an economy of games

robocraft 6.370 Final Round 6.370 Banner 6.370 Banner 2

At MIT, there is a winter fantasy month called IAP, between the fall and spring semesters, which just ended. When I was an undergrad, I ran (into the ground) this programming competition during IAP for the local ACM/IEEE chapter called 6.370 (in Tech-speak, everything has a number). For reasons I’ll explain below, I did this gig for three years. But it was only after I turned it over to David+Aaron+super_allstar_team that the contest gained this cinematic quality that still packs the largest lecture hall at MIT (10-250) for the live final round. Not to watch people play a video game (like some Halo LAN party), but to watch computer programs play against each other.

It also bring in thousands in filthy lucre from sponsor companies. In 2002, we brought in $7,000 and spent about half that; the first place prize was $500. This year, they brought in at least $18,000 assuming they kept the old sponsorship structure; the first place prize is not posted yet, but last year’s was $5,000. $5,000 to one person, for 3.5 weeks of a contest, and you are allowed to participate every year that you are a registered student. I’m not saying it’s wrong to motivate student contestants with traditional software license money. If you’re going to work for the software barons after graduation, you’d just be getting a head start. I’m not even strongly suggesting that it’s highly suspect.

Assuming you weren’t some kind of pinko, hippie, un-American non-profit, imagine how much money you could make from running a professional competition and pimping out the winners to big companies. TopCoder is still around, so I guess they’re doing pretty well. They tend to draw their problem writers from the same pool as their contest winners, and their primary clients want enterprise .NET web service beanlets… or something. Say you’re given this feedback loop and a corporate aristocracy which is 99% upper-middle class white males. You might imagine your contest format will exclude a lot of people, who go home thinking that the software industry is all about high-speed, individual problem solving or engineering generic widgets. Just to be snarky, how many women do you think are at the top of TopCoder’s ladder? How many women do you think are the richest online citizens in Second Life?

Maybe you never thought of Second Life as a contest, or TopCoder as a virtual world, but the difference between them isn’t just a label. A virtual economy is a contest for profit, a contest ladder is an economy of high scores. Linden dollars aren’t real currency, but a language arts teacher in China can still use them to turn 95 USD into 1 million in less than 3 years, and hire 17 real-life employees. It turns out you don’t need a central authority or corporate sponsorship to create skill or wealth. Linden Labs gets the wisdom of crowds to condense value out of thin air for them, but this is never presented as a viable career option to the budding student programmer. This collaborative approach simply follows hundreds of days of socialist Internet tradition.

Sidestepping what is or is not taught at universities, it was never about the money for me. I never touched 6.370’s surplus cash, because I didn’t need to. The second year was magic. The mailing list, the zephyr instance, both devs and contestants staying up all night. I’m the first to admit, there was epic suckage, watching the webcast still makes baby Jesus cry, but people actually wrote to thank me at the end. They could tell that I had a nervous breakdown and that I finished it anyway, mostly, sort of. There was a sense of community that I never got back, and I was even organizing the contest remotely from England at the time. Anyone who cuts themselves a check at the end of January never felt that.

I always wanted to make 6.370 massively multiplayer and 3D, on an open source platform, so that anyone could have their own currency, backed by virtual resources on their servers. That would be my ideal first prize, 22 virtual acres (or hectares), in perpetuity, in an economy where digital data can finally create resources that are truly scarce, not like software licenses. I definitely wouldn’t settle for $5,000. But then I thought I had to do something more rigorous, go to grad school, solve fundamental problems, etc. It will take me a long time to overcome this prejudice.


~ by Paul Pham on 5 February 2007.

2 Responses to “an economy of games”

  1. Found your blog while tag surfing.

    I’m holding a quiet beta playtest of a game – ChipWits – that is in the same genre of bot-coding. We are releasing a standalone version that is a revival of our ’84 game, then we are going to build it into an online game community.

    Drop by my blog and and take a look.

  2. There are a number of problems with this essay, and in general with my approach of thinking of the jokes first and then filling in the Serious Ideas afterwards.

    First, I don’t really have a problem with traditional software licenses in the absence of a monopoly. In ideal “invisible hand” capitalism, the consumer is rational and has perfect information, neither of which is true in real life. The details can wait for a future essay, but if I were to run a competition today, it is likely I would accept money from an indie game startup and reject funds from an old school desktop software factory.

    Second, I unfairly elevated Second Life against TopCoder, but TopCoder actually provides valuable and much-needed “accreditation” for the programming profession. My objection is against the view that companies are only looking for the rapid-fire hacking skills being tested in current competitions. You may argue that university CS majors are not being unduly biased toward any particular business model or employee stereotype. This turns out not to be the case due to (implicit) choices of sponsors and contest promotion (e.g. “You could win an internship at Foobar Corporation, where we look for the skills tested in this competition”). 6.370 gives out $5,000 for the first prize winner and job offers from sponsors. YCombinator gives out $6,000 per founder to start an entirely new company. Which one has more potential for both the winner and the economy?

    Finally, there was a snarky remark at the end about contest organizers paying themselves. I support using sponsor money to pay for ongoing development of a contest, including paying for dedicated organizers and programmers in non-profit style. I don’t approve in paying for unrelated compensation from competition funds, or for payment disproportionate to effort. Like everyone else, my morals only seem fascist and evil to the outside; to myself, I am a benevolent dictator and eminently just.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: