Today’s talks:

10:00 am: How to Manage an Exploratory Development Process, by Kellee Santiago and Robin Hunicke.

I heard Robin speak in the microtalks last year; she’s at thatgamecompany now. Good talk, and surprisingly moving: they started off by talking about how, at E3 2008, they debuted Flower, the press and their publisher loved it, yet the team’s morale was at an all-time low! Their development process was putting all sorts of stresses on them, which they needed to fix lest the company implode soon. (Which, I should emphasize, isn’t anything particularly negative about their company, just typical game development culture.)

Some of the key points for how to build a sustainable development process, so they can keep making the games they love indefinitely instead of burning out after a few years:

  • They needed to distinguish between commitments and estimates.

Both are important; but, if you don’t hit an estimate, that may be a learning experience (or may not, variability is natural, especially in creative work), but certainly isn’t anything to beat yourself up over (or for others to beat you up over), whether consciously or subconsciously.

Also, in an exploratory phase in your process, who knows how long it will take to find the game’s real core mechanic. (Shades of Iwata’s keynote last year.) So estimates there have a particularly large margin of error. Be confident that you still have a vision for the game inside of you; it will come out eventually.

  • Planning and feedback is important.

They have a big board sketching out milestones over the entire lifetime of the project; on a smaller level, they have two-week sprints, with standups three days a week. Their sprint goals aren’t as rigid as some teams do them, but it’s important to have the task cards up there, as a point for discussion.

Which leads to what I saw as their main takeaway:

  • Have conversations; don’t shy away for them, especially if they’re difficult.

The point of having task cards is so that you can see and talk about them if they’re not moving as you expect: are they harder than expected, are they unpleasant, are they simply something that nobody cares about? Or, if your explorations suggest that expected mechanics won’t work or that you’ll need more exploration than expected, don’t hide, even though you’re really nervous: talk to your publishers, talk to your marketers. And don’t just give the bad news: talk about your vision, talk about why you think this will make the game better. More broadly: remember that conversations aren’t part of a zero-sum game: they’re all on the same team!

One other amusing bit from the talk: we call a task easy in one of three cases: 1) we really want it to happen; 2) we have no idea what it entails; or 3) we don’t have to do it ourselves!

And, finally, the summary:

  • Be honest.
  • Wandering is okay.
  • On every project.
  • Until you die.

11:15 am: Kids and Parents Playing Together Online: the Next Frontier of Casual Gaming, by Jesse Schell.

Lots of good points here, starting from the basic point that you have to decide to design for this, not just (for example) build a kids’ game and hope that the parents will like it. Looking through my notes, I don’t see any huge insights that I’ll be able to provide beyond what you’d get from his slides, so I’ll just embed them here.

11:45 am: Lessons Learned building Moshi Monsters to 15m Users, by Michael Acton Smith.

Moshi Monsters was done by a company that had just burned through most of their venture funding, on a real-world cross-media game PerplexCity that sounded fascinating but wasn’t profitable. So they didn’t have much margin of error. Failure was still okay, but they needed to fail fast: which they started off doing, by trying a Webkinz model of physical goods linking to an online world. They stopped doing that, and immediately got a lot more signups, which they eventually managed to monetize quite successfully.

They monetized using a subscription model; he said that a good subscription model is sliced (so free players get to see parts of it), and visual (so free players can see what subscribing players have). The price points matter: they had 1 month, 6 months, and 12 month plans, and initially set them up so the longer plans were a modest savings over shorter ones. That led to most people signing up for 1 months; so they switched to putting a much narrower gap between the 6 and 12 month plans (5 / 24 / 30 pounds, respectively), and all of a sudden almost everybody signed up for 12 months! Another anecdote along those lines: when the Economist had an online-only subscription for $59 and a print and online for $125, then 68% of people chose the online-only version; but when they added a print-only option to their price list, also priced at $125, then it switched to 84% of people choosing the print and online option!

1:45 pm: The Convergence of Flash Games and Social Games, by Daniel Cook.

I went to this talk because I love Lost Garden and because slides from talks he’s posted there before looked good. But it was mostly focused on how to use ideas from social games to be successful on Flash portals; not so interesting for somebody who has been reading Dan’s ideas on his blog for years, who is already in the social game space, and who doesn’t want to be successful on Flash portals.

3:00 pm: Seriously, Make YOUR Game!, by Paolo Pedercini and Jason Rohrer.

Paolo’s bit was amusing theater. Jason’s went into the details more: it turns out that his upcoming African diamonds DS game didn’t have its origin as a social statement, but rather from thinking of what multiplayer strategy video games can do that board games can’t, deciding that hidden knowledge (e.g. spying on the other player’s position) is one possibility, and coming up with a game mechanic where hidden knowledge about knowledge would work. (Having agents in the field buying diamonds, who can possibly be bribed.)

Also, when playtesting the basic purchasing mechanics with a physical game, he found that they weren’t much fun; going into it a bit more, he discovered that they’d stumbled on Nash equilibrium behavior. He broke that in two ways: by introducing a nontransitive relationship (basically, a hidden rock-paper-scissors mechanic, to avoid a dominant strategy) and by having bids where both the winner and loser pay (to avoid having the bid price settle at the item’s value).

4:15 pm: Indie Gamemaker Rant!.

No super standouts, though certainly better than yesterday’s AI rant. I guess the part that clicked the most with me was Robin Hunicke talking about the many ways in which having a male-dominated industry is a problem. Certainly her and Kellee’s talk earlier this morning struck me as a good example of how bringing a broader set of perspectives is likely to have a positive effect on your game’s development.

So that was today’s talks; my first time going to the Summits part of the conference, it’s mellower than the main conference. Very nice dinner, too, and I was glad to finally get to meet Jorge Albor!

Post Revisions: