 |
|
|
Today we resume our exclusive two-part interview on games research, the new frontier of gaming! We look at major breakthroughs, some experimental games, and how a games research lab works. Read Part One here. We speak to the heads of the Singapore-MIT GAMBIT Game Lab: Professors Philip Tan (US Executive Director; project manager with Media Development Authority of Singapore) and William Uricchio (Lead Principal Investigator; Director of MIT Comparative Media Studies)...
What are some of GAMBIT’s major breakthroughs so far?
I’m hesitant to identify any one of them as “major” because of the range of different fields we address. We’ve made a Wii Remote game for the blind. We have a whole series of fun educational games that deal with really difficult math and science concepts. Some of our games demonstrate technologies that typically take animators hours or days to accomplish, and we get good-enough results in seconds. A number of our games gather data from gamers as they’re played. We have games that tackle topics like addiction and oppression, or implement new approaches to game dialogue or artificial intelligence. Which one is “major”? It heavily depends on which field you’re coming from. I think our biggest accomplishment is developing more than 6 really polished games each year, and doing that over miniscule time frames (as short as 9 weeks) to tackle really tough problems. We’re also very serious about maintaining a culture that discourages ‘crunch’, based on fairly convincing research that productivity actually drops when you work more than 40 hours a week.
The other big change was in the academic mindset, I believe. Some researchers think it’s a waste of time to make software that looks good, runs stably, is easy to distribute, and is rigorously tested on non-academics. However, because we can do it so quickly, the faculty working with GAMBIT understand that we can have our cake (solving tough problems) and eat it too (easy to run and engaging to play). This means more people are exposed to our ideas, particularly in the industry. It’s a tough process: every idea has to be scrutinized under the harsh light of practicality. However, we’ve ended every year with even more new proposals on the table. So researchers are definitely seeing the GAMBIT style of game research development as an upside. You have a policy of avoiding ‘crunch’ deadlines? How do you manage that when even big game companies cannot avoid it? Any professional game developer must learn how to eliminate all the ‘nice to have’ features as early as possible so that there is enough time to really polish the ‘must have’ features. People don't develop this skills if their only answer is to work longer hours. So we teach people that overtime is a signal that something has gone wrong. Perhaps there was an estimation error or problem with project management. Perhaps an unexpected crisis arose in the middle of production. Perhaps people were goofing off… but this comes up less often than you expect, because folks in the game industry take great personal pride in the quality of their work. Can you give examples of projects that have been approved or completed under the ‘harsh light of practicality’ criterion? Just about every project we have begins as a vaguely-defined and over-scoped idea. Take AudiOdyssey in 2008, for instance. The initial idea was to have a multiplayer game where lots of people could competitively play online, regardless of whether they were visually handicapped. In fact, you weren't even supposed to know if your opponent was sight-impaired. This is a huge, well-intentioned idea, but the 9-week deadline was unreasonable.
After several rounds of discussions and feature-cutting, we found an achievable core: to demonstrate that both blind and sighted players could play the same videogame with equal challenge, and have fun doing it. The game was a simple rhythm-matching exercise, but you needed really good timing and hearing to do well. And gamers could still play together by taking turns at the Wii remote.
So by trimming down our scope drastically, we ended up with a game that people can actually download and play, as opposed to a spiraling project of doom that we would probably still be working on to this day. What career prospects are there for someone who specializes in games research? Most folks proceeding along the PhD route are heading towards an academic career. There’s a lot of attrition along the way, though. There are more people trying to become a university professor than there are positions available. For some folks, though, it’s the lifestyle most suited for them. (And it does take over your life) Anyone who regularly develops games, whether in industry or academia, has the option of joining a game company. It depends greatly on the specific company, though. Huge MNCs probably prefer to hire entry-level graduates from universities, particularly those who have interned with them before. I would imagine that there are more entry points for an academic to join a developer than a game publisher. Ultimately, this is highly contingent in your ability to network and know people. There’s a small trend of game researchers spinning off their own companies. It makes sense, if you think about it – they’re in research because they want to do something different, which implies that there isn’t a company already doing it. When a scholar joins an existing company, it’s often because the company is trying to expand its capabilities in the same direction as that person’s research. Small start-ups are attractive to PhDs, as they are most in need of new, game-changing ideas.
There are also industrial research institutions. Larger companies such as Microsoft and Adobe have their own research divisions that regularly hire academic researchers. Defence and technology research labs are often interested in folks who understand digital games and can come up with new applications for game technologies and design principles. Such careers are largely defined by one’s ability to invent and patent.
For someone aspiring to go into games research, where is the starting point? Think about what aspect of games you want to improve, then which institute you want to join. For GAMBIT, we have both research and development entry points, so the qualities and qualifications would differ. For development, it’s not dramatically different from any other job application in the game industry. Have a resume and a portfolio that demonstrates that you’re good in your field, be it programming, illustration, animation, music, sound effects, team management, or game design. Show that you have an analytical eye for detail and the ability to communicate your ideas.
One way to get a taste of game research is to join us as a graduate student. Not all our researchers have PhDs, but those who don’t were specifically selected because of their experience in the game industry. In such cases, we look very closely at their ability to teach, to understand and engage in academic concepts, and the problems they have solved before. It helps if you’re the kind of person who thinks that the game industry could be more ambitious. Those who have PhDs come from all the fields I mentioned previously: education, computer science, comparative literature, and so on. Most importantly, we look at your body of work as a graduate student. Are you actually interested in games? What is your current research focus, and what would you teach? Maturity and professionalism are highly valued – demonstrate that you can pace yourself. If we have to work with you day after day, we don’t really want prima donnas.
But after all, many academics are notorious for their personalities!
Note that it's important that academics keep the privilege to hold unpopular views, as long as they are willing to defend them in public. If experts withhold their opinions because they risk losing popularity, then they will contribute little to their field. However, if you're into game development, you are choosing to work in an environment where you are surrounded by people of vastly different skills and persuasions. GAMBIT is structured along these lines: we're an interdisciplinary lab, and we believe that better solutions can be found by having people with different expertise work with each other. So we prefer to have people we're happy to see and talk with every day. It's not the one and only way to conduct game research… but that's just the way we roll. Here’s a bunch if important links if you want to learn more!
|
Comments
RSS feed for comments to this post.