this is where I used to post Mystery Hunt recaps and I don't have anywhere better
2017 was my 15th (!!!!) Mystery Hunt. I haven't been able to make it back to Cambridge for 5 of the last 6 years, though, and I hunted from the Bay Area again this year. Every year I say "next year in Cambridge", but that might not happen again until Ida's a bit less of a baby. I had previously hosted the Metaphysical Plant bay area HQ, but I couldn't host this year and there was more critical mass around a Bay Area crew for Hunches in Bunches, the slightly-younger-Random-alum team, and I was excited to hunt with some old friends who I'd never hunted with before, so Hunches it was! I had a lot of fun and really appreciate being invited. Plus, while I always enjoyed hunting with Plant, my meta-solving skills are more in line with other folks on Hunches; Plant tended to tear through metas in person super fast as soon as there were enough answers and I didn't get much of a chance to contribute. This year I got to make some tangible contribution to at least half of the metas, which was great!
(As a note --- I think it's nice that it's possible to experience Hunt remotely, but I firmly believe that Hunt is an MIT event that should focus on people in Cambridge. If the writing team declared that Hunt can only be done in person, I'd figure out if I could travel or take the year off and not be too upset about that choice. The team I helped organize for 5 years was pretty firmly a "everybody fits in this big room, we don't have to put effort into making it easy to collaborate over longer distances" team, and I loved it! When my team opened the Treasure Chest round two years ago, I wistfully wished I was in Cambridge to see it but did not feel in any way that it was my right to be able to fully experience the Hunt from the comfort of my own home.)
Obviously the most salient fact on everyone's mind about this Hunt is how short it was for many teams. Our team finished around 8th on Saturday afternoon, and Death and Mayhem found the Coin before I went to sleep Friday night. I'm pretty OK with all that. I love that so many teams got to see the entire Hunt! One of Dan Katz's posts reminded me of something I'd forgotten: on my very first Hunt (with ACRONYM in 2003), having a runaround halted midway because Kappa Sig had found the coin. (Admittedly this was a really long hunt already, and our team hadn't figured out how the structure necessary to get to the second runaround worked.) I much prefer clean non-broken Hunts where lots of teams get to see the ending. And that does seem to be the direction Hunt is heading: the past 4 years have seen 8, 11, 4, and 17 teams finish Hunt by my count.
Our team backsolved a lot of puzzles, especially in a few rounds where we found the metas easy and easily backsolvable. Fortunately, that just meant more puzzles to solve after we found the coin; I spent my surprise extra free time trying a few of them. Well, that and spending time with my baby daughter :) That said, the hunt structure (where we couldn't get more quest puzzles without solving quest metas, waiting for events, or "solving" character puzzles) really incentivized us to backsolve character puzzles; we had plenty of answers ready even before puzzles were unlocked and plugged them in immediately so that we could unlock more puzzles in the rounds where we hadn't solved metas yet. Usually in Hunt, backsolving is something you do to puzzles you're stuck on, not puzzles you haven't seen. That said, Hunt is a big complex beast that's experienced differently by every team, and I don't actually think SETEC should have done anything differently here.
Some thoughts on specific puzzles:
- Wizard meta: somehow we got this one off of ?W???MI????. I don't believe it either.
- CHINA KILNS: I loved how the mechanism of this puzzle meant that the answer phrases could be completely ridiculous and unguessable without understanding the puzzle structure.
- Asymmetry: The basic concept of this puzzle is sorta uninspiring: listen to two clues being read over the two channels which clue words that differ in a letter. But the execution was amazingly hilarious, with so many tricky homophones everywhere. Sometimes the brilliance is the execution rather than the idea.
- Ye Olde Seventhe Duck Konundrum: The Cones of Duckshire: Daniel, our host in Berkeley, made it really really clear he wanted to do the Duck Konundrum. So of course we unlocked it while he was running an errand. Fortunately, we managed to stall without Cambridge taking over the puzzle until he got back (and it took us a while to print out a multi-page PDF for a big board anyway). As always, the Konundrum was lots of fun. My only regret is that we made a huge mistake in solving it --- we forgot to watch the original Cones of Dunshire clip first! I have since rectified that problem.
- Out of the Mouths of Babes: Unfortunately there are no famous "Baby" songs that only consist of the words "mama", "dada", "cat", "dog", and "Dandy", so my training was less help here than I would have hoped.
- The Leaning Tower of Sheshach: Oh man this puzzle. We spent a long time on it, even though we got the basic Atbash idea early. We thought for a while that we just wanted to find pairs of words that directly translate to each other in a third language, and that we'd just use Atbash on the big letters at the end. After all, there's no way that there would be 20 pairs of words in these languages that map to each other via the horribly ugly Atbash cipher, right? Nope. Human languages are amazing. All random series of characters are actually Polish words. Plus, any puzzle that ends with building something is fun.
- Changing Rooms: I love cryptics, and Lead With Hydrogen was a fun easy one to start out with. But this one... woo. That was pretty tough! (And only partly because of the ridiculous crane word.) But it was ultimately satisfying, and plus it's the first time I've done a crossword with my old friend Kaia, so that was nice. Even nicer was that some folks in Cambridge were far far better than I at Star Battle and LITS; I identified the puzzle types but was happy to let them take over.
- An Hunt MIT Mystery Puzzle: One of the few we never solved, and one where we apparently were trying too hard. We easily identified the movies, but were incredibly focused on the few words that were changed from the original quotes (mostly names, but also a couple other random words like adding "unobtainium" to the Up quote). Oops!
- Crafty Criminal meta: Sometimes there are Hunt metas that I love everything about except for how they lead to really bad puzzle answers. (There were a couple like that last year --- KARIN INGEL-ANKER, I'm looking at you...) I'm impressed that this tightly constrained meta could be constructed with answers that were more or less recognizable! (Even the oddest one consisted of words I knew from Michael Chabon's Gentlemen of the Road.) Well done.
- Modest Minstrels meta: Oh, the bad idea alleys I went down on this one. We were all so sure that the weird time signatures were relevant. I was pretty sure for a while that this was somehow a Pictures at an Exhibition meta. But the real method was basically perfect. I marveled at my teammates' ability to recognize modes by ear; I could pull it off with a piano at least.
- Chromesthesia: I did not work on this, I merely got to hear my teammates' minds get repeatedly blown over the course of this puzzle, wander over, and get my mind blown as well. Codecs are amazing.
- Elf/Dwarf Brawl: Did not see during Hunt, have watched video far too many times since.
- Schoolhouse Rock: So I could imagine a Hunt puzzle that's like this except you have like 10 separate songs to do this to, and you get one syllable per song. (And with some Hunt philosophies, that's the only way you'd see this.) And part of me says --- this puzzle was so fun! Three of us singing along and searching for the words for 15 minutes was great; maybe 2 hours of it would be great too! But the rest of me says --- um actually this puzzle was just perfect. Fun to solve and finished before it gets tedious.
- Smart Moves: OK, first of all, some of these dancers are AMAZING. That said, this had the biggest BE NOISY moment for me I've ever had. We weren't able to find the PROTEINS video for a long long time (that video has many scenes that look very different and it was easy to skip over it). Now look at that answer grid. There's only one column that passes through all the entries other than ENTRIES: the one to its left. If you're missing PROTEINS, then that column reads "BESTT?N". It was very very very hard to not treat this as clearly reading "BEST TEN ENTRIES". There is definitely no list of best ten Dance Your PhD entries anywhere online. I know because I looked for a long time. So SETEC, that's why we called in BEST TEN and BEST TAN and BEST TEN ENTRIES and many more things like that.
- The Puzzle at the End of this Book: There are a lot of puzzles that I wish I had solved fresh after Hunt without seeing anyone else's solutions. This is probably the one that applies most to.
- Dispatches from the Foreign Service Units: This one was just great. And I learned a lot, like the fact that despite Unicode not having enough space to put Chinese and Japanese versions of similar characters as separate codepoints, it apparently has room to have an entire block of characters that are literally just from one object we dug up called the Phaistos Disc.
A note on solutions: I appreciate that not only were solutions ready post-Hunt, but that for computer-aided solutions, sample programs were included as well. There are some elaborate puzzles in previous hunts that require computer help where I never felt like I understood how you could actually go about going from figuring out how the puzzle worked to actually solving it (say, "In the Details" from 2013, which I finally understand as of yesterday); actually having sample programs is super helpful and inspiring.
I'm pretty excited to see what Death and Mayhem does next year! I generally prefer Hunts from more MIT-connected teams, and a lot of my old college friends are on this team. This is the first time a team I've ever been on has won (though I've joined two teams in the post-victory-not-trying-to-win-this-year stage). Honestly I'm pretty optimistic. Death and Mayhem has enough experience under their belt that they know what a generally enjoyable and non-broken Hunt looks like, but is also new enough that I suspect they'll be some creative new surprises up their sleeves. And frankly... the worst case scenario is we get another kinda-too-hard-kinda-broken-still-full-of-interesting-and-fun-moments Hunt, like in 2004 and 2013. Periodic semi-broken Hunts are part of the cycle of life, and if they happen about once in every undergraduate career, it's not the end of the world. Don't get me wrong --- I hope and expect that 2018 won't end up like that. (There's a fair number of folks on the team who've written before, anyway.) But even if it does, it's just a university puzzle weekend, and far be it from this aging alum to be too upset. (Even if it is the best university puzzle weekend!)
(As a note --- I think it's nice that it's possible to experience Hunt remotely, but I firmly believe that Hunt is an MIT event that should focus on people in Cambridge. If the writing team declared that Hunt can only be done in person, I'd figure out if I could travel or take the year off and not be too upset about that choice. The team I helped organize for 5 years was pretty firmly a "everybody fits in this big room, we don't have to put effort into making it easy to collaborate over longer distances" team, and I loved it! When my team opened the Treasure Chest round two years ago, I wistfully wished I was in Cambridge to see it but did not feel in any way that it was my right to be able to fully experience the Hunt from the comfort of my own home.)
Obviously the most salient fact on everyone's mind about this Hunt is how short it was for many teams. Our team finished around 8th on Saturday afternoon, and Death and Mayhem found the Coin before I went to sleep Friday night. I'm pretty OK with all that. I love that so many teams got to see the entire Hunt! One of Dan Katz's posts reminded me of something I'd forgotten: on my very first Hunt (with ACRONYM in 2003), having a runaround halted midway because Kappa Sig had found the coin. (Admittedly this was a really long hunt already, and our team hadn't figured out how the structure necessary to get to the second runaround worked.) I much prefer clean non-broken Hunts where lots of teams get to see the ending. And that does seem to be the direction Hunt is heading: the past 4 years have seen 8, 11, 4, and 17 teams finish Hunt by my count.
Our team backsolved a lot of puzzles, especially in a few rounds where we found the metas easy and easily backsolvable. Fortunately, that just meant more puzzles to solve after we found the coin; I spent my surprise extra free time trying a few of them. Well, that and spending time with my baby daughter :) That said, the hunt structure (where we couldn't get more quest puzzles without solving quest metas, waiting for events, or "solving" character puzzles) really incentivized us to backsolve character puzzles; we had plenty of answers ready even before puzzles were unlocked and plugged them in immediately so that we could unlock more puzzles in the rounds where we hadn't solved metas yet. Usually in Hunt, backsolving is something you do to puzzles you're stuck on, not puzzles you haven't seen. That said, Hunt is a big complex beast that's experienced differently by every team, and I don't actually think SETEC should have done anything differently here.
Some thoughts on specific puzzles:
- Wizard meta: somehow we got this one off of ?W???MI????. I don't believe it either.
- CHINA KILNS: I loved how the mechanism of this puzzle meant that the answer phrases could be completely ridiculous and unguessable without understanding the puzzle structure.
- Asymmetry: The basic concept of this puzzle is sorta uninspiring: listen to two clues being read over the two channels which clue words that differ in a letter. But the execution was amazingly hilarious, with so many tricky homophones everywhere. Sometimes the brilliance is the execution rather than the idea.
- Ye Olde Seventhe Duck Konundrum: The Cones of Duckshire: Daniel, our host in Berkeley, made it really really clear he wanted to do the Duck Konundrum. So of course we unlocked it while he was running an errand. Fortunately, we managed to stall without Cambridge taking over the puzzle until he got back (and it took us a while to print out a multi-page PDF for a big board anyway). As always, the Konundrum was lots of fun. My only regret is that we made a huge mistake in solving it --- we forgot to watch the original Cones of Dunshire clip first! I have since rectified that problem.
- Out of the Mouths of Babes: Unfortunately there are no famous "Baby" songs that only consist of the words "mama", "dada", "cat", "dog", and "Dandy", so my training was less help here than I would have hoped.
- The Leaning Tower of Sheshach: Oh man this puzzle. We spent a long time on it, even though we got the basic Atbash idea early. We thought for a while that we just wanted to find pairs of words that directly translate to each other in a third language, and that we'd just use Atbash on the big letters at the end. After all, there's no way that there would be 20 pairs of words in these languages that map to each other via the horribly ugly Atbash cipher, right? Nope. Human languages are amazing. All random series of characters are actually Polish words. Plus, any puzzle that ends with building something is fun.
- Changing Rooms: I love cryptics, and Lead With Hydrogen was a fun easy one to start out with. But this one... woo. That was pretty tough! (And only partly because of the ridiculous crane word.) But it was ultimately satisfying, and plus it's the first time I've done a crossword with my old friend Kaia, so that was nice. Even nicer was that some folks in Cambridge were far far better than I at Star Battle and LITS; I identified the puzzle types but was happy to let them take over.
- An Hunt MIT Mystery Puzzle: One of the few we never solved, and one where we apparently were trying too hard. We easily identified the movies, but were incredibly focused on the few words that were changed from the original quotes (mostly names, but also a couple other random words like adding "unobtainium" to the Up quote). Oops!
- Crafty Criminal meta: Sometimes there are Hunt metas that I love everything about except for how they lead to really bad puzzle answers. (There were a couple like that last year --- KARIN INGEL-ANKER, I'm looking at you...) I'm impressed that this tightly constrained meta could be constructed with answers that were more or less recognizable! (Even the oddest one consisted of words I knew from Michael Chabon's Gentlemen of the Road.) Well done.
- Modest Minstrels meta: Oh, the bad idea alleys I went down on this one. We were all so sure that the weird time signatures were relevant. I was pretty sure for a while that this was somehow a Pictures at an Exhibition meta. But the real method was basically perfect. I marveled at my teammates' ability to recognize modes by ear; I could pull it off with a piano at least.
- Chromesthesia: I did not work on this, I merely got to hear my teammates' minds get repeatedly blown over the course of this puzzle, wander over, and get my mind blown as well. Codecs are amazing.
- Elf/Dwarf Brawl: Did not see during Hunt, have watched video far too many times since.
- Schoolhouse Rock: So I could imagine a Hunt puzzle that's like this except you have like 10 separate songs to do this to, and you get one syllable per song. (And with some Hunt philosophies, that's the only way you'd see this.) And part of me says --- this puzzle was so fun! Three of us singing along and searching for the words for 15 minutes was great; maybe 2 hours of it would be great too! But the rest of me says --- um actually this puzzle was just perfect. Fun to solve and finished before it gets tedious.
- Smart Moves: OK, first of all, some of these dancers are AMAZING. That said, this had the biggest BE NOISY moment for me I've ever had. We weren't able to find the PROTEINS video for a long long time (that video has many scenes that look very different and it was easy to skip over it). Now look at that answer grid. There's only one column that passes through all the entries other than ENTRIES: the one to its left. If you're missing PROTEINS, then that column reads "BESTT?N". It was very very very hard to not treat this as clearly reading "BEST TEN ENTRIES". There is definitely no list of best ten Dance Your PhD entries anywhere online. I know because I looked for a long time. So SETEC, that's why we called in BEST TEN and BEST TAN and BEST TEN ENTRIES and many more things like that.
- The Puzzle at the End of this Book: There are a lot of puzzles that I wish I had solved fresh after Hunt without seeing anyone else's solutions. This is probably the one that applies most to.
- Dispatches from the Foreign Service Units: This one was just great. And I learned a lot, like the fact that despite Unicode not having enough space to put Chinese and Japanese versions of similar characters as separate codepoints, it apparently has room to have an entire block of characters that are literally just from one object we dug up called the Phaistos Disc.
A note on solutions: I appreciate that not only were solutions ready post-Hunt, but that for computer-aided solutions, sample programs were included as well. There are some elaborate puzzles in previous hunts that require computer help where I never felt like I understood how you could actually go about going from figuring out how the puzzle worked to actually solving it (say, "In the Details" from 2013, which I finally understand as of yesterday); actually having sample programs is super helpful and inspiring.
I'm pretty excited to see what Death and Mayhem does next year! I generally prefer Hunts from more MIT-connected teams, and a lot of my old college friends are on this team. This is the first time a team I've ever been on has won (though I've joined two teams in the post-victory-not-trying-to-win-this-year stage). Honestly I'm pretty optimistic. Death and Mayhem has enough experience under their belt that they know what a generally enjoyable and non-broken Hunt looks like, but is also new enough that I suspect they'll be some creative new surprises up their sleeves. And frankly... the worst case scenario is we get another kinda-too-hard-kinda-broken-still-full-of-interesting-and-fun-moments Hunt, like in 2004 and 2013. Periodic semi-broken Hunts are part of the cycle of life, and if they happen about once in every undergraduate career, it's not the end of the world. Don't get me wrong --- I hope and expect that 2018 won't end up like that. (There's a fair number of folks on the team who've written before, anyway.) But even if it does, it's just a university puzzle weekend, and far be it from this aging alum to be too upset. (Even if it is the best university puzzle weekend!)