It's a "Puzzle Design: Surprises and Aha Moments" a talk by Greg Flipus, who wrote the Research Triangle #Octothorpean puzzles along with some feats he mentions in the "background" section 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. Original videos and slides at this here link.
I was pretty glad to see this video. As you'll notice, the sound quality was pretty echo-y. At the live event, at any given moment I was only halfway aware what Greg was talking about. It's clearer in the video.
- Greg's background: Mostly the Microsoft Intern Game, the weekend-long driving-around event for MS interns (not to be confused with Intern Puzzleday, a shorter event on MS campus).
- Dig at a bit
of self-deprecation in Ian's talk: recent Intern Hunt
had a quark-flavors puzzle that went well.
- This talk is kind of a variant on Ian's: Where Ian reminds himself not to over-focus on "Wow" over "Fun", this talk revels in some "Wows" that worked.
- What's good about Ahas?
- More memorable; whether or not they're "the best," folks'll talk about them later.
- Maybe your puzzle comes in a 3.5" floppy disc. Data on the disc makes
a message; but the message says ANSWER IS INSIDE DISC CASE. So now you
need to break that open.
- This "aha" has been used plenty of times in the Intern Game; by now, it's a cliche: a team receiving a break-apartable thing with data on it expect they'll break that thing apart.
- This works because players have blind spots.
- [I'm not sure "blind spot" describes my feelings towards clue destruction. "Strong aversion," more like. I still remember how nervous I was destroying-revealing a clue for REDACTED, and that was a few years ago.]
- Besides don't-destroy-the-clue, what other "blind spots" do players have where we can hide Ahas?
- Blind Spot: If there are multiple copies of info, share them around.
- Ah, but what if those "multiple copies" aren't exact copies, but have differences? What if that's the puzzle?
- Blind Spot: tough constraints are tough.
- If your puzzle seems impossible to construct, that makes it tougher to solve, since teams think "Well, they couldn't have done that...
- [Hmm, that sounds like wandering into not-fun-but-you-left-it-in-to-impress-puzzle-designers territory.]
- Oh, but the examples aren't getting into un-fun puzzles.
- E.g., No way could they get famous actor Neil Patrick Harris to record video footage for this game.
- Example from old Intern Game: team received a puzzle; upon solving, discovered that the answer was customized to their team. Wow, spooky! Unexpected enough to surprise teams.
- You've been carrying the answer with you for the past few hours.
- Example from intern game, in which a sheet of paper labeled "rules" wasn't just the rules of the puzzle, but actually important since the puzzle worked by dropping letters and anagramming: which meant those "rules" could become a LURE.
- But remember: don't make teams feel stupid.
- And it's rough when teams need a hint, and the only hint you can give them gives the whole thing away.
- And what if a clever team gets a couple of Ahas in a row, leapfrogs through a couple of puzzles, and now your schedule is b0rken?
- Ian Tullis: Yeah, I worry about clever teams breaking the schedule.
Like, if you use that perfect thematic answer, what's to keep teams from
just plain guessing the answer in the first minute?
- Greg: Yeah, I worry about that, too. It never seems to actually happen, though. Well, wait. We had a hunt with a couple of zodiac puzzles. For the first such, a couple of teams tried entering all the zodiac symbols as answers. (But the answer wasn't a zodiac symbol.) For the second such, no team tried that. Which was lucky, because the answer was ARIES.
- Ian Tullis: Yeah, I guess that doesn't really happen so much. I worry about that, and I worry about someone else "scooping" my puzzle idea. But that doesn't really happen either.
- ???: Except that year when all the hunts were themed on the Mayan apocalypse.
- Ian Tullis: Could you say more about designing for the intern
game versus other puzzlehunts? Like, I know you can't use too much
culture-dependent stuff because MS gets interns from all over the world.
- Greg: Quite a range of experience—some have only learned about this stuff from playing in Puzzleday a little while before. Some have helped run the MIT Mystery Hunt. In general, we aim for the easier end of the spectrum. One nice thing about having such a restricted audience: it's a lot easier to get playtesters and volunteers when most of the area's puzzlers aren't allowed to play.
- Todd Etter: That kind of goes back to worrying about teams getting the "Aha" too early. The experience of the designer isn't going to be the same of that of a solver. If I'm playing in a casual event, I'm not going to be so aggro about rushing to the end of a puzzle.
- ???: I went through a phase where, when there was a puzzle in front of me, I'd try "all the tricks." But it never worked and it wasn't fun, so I got over that.
- Greg: Like for that floppy disc puzzle‐the info inside the disc wasn't actually the answer; it was a password to give to the program that you got from the disc. If you'd started just by opening up the disc, it might not have saved you time.
- Ian Tullis: On destruction: playtesting a puzzle for The Griffiths Collection, I got a puzzle with some crayons. Out of desperation, I started eating the crayons.
- Brent Holman: For the Espionage game, we broke open a clue to get the answer. We got the answer quicker, but that clue was a work of art. Other players have a cool memento for their trophy case. In hindsight, I regret that destruction.
- Todd Etter: Playtesting the Famine Game, there was an awkward part. We got a tower of Legos with the instructions: figure out what's going on, don't take the tower apart. The pattern of bricks made a message: take the tower apart. So… it's possible to worry so much about teams "jumping ahead" that you give instructions that make your puzzle seem pointless.
- Greg: Yeah, you need to be able to rely on GC. And there are already so many opportunities for that go wrong. Maybe someone on a team calls the hint line, gets some info from GC—but garbles an important detail when relaying that information to the team. Later on, when folks figure out that was wrong, they say "GC lied!"