Tuesday, April 6, 2010

Share and Share Alike

Making games is hard, but it is hard for reasons some may not expect. Technology isn't the biggest challenge, deadlines and budgets aren't the biggest problem. I've learned that the hardest part of making games is making games with other people.

I do not mean this in a negative way, grousing about how difficult other people are. Rather, for me at least, transitioning from academia where you are basically the only one responsible for (or, let's be honest, who even cares about) your work to an environment where two dozen people all have to execute the same idea is a big sodding change. One of the things I took away from GDC was that this is a bigger deal than a lot of people are willing to acknowledge.

Honestly, after spending all my years in university (and even at one crappy start-up) doing primarily solitary work, I never want to do so again. It's not that working with other people is unpleasant; I love it and I'd go totally barmy if I couldn't do so. But it is different and learning to recognize and even harness those differences is not necessarily obvious or easy.

I think at the end of the day, it's really about trust. You have to be able to trust that your teammates are all working toward the exact same goal (make the best game possible). Everyone has to feel some degree of ownership for the project. But at the same time, solitary ownership of ideas can be very dangerous.

The problem is ideas that are distinctly owned conflate the person and the idea. If we determine John's idea isn't viable for our game, some could see it as a reflection on John's competence. John may end up defending a clearly flawed idea, simply because he's worried if it's not accepted, it will reflect poorly upon him. It takes a mature and trusting team to simply see an idea as a hypothesis to be tested.

Matthew Burns (aka Matthew Wasteland) wrote an article for Game Developer Magazine examining the production processes at Harmonix, Treyarch and Valve. (It's in the issue that was given at GDC and is a great article, so copies abound and you should find one) Within the article, Valve's Robin Walker provided this quote, "The goal isn't to get my particular idea in the game - the goal is to be right." This is what we all should aspire to.

The question becomes how to move toward that state. One practice I've found useful is to literally swap ideas. When working on a pitch, a new feature, whatever, literally exchange your ideas with a teammate's. Have them send you their write-up and send them yours. Work in isolation for a little while and then meet. This structure has the added benefit of knowing someone else will examine and tinker with whatever you come up with, so it ought to be your best work at its most cogent.

Not only does this give you better perspective on what they are thinking, but it allows you to start discussion at the points of intersection (or even points of divergence). It separates you from your work. Being wrong is fine (I'm wrong *all* the time, believe you me), as long as we learn something from being wrong that will move us closer to being right.

And from a purely logistical standpoint, bottlenecks are terrible. Having one single person responsible for the lion's share of decisions does not work in all but the most unusual of circumstances. Even if it does work once, I'm deeply skeptical of it being sustainable. We've all heard of tales where a single, all-important stakeholder ends up looking away from a project for a period of time. Upon return, they see all kinds of changes not to their exact liking and all hell breaks loose. Maybe I'm in isolation, but I can't imagine wanting to continue working under such circumstances.

The most important quality a team can possess is trust and it's also the hardest to cultivate. Of all the discussions about team and culture I heard at GDC, this aspect echoed most frequently. I certainly can offer no sage-like wisdom that will suddenly imbue trust in a group of people. But I do believe that the importance of a culture of trust is grossly underrated. Working to separate people from ideas is a single way to move toward a more trusting culture. If you have any more experiences with building trust on a team, I'd be curious to hear them.

Labels: ,


Anonymous Anonymous said...

Great article! I couldn't agree more with this part:

"The goal isn't to get my particular idea in the game - the goal is to be right." This is what we all should aspire to.

It's often interesting to go by the saying "there's no 'I' in a team". Any idea is the team's idea. Provided that the right decisions are made for the game, everybody can feel equally good in the end!

April 6, 2010 at 12:00 PM  
Blogger Mr Durand Pierre said...

Everything in life can be learned from that picture.

April 6, 2010 at 3:18 PM  
Blogger Nels Anderson said...

@Hugo Some may say that's trite, but I think it's vital in ways most people don't acknowledge. Disagreement is fine, important even, but you have to be disagreeing for the right reasons.

@Mr. Pierre You'd think a simple CC Flickr search for "sharing" would reveal such splendour, but here we are.

April 6, 2010 at 9:25 PM  
Blogger David Carlton said...

Did you attend Kellee Santiago's and Robin Hunicke's talk? If not, you might want to dig it out of the vault - it was the best talk on this subject that I've heard in a while.

April 7, 2010 at 6:53 PM  
Blogger Nels Anderson said...

@David Haha, yup. I saw their talk during the IGS and it most directly inspired this post. Others discussed this too, but their talk was absolutely the most honest and useful of the bunch.

April 7, 2010 at 10:27 PM  
Blogger Graham said...

I strongly agree with the concept of swapping ideas. A variation on this is to rally around ideas one at a time: Have all the stakeholders in a decision "prove" one side, then the other. Afterwards, the right course of action is usually pretty obvious, and everyone is invested.

The catch is that these techniques both apply more easily to large decisions. And trust is what fills in the gap.

What's worked well for me on the past (speaking quite subjectively) is to make someone "responsible" for decisions. This requires at least a foundation of trust already, but it yields much better results that someone being "in charge" of decisions:

Basically, the individual acts as the bucket into which the team pours their ideas. And more than ideas, the team pours in their goals and ambitions.

The team has to trust that this person will make fair judgments, but if that person does their job correctly, they turn into an almost mechanical decision-making machine, with the other team members as inputs.

To me, this is what a designer should be 'doing' on a team.

April 12, 2010 at 9:00 AM  
Blogger Nels Anderson said...

@Graham Having folks defend both sides of a proposal is a really good structure, IMHO. The more people believe all ideas have some merit and the appropriate evaluation is terms of advantages and disadvantages, decisions can be made more clearly, I feel.

And yeah, having a single person (or better, several people with mildly overlapping areas of responsibility) that everyone trusts can help alleviate that vicious bottleneck that can so easily reduce progress to a crawl.

April 12, 2010 at 11:00 PM  

Post a Comment

Subscribe to Post Comments [Atom]

<< Home