Tuesday, April 22, 2008

The Paradox of Student Failure

Reading over my notes from GDC, I just realized that I commonly hear two pieces of advice for teaching:
  • Encourage students to fail early and often. Being in school is the one time where you can do this without losing millions of publisher dollars in the process. BUT,
  • Punish students harshly for failure. It's a tough industry, and classes should reflect that.

These aren't necessarily mutually exclusive, although it seems like it at first glance. The former is primarily concerned with taking creative risks: trying forms of gameplay that have never been done before. The latter mostly involves setting and achieving reasonable goals: controlling the scope of a project, keeping to a schedule and meeting deadlines.

However, the two viewpoints collide when you're teaching a studio class where the output is a complete game -- if the students try hard, but end up making a game that is just not fun or interesting (in spite of their efforts). As a teacher, do you grade them harshly, because a comparable professional project would mean that their studio would be out of business and they'd all be looking for new work? Or do you grade them generously for their ability to try hard, stick with a process and complete the project? Either way would seem to send the wrong message.


josh g. said...

That is a tricky question. There probably isn't a single Best Answer, but a couple of possible solutions came to mind.

1) Create a marking rubric that reflects both sides. Give half the grade for project management concerns, and another half for creative achievement. It's a thought, anyway.

2) How often will a game that's fundamentally un-fun survive to the end of a repeated rapid iteration design/playtest cycle? If the course is meant to reflect industry-style criteria, maybe games should meet playability and fun standards in early prototypes by the halfway point of the course. Have them self-evaluate honestly, or use peer evaluation, and if something is failing creatively at that point, cut it. Don't let them continue working on it, but force a hard reset where they now have to come up with a creative project that fits in the bounds of the second half of the course. If the second-attempt project turns out to be great (for a game made in half the time as the others), then they could still get a good mark out of the deal.

I'm not sure if this solution really makes the marking a lot easier, but it does create a stronger sense of "fail early".

Chris Dodson said...

Good question. I think the beginning students should be punished less for failure and experimentation, but the seasoned students have to start learning that a bad game is (to a large degree), failure. Though it seems to me that through the design process they ought to be testing and iterating the game, so in a sense, you are also grading them on how well they are paying attention to player feedback and modifying game play based upon it.

I personally like grading based upon a set of factors and averaging them. Really this comes down to grading based upon game play, aesthetics and level of effort(expressed as detail). Isn't this sort of how the customer judges a game? In fact, we might use a rating system just like a game reviewer would use (great graphics, awesome game world but the game was just no fun, etc).

This discussion reminds me of Ernest Adam's talk at GDC: http://www.gamasutra.com/gdc2008/index.php?id=17446

Ian Schreiber said...

@Josh: Good point about forcing students to start over from scratch if I see the game shaping up to not be fun. This is essentially what Nintendo does, as I understand. However, I wonder how demoralizing it would be to take a class that I've given creative freedom to, and then take that away by forcing them to throw away something that they're surely attached to.

@Chris: Excellent point, and I remember Ernest talking about this, where an early project is different from a late one. In my case, the difficulty is that I'm teaching at a two-year school where there's only time for one project, and no guarantee that they'll ever have another one.

Anonymous said...

Naturally, you would expect more from your senior class than you would your freshman class.

I don't think there is anything wrong with providing 'harsh' feedback on their work. These people are going to be dealing with, possibly, much worse when they get into the industry. They need to learn how to take flack as much as learning out to identify/make a fun game. Don't sugar coat it for the sake. I would personally rather have someone tell me that my game isn't fun and giving me some ideas that could possibly make it fun, than someone not saying anything and leaving me blind.

"Hate me now, Love me later"

As long as your respective and show the willingness that you are really trying to help, I'm sure they'll admire it.

You only have 2 years, and that time moves fast.

I'm pretty sure a common phrase I heard at the GDC was, "Start Early."