list if I ever compile these into an article, because it's very important. Maybe the most important. . .
Game development is not different!
That's right. Game development is not different from developing financial, database, or text-handling applications. Anyone who tells you otherwise is being too lazy to learn or is being dishonest with himself. Game development requires development plans, design documents, project management, schedules, resource management, design reviews, and all that other stodgy suit-stuff that you thought was irrelevant because you're gonna write games.
Once again, the problem is that developing games is the most seductive part of programming. No high-schooler has grand visions of writing the next great financial report generator, and the industry is plagued by folks who think that because they know how to optimize fill-rate, that they're ready to release the Next Big Game. Trivial things like how to write a coherent development plan, how to manage a project, how to write a design document that doesn't become obsolete immediately, and how to schedule are viewed as being much less important than the ability to write code and pitch your grand vision to a publisher.
Game development is only art if you don't plan to make any money. If you've got a publisher funding your product or you plan to pitch your product to a publisher, then you're not an artist. The era of "Hey Mr. Publisher. Gimme a million dollars and I'll give you a great game sometime next year!" died in 1999. Learn to live without it.
The funniest anecdote of the entire conference came from a project-manager at another unnamed company (let's call 'em FreefallCo) which is currently so far in debt that their next three products are gonna have to outsell Myst to keep 'em off the auction block. At one point, he rattled off a long list of publications that had interviewed him in the past year as an example of how "there is no bad press". The only thing I could think of was "the only reason they are interviewing you is because you are part of what is shaping up to be the gaming market's equivalent of Heaven's Gate."
Sorry, guy, but there is indeed bad press. Reporters are interviewing you for the same reason that people slow down near a traffic accident.