It's a talk by Andy Rich, who was on GC for Microsoft Puzzlehunt 14. Well that link is a link to a video of his talk. This here is my notes. [My rambling asides are in italics] and I take some pretty egregious summarize|rephrase|totally-change-meaning liberties with other folks' words, too.
- He's from Team Liboncatipu. Oh hey, Liboncatipu isn't just the MSPH14 team. They were also the Seattle team who wrote Zombie SNAP. [I know they simulcasted that: I was on the Bay Area side of that simulcast, where we called it BANG 19.]
- Oh man, he said I put together the bay area simulcasters. [Actually, that was Alexandra Dixon and Joe Fendel. Those folks should write more so that far-away people remember their existence instead of attributing their deeds to me.]
- Someone corrected him: Actually that was Golden Golems (Joe) and Mystic Fish (Alexandra) [my memories are hazy, but that "someone" might have been me grabbing a microphone in Mountain View: That's why you need to attend GC Summit, to avoid getting mixed up in some kind of misattribution thing leading to some kind of blood feud. GC Summit: It could save your life.]
- Shout-out to Dan Egnor and the Demonic Robot Tyrannosaurs (combo team of Burninators, Coed Astronomy, and other geniuses) for simulcasting MSPH14
- MSPH is not The Game, not exactly. No locations, stuff happens on MS campus. More about the puzzles than the amazing experiences.
- Bigger scale: more than 80 teams, a dozen people each.
- [That's something that made flying up to Seattle to help run a MSPH so amazing: standing at the front of an auditorium looking out at so many people.]
- Finding simulcasters Ask around. If someone says they're not interested, ask if they know someone who might be interested.
- Seattle puzzlehunters have a Facebook page, and it's a good place to start grapevine-ing.
- Use your simulcasters as early playtesters! The bad news: the later they playtest, the later they can start on ther own preparations. So early playtesters.
- And especially use your simulcasters as playtesters if they're a combination of the Burninators, coed astronomy, and other geniuses because they're frickin' puzzle-designing geniuses.
- Why is it hard to be a simulcast team?
- The crap-work, not the fun-work. Location wrangling instead of puzzle design.
- Relying on main GC [but too far away to grab them by the lapels and shake answers out of them]
- Since Liboncatipu used Dropbox to store puzzles, nice side effect was that remote GC could also see latest versions.
- Separate mailing lists so that, e.g., Californians didn't see tons of mail about reserving rooms on MS campus. But a couple of folks from each group would hang out in both mailing lists to make sure important stuff got relayed.
- Oh man, we wrote special server software for this game, and that was late. I bet the simulcasters were nervous about that.
- Communication during game, not just before game is good. Californians found bugs in puzzles during the event and were able to get word up to Redmond—but had a hard time because folks weren't checking the mailing lists during the event, had to scrounge up someone's personal contact info.
- Content is King Don't just write down the puzzle, write down solution info, too. Simulcasters will want it. And you were going to need to write all that crap down for the hunt wrapup web site anyhow, right?
- Portability Like Scott said, if you wanna simulcast your event, don't fill it up with location-specific puzzles.
- Remote GC made an all-remote game. They only used the puzzles that could be delivered over the internet. This meant that they skipped a couple of puzzles. There were a couple of puzzles where they had "This is what you would have seen if you'd been there" to deliver "environmental" information.
- Some puzzles solved to MS Campus stuff like Go to such-and-such conference room. Remote GC had to tell their teams: if a puzzle seems to tell you to go someplace on MS campus, call us instead so we can tell you what you would have seen.
- Didn't want their hunt software up on "the site" while other Seattle events going on, because GCs of other games would thus have access to hunt data. [Wha? Not sure I follow this part. Do all Seattle GCs share the same machine for serving hunt answer checkers? That doesn't make sense.]
- Wanted remote GC to serve their own copy of the site, so that load spike from one city wouldn't overwhelm the game for everybody. BUT game client (a silverlight app) had hardcoded URLs, so had to relative-ify those.
- Original hunt software used players' MS corporate login to authenticate. Whoops, that doesn't work in the outside world where not everybody works for MS.
- Don't be working on the hunt software the night before the hunt. That is not the path to a happy life.
- (Audience rueful chuckles from folks who worked on hunt software during the hunt)
- Letting Go The remote GC is gonna need to change some stuff. You invested part of yourself into those puzzles, but you gotta step back and let them change it nonetheless.
- Q&A
- Dan Egnor of remote GC points out that something that helped a lot was having a single point of contact for each group. If you were in CA and needed to get word to "the right people" in WA (or vice versa), you'd talk to the WA "liaison" who would know who to pass the word onto: just mail arich and the right things would happen.
- Yeah, I think you pretty much need someone in that role. If you just tell folks to mail some mailing list, stuff falls through the cracks.
