devjoe (devjoe) wrote,

MIT Mystery Hunt 2020, Part 1

It's time. The Hunt ended 4 days and some hours ago, and it's time for me to start writing up my recollections.

Before the Hunt

There was a bunch of drama before the Hunt even started. MIT's latest attempt to kill the Hunt was a policy change prohibiting us from solving Hunt on campus between 1AM and 7 AM. Also, a week before hunt when we heard this, we still didn't have our rooms. This sent us scrambling for alternate headquarters and we secured a ballroom at Le Meridien (which is right behind The Asgard). We got a good price, but this did not include any food and we were told that, while they wouldn't prohibit us individually bringing food into the room, we could not serve food at the hotel unless it was from their (expensive) catering. So instead of stocking our HQ with hundreds of liters of soda and tens of pounds of snacks along with mass takeout orders from nearby restaraunts, we told teams they were on their own for food, and gave them suggestions including a number of local restaurants and places they could buy food to stock their hotel rooms with. As for me, I rented a room (Friday and Saturday nights only, since I'm local-ish) and stocked the minifridge with cokes and sandwich fixings. When I wanted food, I took a short break and went upstairs and got it. And I slept. Real sleep, 5-6 hours each of Friday and Saturday nights, but starting around 11 rather than on MIT's schedule, which kept me more on my normal sleep schedule.

A couple days later this new policy got changed to: You can solve in your HQ between 1-7 AM only if you have a current MIT student in the room, still no sleeping in rooms and no solving around the rest of campus these hours.

And I totally understand the no sleeping in classrooms. That has actually been a rule for quite some time, one I have been known to violate involunatrily by attempting to stay up for too many hours at once. The 1-7AM is a weird curfew, but I guess they had to pick some time and they wanted to make it for enough time to encourage people to get a reasonable amount of sleep.

The change didn't really help us, as we had only one current MIT student and we couldn't really force him to stay up those hours, so we stuck with our alternate plans. We'd gotten our room assignments at this point, in building 66. While we could have spent a lot longer indoors on the trip starting on campus, this is almost as far from Kresge as Le Meridien.

And then they told us that due to some unforeseen circumstances they might be changing our room assignments on Thursday, one day before Hunt. Since we'd gotten a room way out in the far reaches of campus with tablet desks and a smaller room with a table that was good for maybe 8 people, we decided we were going to stick with Le Meridien and we'd give both our rooms back if they could give us one room closer to us. We ended up getting something in building 8 I never visited. In fact, I stayed at Le Meridien the entire time, except when we came to do the endgame en masse on Sunday; my experience with the MIT campus was not needed this year.


There was another kind of drama before Hunt started, and that was the wedding invitation.

This one looked a lot like the wedding invitation from 2011 except the room and date were different, and it was M and G rather than M and P. 2011's wedding turned out to be Mario and Princess Peach, a wedding interrupted by Bowser doing what Bowser does - kidnapping the princess - to introduce a hunt that wandered from Mario's world into the worlds of certain other video games. This year ... it was a real wedding! Mark Gottlieb and Gaby Weidling, who've been a couple well-known in the puzzle and game community for several years now, finally got married and did it for real at MIT in front of a standing-room-only crowd in Kresge and who knows how many others via live video feed (preserved on Youtube).

While Mark and Gaby have been known as a couple for several years, Mark's association with Mystery Hunt goes back much further. His 1998 MIT Thesis, cited in Wikipedia's article on the Hunt, describes some of the chaos from the 1996 Hunt which his team ultimately won. When my team, as The Archivists of Beginners' Luck, worked to fill in the real gaps in the archive after the 2010 hunt, we used this thesis to help solve parts of the 1996 hunt to supply missing answers for the web site, in particular for parts of the Hunt that were literally unsolvable due to not having the non-paper parts of the puzzles

While the wedding was briefly interrupted by Cupid giving us a fake theme about finding the heart which all love emanates from, after they got married we got the real theme, based around Mark's desire to have their honeymoon at a theme park he visited as a kid, Penny Park. We learned the story of how Penny Park was named for the founder's granddaughter Penny, and 28 years later, that granddaughter now ran the park. We also learned that the various mascots, one of whom would ceremonially open the park each day, were still serving in these roles - but each of them had their own problems that kept them from doing their jobs correctly. And the park itself was not doing well and was about to close for good! So of course, we were going to solve the park's problems by doing what we do every year - solving puzzles!

The 2020 Hunt was located at (and a static version with solutions still is, though I don't know how much longer it will be there; ultimately it should move into the archives at MIT). At 1:30 PM on Friday, it opened for us to start solving.

Storybook Forest

This small round was where most of our very first puzzles came from, though we were getting puzzles within multiple open rounds almost the entire hunt, but I'm going to keep things simple by going a round at a time rather than trying to go linearly. Since there were few puzzles open at the time, most of these were piled on by many people and solved fairly quickly. I worked on only two of these puzzles:

Hackin' the Beanstalk featured an interactive terminal where typing one letter caused one or more complete lines of Python code to appear. Ooh, a Python puzzle! My favorite language! Let's go check it out. Well, there was a time when I might have been a team's Python expert. I learned Python back when Python wasn't cool, essentially on a dare in 1997, as part of a challenge proposed by another player in the nomic game Ackanomic. It was Python 1.5. Now it's 2020, and you can find Python programmers on every street-corner (at MIT, anyway :-), and this one was getting dogpiled as much as the others. I participated in an early effort to catalog the code snippets that came from pressing various keys, and then once we figured out that typing the name of a computer scientist produced code implementing an algorithm named for that scientist, I managed to contribute just one name by looking up named algorithms starting with one of the letters we had, that name being POLLARD. Then my teammates figured out there was a bug in each program, and quickly extracted letters and solved the puzzle.

Toddler Tilt describes a rather unsafe and demented sounding amusement park ride: 64 toddlers with letters painted on their heads stand on a platform, which tilts, causing them to slide in one direction, and one of them to fall into a trench. This is repeated until they all fall off. You can read the puzzle. I got in on this one near the end. Teammates had GANG BOSS CAPONE which of course clued AL, and I helped them fill in some missing letters in the others to spell ANTONYM OF WOMAN (MAN), DOMAIN OF CYPRUS (CY), and ultimately A BUCK IN BRUGES. They had guessed A DUCK... and were trying to translate into Dutch. Also, without this answer, they were trying to use the other parts phonetically to make CY + MAN + ??? + AL = SIMON COWELL and other wrong Simons. I checked and say the D they wanted was impossible from what they had left, got the correct clue, and after a short qhile got the correct ordering and submitted the answer.


In Space, No One Can Hear You Sing arrived during my shift as the team's official answer-caller-in and callback person, so I took the callback that requested two karaoke singers be on campus half an hour from that moment, and quickly found some people to go. When they got there ahead of their assigned slot, we got the URL for a Youtube stream (available now in the puzzle) showing our two singers and those of other teams. I kicked off the effort to start recording the song titles, masks that some sort of video filter was putting on the singers while they were singing, and scores. And a teammate who, unlike me, had seen more than a small part of one episode of this series, identified that the songs and masks were from The Masked Singer. We identified the songs, figured out the masks and scores were repeating so we only really needed one cycle, and figured out the loop we had to form by linking the songs one contestant on the show sang to the mask he wore while singing them, extracted the two letters from each song given by the scores, and then got stuck for a while on IANO ROCK BAND BALDWINS, thinking we'd somehow missed a mask that would have given us a P to make the first word PIANO. Eventually somebody parsed it correctly as BALDWIN SIANO ROCK BAND (who knew SIANO was a name?) and quickly found the name of the band these two were in.

In Stress Test, I helped solve some of the clues which provided words spelled the same way but pronounced differently with the stress on different syllables, but that was pretty much all I did. Other people figured out the correct extraction (which took as a while).

And that's all I've got time for tonight. There are a bunch more rounds with about 8 times as many puzzles in them as this.
Tags: mit mystery hunt 2020 puzzles puzzlehunt

  • MIT Mystery Hunt 2020 Part 4

    The epic conclusion! (If you like, revisit part 1, part 2, and part 3.) I've left out some other details, including the events and how we…

  • MIT Mystery Hunt 2020 Part 2

    Picking up where the first part left off... Wizard's Hollow Charming: I recognized that this puzzle was cluing Harry Potter spells and ..…

  • Mystery Hunt Index Updated

    It's only early October and I have all the 2020 puzzles indexed. Still slow but better than some years. I've addressed some feedback you…

  • Post a new comment


    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded