(VERY SPOILERS) Requiem: A 2024 MIT Mystery Hunt Write-up

So, we¹ wrote a hunt.

It was quite the experience to be doing this from the writing side for once (and right in the thick of it at MIT), so I wrote a lot about it.

As always, read at your own peril, for spoilers will follow. Opinions are my own, and not reflective of those of TTBNL or Puzzle Club. And of course, this post is not a puzzle. I just like framing my life with classical music. It’s not worth reading too deeply into that.

Rhapsody on a Theme of Paganini

The process started practically as soon as we got back to HQ with the coin in tow. Charles, the team’s captain, held a debrief about hunt that I barely listened to due to having been awake for 21 hours and also due to my attention span being taken up by the jigsaw puzzle². Hunt was now an active thought that would take up a significant* proportion of my time for the next year, so I tried my best to keep it out for the next week or so to avoid puzzle overload. I enjoyed ANE³ in the meantime, trying very hard to not accidentally leak any theme proposals to the two puzzlers I shared Ubers with, before development kicked into full swing.

Charles soon invited me onto the executive team. I had absolutely zero idea what this could entail and made this thought known, though I did air a preference for an editing role. For now, though, I was primarily a liaison with MIT Puzzle Club, which I was already on the executive team for.

Of course, the first thing a hunt needs to decide on content-wise is the theme, since the choice will cause a ripple effect throughout the rest of the hunt.

Sara, our creative lead, took a lot of initiative with getting this particular ball rolling; the plan was to get the theme nailed down by the start of February, which was impressively early⁴ considering the timelines we were shown during the debrief meeting.

We got some 70 elevator pitches for themes, which was too much decision paralysis for me to be comfortable voting for or against anything.

Through that preliminary round of voting, the 70 elevator pitches were whittled down to 12 proposals that were going to be fleshed out into a formal pitch: storylines, round ideas, potential strengths and weaknesses, a way for smaller teams to not be bottlenecked⁵. I didn’t actively contribute to any of the theme-making, but I definitely read through all of them—and one I was particularly convinced would make a good theme, which would eventually become the main theme of the hunt. I’m a bit of a nut for Greek mythology (though this comes mostly from having read too much Percy Jackson as a middle schooler, and then having watched too much OSP as a high schooler), and that theme got my idea generator rolling the most. Which also meant I was going to ask some questions that needed to be addressed in order to strengthen it as much as possible. I was struggling a bit on deriving second-stage round ideas, for instance, but that seemed to be less of a problem than I projected.

Then came the voting stage again, which was a mixed bag of excitement and despair. I would have been happy with writing for a decently large proportion of the themes, but I could definitely feel tension as I looked through the voting results—enough that I explicitly called out a schism as a likely outcome. We eventually deemed the end result “too close to call” between two particularly front-running themes, which derailed our plans for two days. The third vote was equally fraught, and the two themes we chose for the runoff were neck-and-neck for most of the process. The idea termed “Puzzlympus” at the time won out by a singular vote. (I won’t elaborate on the other potential theme here, considering it might resurface for a future hunt.⁶)

Which really speaks to the passion that people have for their vision for the world’s largest puzzlehunting event. I saw large floods of images being posted into our creative channels advertising their particular vision to prospective voters, including many, many GIFs of some kind of dog trio doing cute things.

There were some hurt feelings at the decisions, but (fortunately for us) no particular sign of schism. We set the puzzle-writing infrastructure up and moved into February ready to write puzzles.

Spring Sonata

Life stops for no one, of course, so my junior year plodded on unimpeded with a three-class load⁷ and an all-too-familiar night owl sleep schedule.

I was approached by Craig at the start of the month to gauge my interest in becoming one of the chief editors for the hunt. Of course, I could not pass up that particular opportunity, but I’ve known myself to overbook heavily for lengthy periods of time. Whether this had any particular effect on my morale has yet to be seen.

Of course, after themes comes metas—and as one of the chief editors I got to see basically all of the ideas coming through.

The editors had a meeting every Wednesday to talk about things—and for the first few months this was mostly meta ideas. Many worked better as feeder puzzles (and a decent number of those made their way into the hunt). A few lost track of the plot pitch and themed metas around the ostensibly-dead-but-oops-not-for-this-meta Pluto. We had a lot of music and a lot of sports. For every meta idea that made it in there were another three that got axed for some reason or another. Obviously I can’t talk about all of them here, since (a) they might come up as future puzzles and (b) there are simply Too Many Ideas on the cutting room floor.

I also started floating around ideas of my own at this point—mostly for feeders, because I was coming up blank on meta ideas. Both of the main ideas I had within the span of this month lent themselves well to minipuzzle suites. The main puzzle that came out of my first round of thought was Modern Architecture, inspired by a walk back home from SGS⁸. The second idea was themed around the five rivers of the underworld.

February flies past alarmingly quickly—it’s been a month since the theme was finalized and we had no metas firmly in the ground. The first meta to do so wouldn’t come for another month. We put a couple of metas in testing and they keep getting cycled out for one reason or another—mainly “too much research” but also “inelegance” and “red herrings everywhere oh crap that’s a lot of red herrings”.

So I try my best to think up some meta ideas around the start of the month. One of the ideas I had was for a Styx round, though I felt it treaded a little close to another meta in passive development (the original idea for the Persephone Meta, back when Persephone was still supposed to be in the Underworld, which eventually became a subpuzzle of Steam Library), so I didn’t submit it into puzzup just yet.

The other was a riff on Mike Sylvia’s Prodigious Riddle of Juno and its sequel, which I originally called “The Piercing Rhetoric of Hera”. I didn’t realize the “PR” thing was an entirely deliberate signature until Craig pointed out the initialism.

Thankfully the other title was easy to ideate.

The idea was essentially unchanged from conception (the only major change being switching from Latinate components to Greek ones). Of course, the conception of this made construction a tall order, and I wanted a meta answer nailed down before funnelling time into construction, which was a second tall order⁹. I spent around a week constructing around a mediocre answer I came up with on the fly as a proof of concept before hitting upon what I saw as the perfect pun for the puzzle (with the only downside being that it constrained my construction a bit more, leaving only four transformations between number and letter). Constructing around that took another month and a lot of my sanity, as well as a lot of bashing queries into word tools… which still got me only 85% of the way.

However, the idea ended up being back-burnered, as a different take on the fish meta had also appeared around this time and was advanced instead—mine was the backup.

I interspersed construction with typical editor duties, though of course MIT courseload meant I could only have so much on my plate. I was assigned the editor role for the Rivers/Charon meta, originally a semantic meta based around the White Star Line, around the time we decided we had too many semantic association metas; we eventually came up with the idea of creating a literal river on a grid out of the names of underworld rivers. Completely incidentally, I had an idea using the same rivers, which meant that idea was now dead in the water¹⁰.

The first draft was seven letters wide and called for a completely pure meta; testsolving revealed this to not be at the right difficulty for an intro round and we added the grid. We then had to redraft it a few times because our initial construction had half an Error That Cannot Be Named and a number of subsequent drafts had inelegancies that affected the solve in one way or another. Eventually I bumped the size to a nine-wide and drafted something that I found reasonably aesthetic; a slight revision passed testsolving in early May.

At this point, we were still only five metas in.

Tod und Verklärung

Writing a hunt about the underworld definitely got me thinking about the nature of death a lot.

I’m told that TTBNL hunt themes tend to leak into the real world. TTBNL has apparently tried to write a hunt two separate times before winning Mystery Hunt. The first was themed around a pandemic; this was planned around 2020, and got cancelled when it wound up presaging The Actual Pandemic. The second was meant to be presented as an unfinished hunt, and… you can guess what happened there. This made some of the top brass leery about tempting fate with the theme proposals back in January—Charles in particular liked to joke about theming the hunt around world peace—but I never took the superstition all that seriously.

Early March, I (re)discovered Spiritfarer from watching a streamer I followed and went down an emotional whirlwind of a rabbithole reading through all the information about it. Seeing the strong theme connection, I noted it down as a puzzle idea. This became Asphodel, which started as the meta idea I mentioned earlier. It was first withheld due to the aforementioned Persephone meta development and then further due to an overload of semantic association meta ideas, before I decided to make it a feeder; further development of the idea would be stalled to after the Styx round was fully developed.

The theming of that game being what it is, though, made for the absolute worst possible timing for the lightning to start falling.

It was spring break, and I had moved into a family member’s apartment for the time being to slow down a bit from the typical breakneck pacing at MIT. I was five days into an otherwise relaxing period when my sister broke the news to me.

Drago was, directly or otherwise, the catalyst for my interest in linguistics. I only knew it was a potential subject of study from the competition he spearheaded. The puzzles were sufficiently up my alley for me to seriously consider linguistics as a major. I literally would not be where I am today without his guidance. Even if I only knew him from one or two in-person meetings, to see such a major influence on my life fade out within four years of doing so… the feeling is not something I can put into words.

I wanted to do something in his memory. Tentatively, I decided my best option was to ask the IOL what I could do at some point, but I also put an idea in Puzzup for a puzzle about the IOL. I also precommitted to it being credited under my real name, assuming it made it in—the other puzzle about an Olympiad was under construction at this time.

Either way, life had to continue, grieving or otherwise. March gave way to April, a few more metas passed the bar, I started thinking about Modern Architecture’s minipuzzles and got three of them on paper, and I go to a Logic Puzzle Open held at MIT and meet some more people I only knew from the internet.

A week later, on the third of May, MIT Puzzle Club was talking to a representative from the MIT Museum about setting a puzzle for some event they were holding in June. CJ tangentially mentions that an e-mail had arrived in our inbox from one “Susan Polansky” urgently asking us to contact them.

The surname vaguely rang a bell—it was shared with the real name of Jack Lance, an avid¹¹ puzzle constructor. Out of curiosity, I read the e-mail… and unfortunately, I knew some context that CJ apparently didn’t.

A few days prior, someone had shared a slightly worrying change to Jack’s website to an online puzzling community that all of us were in—all the verbs were switched to past tense. The site had been configured to trigger the change on a timer just two days before.

On their own, this change and that email would have looked frivolous. Seeing the two in such close conjunction immediately led me to one grim conclusion.

I told exec about the context. They got the implication. None of us could believe it. All of us had seen Jack, alive and in seemingly good spirits, just ten days prior at the open.

We passed the email on to the mod team for that online community, knowing we still couldn’t respond to it in a satisfactory manner; they confirmed our worst fears.

The news gradually started spreading around, and the construction team started throwing ideas around for a fitting tribute (what would eventually become ENNEAGRAM). I silently concluded that the TTBNL theme engine was, in fact, cursed. Intrusive thoughts¹² weighed my head down in the middle of the most stressful fortnight in the semester, and I had to push through them as hard as I could. I had to check out a bit from hunt construction—those papers weren’t going to write themselves.

I survived it, or so I hope.

A Midsummer Night’s Dream

Then I immediately got on a plane for Seattle, got stuck in the north terminal because midnight is the absolute worst time for a layover and the airport’s rail system shut down while I was looking for food, started postsolving the Microsoft Puzzlehunt I missed in the meantime¹³ because why not, got on a second plane to Anchorage, hopped onto a cruise ship with my family, and got stuck with crappy internet for a week, all in the middle of the most busy week for meta writing.

The trip did result in me seeing eagles in the wild for the first time, and also netted me one (1) small eagle plush.

On the boat, I contributed a rewrite for the Hydra that didn’t rely on a letterbank; I’d’ve pushed a different draft that extended the pattern but didn’t have time or internet bandwidth to implement it. Besides that, though, I mostly stayed out of the hunt for the week I was on the boat, which meant I missed the majority of the push to complete our complement of metas. By the time I got back to terra firma, that was done, which gave me little idea as to what to do with regards to my own meta proposal. The other divisions seemed to be operating under the assumption that these were all the rounds in hunt, so I asked… and got the response “yes, we expect your meta to be in hunt”.

I did take a look at the other metas that had made it into hunt at this point. While they were all pretty good, some ideas I saw were a bit more… unnatural than others. The part I had the most apprehension about design-wise was Hermes’s three-part structure. Most of it came from the second round’s meta just being a full-on word manipulation conundrum, but I also found the assignment section of the third meta a little bit… confusing. Granted, if you knew going in that they each represented a unique white chess piece, the assignment is more obvious, but no one was going to land on the starting square as their first interpretation of B/W. (I personally would have preferred K/Q, but those are rare letters, and the first part already used five “pawn” answers in addition to the two rooks.)

By the time I redocked and flew home, Craig had set something up to track borrowed answers, so I looked at the Styx list and took one for Asphodel, which was drafted up over two days. Testsolving was… more turbulent. The first one immediately jumped to the conclusion that it was a cipher Nurikabe. Welp, move some of the “clues” around to preemptively defeat that assumption. Also rewrite that flavortext. Second try went clean, but short-circuited a lot and suggested some changes that I put off for a later date.

Eventually, though, it came time to think about what exactly Hera was doing. I still had my 85% of a draft sitting around and spent a considerable amount of time thinking of ways to finish it up, eventually ditching my rule against synonym triangles to close the gaps on lines 1 and 10.

The puzzle was originally just partial fragments of the first and last statement in the puzzle, as well as a handwritten note with details that were eventually rewritten into the middle statements. This was a massive red herring for a number of reasons that were mostly unearthed during the maiden voyage, and I soon rewrote the puzzle into its current form and added the skeleton framework. It passed a testsolve. Then we wanted to test how beginner-friendly our fish meta was, and as it turns out the answer was “not really”—our testsolve got the assembly down after some time but floundered on the punchline. It passed a different testsolve in the meantime, but I was a little uncomfortable with it in its current form, since I decided I had overshot the intended difficulty. We eventually decided that it was fine as is, with an asterisk to hint through any aspiring but struggling teams. I wasn’t perfectly happy, but eventually decided for it after considering what difficulty other “fish” metas landed at.¹⁴

And just like that, we had our meta assembly: 17 rounds consisting of a total of 215 feeders¹⁵. And July was just starting to trot to a halt.

…wait, 215?

Nightmare on Bald Mountain

Yeah, we’re halfway through the year and only, what, 25 puzzles have made it past the testsolving stage? It’s not zero, but it’s still only 10% of the expected value. That’s a lot of ground to cover.

We talked about The Number in the subsequent editor meeting, and didn’t end up committing to a hard decision. We did, however, settle on a plan for what to do with it if we couldn’t reach this number of puzzles.

As such, we started setting targets: M puzzles graduating and N puzzles drafted in a given month for arbitrary values of M and N. We started light with 25 in the latter half of July, but that fell short by half and we then needed to be a bit more aggressive with those numbers. Goals for August and later were in the forties or fifties. Almost none of these numbers were met, but they were generally close, and at some point in September we passed the inflection point of one puzzle graduating per day.

This became slightly more impetus to bring Asphodel to its conclusion—I redrew a lot of its assets to make the flowers more identifiable by RIS, and then redesigned the grid to dodge the short-circuiting. And then I had to make a burner email to make an account to update the Fandom wiki to have the actual shape of the building in a separate inset, because the size entry just defines a bounding box and that tripped up several solvers¹⁶. And then I had to update the flavortext because I didn’t have good inroads for people who didn’t know the game (which thankfully worked the first time; I would have lost my mind if I had to edit it more than once).

I meant it as a quick drop. It became an albatross, taking up desperately needed headspace. Over time incoming testsolve feedback made me a lot less happy about my aesthetic preferences. It was still terrorizing my thoughts as I moved on to drafting other puzzles, and by the time it graduated in mid-September it had already taxed most of my passion for it.

I distracted myself by constructing a different puzzle, drawing up ideas throughout August and starting writing just as I was about to head back to campus. It would take a bit to come to fruition, but I decided it was worth it.

The rest of the team was also hard at work cranking out feeders. Three separate writing retreats happened around the start of August, but I didn’t have the freedom of movement to attend any in person and instead tried to supervise the remote participants. Given my sleep schedule, I had a considerable amount of trouble doing so.

Eventually, it came time to start my senior year. I flew back to MIT at the start of September ready to coordinate the whirlwind of on-site stuff—wouldn’t be an MIT Mystery Hunt without the MIT, after all.

Academic Festival Overture

This, as it turns out, was a bit rough to balance. While I perceived myself as having a light load for the time (three classes again), dividing my attention between everything that warranted it proved to be too much, and I ended up being forced to take a lighter class load (after some consultation to ensure it wouldn’t interfere with the rest of my senior year). Fortunately, it made my Mystery Hunt planning smoother.

With metas in place and MIT life about to get in the way, I started checking out from formal editing duties, only really stopping to comment for the weekly meeting or if I had a particularly strong opinion on a puzzle topic. This slight role shift also served to start accommodating my role as liaison for logistics, which was going to be happening a lot throughout my fall semester. A lot of this was simple stuff—I booked the Bush Room in September, Lumia called IS&T to open up the Discord ports on the Guest Wi-Fi for the duration of the Hunt in October, and we start booking the other teams’ rooms for hunt (fortunately we’re on pretty good terms with the Registrar).

I spent my September writing my other puzzle and visiting the Wiesner Gallery trying to think of something to add to Modern Architecture. I also took the opportunity to scope out billboards for the badge puzzle. September then bled into October, which was more of the same, until an answer freed up that could fit Modern Architecture’s strict cluephrase requirements. Immediately I worked on getting the three parts I knew how to write onto paper.

In the meantime, I also started coordinating runaround details with Nick—and while we agreed on several points of general themata it was hard to even start writing details until we knew which rooms to use. I decided the ideal case for the mid-hunt runaround endpoint was Barker Library (the fifth floor of Building 10, just under the rotunda), but it was closed on the weekend. I set up a meeting with SOLE in mid-September to talk about other things, and as an aside asked for a location similar in atmosphere; SOLE suggested iHQ. I then got swamped by aforementioned work and held off on actually sending in the request until mid-October, at which point I also booked some other rooms we might need for the final runaround.

This was also around the time when events were being finalized; of these, the second event for Saturday afternoon was most concerning with regards to bureaucracy, since it involved acid-base chemistry. Jessica drafted up a safety plan for it, which I now had to deliver to… some authority.

The calendar ticked into November, and still very little has happened. I needed to step up the pace.

I had also talked to SOLE about putting up posters in unorthodox locations for the duration of Mystery Hunt, which has happened several times before without incident. SOLE promised a response once they talked to the right people, then vacillated on the response until I literally walked up to their front door two months later, at which point they still didn’t have a firm response for me. They then asked for a list of locations we wanted posters at, which I gave the next day. (I still didn’t trust them enough to respond in a timely manner.)

Then I got the reply from iHQ (also belated). I expected a simple confirmation or rejection and was instead met by the mild surprise of a $4000¹⁷ invoice in my inbox.

I was never informed of this, and decided on the spot that the cost overhead was a little bit prohibitive. I immediately appended this fact to my original SOLE reply and tried to work out a backup plan, but everything falling apart at such a crucial moment was simply too much pressure to bear.

This was also around when we formally opened registration, which magnified this pressure sevenfold.

Breathe in. Breathe out. One step at a time. First up: dealing with the logistics fire.

I stopped by SOLE again a week later, still visibly irritated. Still no answers about the posters, but I took the time to get event registration details sorted out. Next I asked the CAC front desk about where I could submit a safety plan for one of our events; they pointed me to EHS and I sent the thing. Their turnaround was much quicker, and I came back with a bunch of questions to answer; the next few weeks essentially consisted of me relaying responses between EHS and Bella, and we eventually agreed on a plan of action.

The day before Armistice Day, I met up with Nick on main campus to map out a route our mid-hunt runaround. This runaround was a little less refined back then—our main idea was still doing an Orpheus and Eurydice thing in the Tunnels, but the plan was to separate them, send them down two separate loops, and get them to meet up again near the endpoint¹⁸. We found some interesting paths, including a route through the Stata Garage and the biology buildings that loops around back to Landau and isn’t marked on the official maps. At the exit, though, we noticed a sign that said the building was locked “after hours”. But it didn’t define “after hours” precisely, and the day before Armistice Day was an institute holiday, so we thought this just meant “nighttime”.

I then return to check the precise definition of after-hours availability the next Sunday, and come face-to-face with a locked door. As it turns out, the majority of the hunt is after-hours for the purposes of that route. Guess we were going to have to come up with a new plan. The only thing I could really leave unchanged was our new intended destination room, which I then booked with the Registrar.

Further runaround development was now stalled until Nick could get back to campus, so I went back to ideating puzzles. That Friday I finished writing my other puzzle, meaning I could focus harder on logistics. Shortly afterwards I figured out what I wanted my last minipuzzle on Modern Architecture to be, just before Fall Break put a brief stop to my activities on campus (meaning I would have to wait for Sunday to implement it).

In the meantime, I made a little sketch.

Our in-house art-team was rather sparse, so a lot of the hunt’s art was commissioned. While I do draw occasionally, I wanted to leave the artwork to people more available than I (since I was going to be spending my hunt time editing and writing for the most part). I produced a bunch of art for my own puzzles, primarily Asphodel, and functionally nowhere else.

Well, nowhere else except for this sketch.

The Yellowstone round art was a little bit special, in that it sourced art from across the team—just doodles of mythical creatures, nothing too fancy. I sketched up these Stymphalian birds over a two-hour span of Thanksgiving Friday and sent it in. They now grace the round art, representing Arts and Crafts once it gets solved.

In the meantime, we finally hit a goal: 50 puzzles graduated in November. At this point there were just 49 overworld puzzles left, along with some spare fish. Hunt progress was in good shape.

Winter Wind

40 days and counting.

Just like Asphodel, that puzzle I finished last month testsolved turbulently. I basically used my first testsolver group as a guinea pig for various levels of nerfing. The nerf did solve a lot of problems, and would be tweaked a lot over the later days as the testsolve surfaced issues later on.

In the meantime, we brought Modern Architecture to a finish; I had handed off the crossword-filling portion to Craig, who had finished it by the first week of December, and I did the rest of writing in the meantime. I spent a bit collecting pictures and typing up a document for testsolving, and off it went.

Nick came back to visit for the last two Saturdays I was on campus.

On the 9th, we took a walk along our original route for the mid-hunt runaround. Nick had a new idea to try, so we took a bit looking for distinctive features in the tunnels of MIT that could produce some numbers. We came up blank on ideas for the final; we created a Discord channel to brainstorm over the next week. This channel was mostly silent for the next week.

The 16th was, thankfully, more eventful on this front. We spent a bit brainstorming, and eventually started drawing up a list of the major Olympian deities that we hadn’t used yet (namely, Zeus, Apollo, Ares, Demeter, Hephaestus, and Athena¹⁹). It was then that we struck upon our final runaround plan—we picked suitable locations on campus for four of them, and we had the skeleton for the victory lap within an hour and a half.

Unfortunately, this walking around, combined with a timely statement from fellow EIC Nathan, also brought to my attention the fact that the Dreyfus and Green Buildings—two areas where access was vital for the puzzle I had just written—were closed on the weekends.

I asked SOLE for avenues to pursue, and they told me that my best course of action was interacting with the relevant departments directly. This was also around the time winter break was starting; I didn’t have any finals that required me to be on campus, which meant I had to sort all this out before my flight home on Monday.

No pressure. Just a whole puzzle riding on what I do here.

I visited the relevant buildings several times over the course of my last week on campus. Each time I vacillated on talking to anyone out of fear. Of what, I’m not sure. I sent a bunch of e-mails to possibly-relevant people on Thursday, not knowing if they were the right people to ask. Finally, on Friday, I decided to actually talk to them.

I didn’t know what to expect. I described it as walking into a hornet’s nest, likely an understatement since I already engaged so little with the faculty and other students here. I actively failed at getting words out of my mouth for several minutes after I walked into the office. Eventually, though, I got my words out, and got an answer from both—the departments were amenable to opening the buildings for the weekend, but they would have to discuss things further.

This left no logistics problems on my plate when I flew home on Monday. I was looking at three weeks of stress-free time for Getting Things Done.

Which meant it was prime time for something else to go wrong.

I already knew that my other puzzle was going to struggle during testsolving. I saw only so many people on the team that could feasibly do everything, and so I actively avoided seeking help with construction to maximize the pool of potential testsolvers. It took three months to hammer the content out into a form I was happy with; another fortnight (and a maiden-voyage testsolve) to make it friendlier to solve.

However, the second testsolve had now been dragging for another fortnight. It was provisionally paused on Wednesday morning. At the Editors’ Meeting that night, we decided it was too risky to rebalance—its round had other hard puzzles, and we couldn’t move any of them to a different round without burning another four months we didn’t have on construction.

I stayed silent for the rest of the meeting, lost in my thoughts as I struggled to fill out the take-home final I had that day.

I consider myself remarkably good at keeping my personal woes offline, but real-life Adalbert instantly fell into a depressive slump. This was four months of on-and-off work, aggregating close to a hundred hours, amounting to functionally nothing.

I sat on my feelings for the rest of the week, only occasionally getting up to get necessary edits to Modern Architecture after its first testsolve also struck turbulence. A damper on an otherwise uneventful Christmas. If the edits hadn’t successfully gotten Modern Architecture through two clean testsolves²⁰, I’m not sure how my mental state would have been in January.

Trying to run my creative engine only served to further mire my brain in self-doubt, so I spun my wheels. Inspiration striking now would be an unlikelier event than a lightning strike now, especially with me being as bad an ideator as I was.

Yet that’s exactly what it did.

I occasionally scroll through the IOL problem archive. On one such perusal I was reminded of an IOL colleague’s translation of a book on Transcendental Algebra. Immediately I started piecing it together with the rest of that IOL idea I put in several months prior; while there were concerns given the several other “esoteric equations” puzzles in the hunt, the other editors let it in (given enough spacing). By New Year’s Day I had a working puzzle.

It blitzed past testsolving on what was essentially the first go—the cleanest progression curve among all of my puzzle drafts. I was a bit salty, given the comparatively less effort it took to piece together, but it was a much-needed morale boost as I was preparing to fly back east.

This was one of the last puzzles we needed to finish the hunt’s puzzles. We were finally done.

A few tasks still remained for the team (notably postproduction and a final factcheck), but things that needed editorial decisions were finally going to slow down. We also started a full-hunt testsolve with what unspoiled solvers we had on Christmas, which was progressing. Not optimally, but still progressing.

The 9th of January

My return east, originally planned for Saturday, was delayed to Sunday due to an unanticipated winter storm. Back at MIT, though, were more logistical issues rearing their heads… primarily administration wanting to meet with me every morning.

The rest of Puzzle Club was handling rooms and the like. I got to talk to MIT Audiovisual services and E33 to coordinate kickoff. At 9:30 AM.

We also had to send out the things that typically go out before hunt—liability forms (I was sent a docusign before winter break and we only sent that out to teams on Wednesday (sorry)), Health and Safety Quizzes (which I just forgot existed until I compared to emails from last year), minors policy (which we already dealt with in November), and Tim Tickets (which I set up last week but forgot to send the QR code for).

Wednesday afternoon, after four hours of class, I walked up to Le Meridien to meet up with the growing number of TTBNL people at the hotel. Everything was going great—all the art was done, the website logic was functioning, we had actual coins and merch on-site, and the now-complete complement of puzzles were running through one final factcheck before being released for hunt. The last editor meeting was essentially just a walk into University Park, where we had very little to talk about besides contingencies and the random chemical symbols etched into the stone cubes.²¹

Just one more day to go.

I walked in as early as I could manage on Thursday morning, right on time for the Le Meridien ballroom HQ to open. I brought a rather larger eagle plush along for moral support. Lumia swung by early to drop off the HQ phones, but we quickly found that they didn’t come with power cords, so we had to order some (which fortunately worked). Soon the ballroom was packed with people, each doing their own thing—whether that was reassembling a pinball machine for Game To Be Themed Later or spinning up balls of thread for Encore.

For my part, I was exiled from the ballroom immediately, because MJ was going over events and I couldn’t be there to spoil myself on one of them, just in case I was needed to run a lap for Puzzlympics. I ate my lunch (one (1) bag of instant ramen, because why not) and watched two people create small skeins of yarn in the meantime.

Once I was clear to come back in, I was pulled aside for an overview of the Science Event. For me, this mostly involved mashing blueberries with gloved hands and brewing a tea out of the remains, which thankfully worked the first time, giving me a nice set of colors.

I then went to do factchecking things, and was pulled back into the events loop to do a quick runthrough of Nero Says. I guessed at first position from what I knew from postsolving MSPH, then promptly forgot what an arabesque was in spite of learning that much earlier in my life.

I got back to checking postprods. We then had a team-wide meeting, with a bunch of admins giving a brief précis of what we had achieved throughout the year. I made a point of collapsing dramatically when they mentioned there being one (1) MIT student on the team.

As they were wrapping up that presentation, I gave my email a brief glance, and I found a worst case scenario waiting for me, almost as if to mock me on that point.

Not only did we repeat the mistake of having a non-student talk to CAC, CAC decided to impose itself onto the Science event, and it was on the verge of not happening. Hunt was literally going to start in 18 hours, and there’s already a flaming wrench flying our way.

We legitimately thought EHS was the only organization we needed to converse with on this front. We forgot CAC owns the room. The good thing about this is that we already had our conversation with EHS, so we just CC’ed them into the thread (along with a rather lengthy apology²² email) and hoped they would work it out between them.

Still didn’t help my mood though. Several more physical manifestations of my mental fortitude falling apart happened over the course of dinner.

Also not helping my mood? The full-hunt testsolve. Which was more of a three-person mass-solution-read at this point since the actual solves had slowed down. And whose leader was now loudly complaining about broken puzzles left, right, and center—complaints about load-bearing flavortext or factual inaccuracies or whatnot—and seemed to specifically have a grudge against A Radical Fishing Trip. Why? He claims to know Japanese and is unhappy about several things he thinks constitute incorrect Japanese.

Dude. We had at least two separate Japanese speakers look at this. One of whom is yours truly and also an EIC.

Which meant it fell on me to talk him down, and I was already not in the mood to entertain any BS. So I pulled up Jishō on my phone to back up my extant Japanese knowledge and talked a dozen points down to one easily-fixable point, which we promptly fixed.

Eventually, though, it came time for me to head back to the nest. I grabbed a badge and a staff T-shirt on my way out. Before I hit the hay, though, I had one final task to do.

Being the one MIT student on the team meant I also became the point person for anything to do with printing. And a lot of things needed that printing—not just the on-site puzzles, but also the hundred-or-so liability forms and other miscellanea for the chemistry event, the entry passes for the night club, and a semaphore reference diagram for Nero Says.

So I waltzed down to the printer in my dorm and babysat the printer in my PJs for 10 minutes. Unfortunately, I was not going to be able to carry two hundred documents in my backpack without damaging any of them.

Fortunately, I had a solution sitting right on my bookshelf.

Symphony of a Thousand Puzzlers

Friday morning. Here we go.

I unfortunately missed my intended wake-up time of 7, instead bolting out the door at 9 with the eagle plush under my shoulder and a box full of printing in my backpack. I distributed everything as necessary once I got to Le Meridien, after which we immediately went to Kresge to practice the skit for kickoff. I was asked to go print another 330 pages of script for the practice, so I diverted myself to the Student Center and babysat another printer for ten minutes. I’m so thankful I thought of the box I got to keep from last year—I found out it was just the right size for a ream of paper when I packed up my dorm room in May.

I checked my email. No response from Meredith.

I run to Kresge with the box and join in on the rehearsal. Fortunately, my part was essentially 90% reading off a script and 10% self-directed ad-libbing, so most of it was watching everyone else.

I check my email after my lines. No response from Meredith.

I then ran out as soon as I decided I wasn’t needed for the rest of practice in an attempt to meet up with my online friends again. As it turns out they weren’t in the Flour this time and had moved across the street, and by the time I actually got there they had all but dispersed. Welp.

I got back to Kresge in time for a full run-through. We then needed a run-through with all the tech stuff, which pushed actual kickoff back by twenty minutes or so. When we finally opened the doors, I could hear a chorus of a thousand puzzlers stream into the auditorium, milling about and trying to find good seating.

Eventually, the lights dimmed. The crowd quieted down.

Breathe in… Breathe out.

Kickoff, I’m told, was a smash success.

I met up with some of the puzzlers that I knew from online afterwards. We talked about a bunch of things, including the late Jack Lance. I’m told Jack’s parents came to kickoff; they wanted to experience the hunt as Jack would have.

I checked my email afterwards. Still no response from Meredith, and the event is literally in 24 hours.

CAC would only be open physically for today. Now or never to visit. Paul decided to tag along; I’m thankful for his support, because my beak is hardly naturally coherent even on the best of days.

I walk into CAC’s office and ask for Meredith. She’s not there, but someone who is offers to look into it for us.

My heart is racing at a brisk allegro, but I tried to project a calm face. No pressure. No pressure. Play the Largo in your head. Everything will be fine.

It seemed like an eternity had passed when the CAC person finally responded.

We were clear to continue.

A little bit of cleanup work later, I walked out of the CAC office, much relieved and in a significantly better mood. I had finally gotten the last of the logistics work off my back, and now I could enjoy the hunt to its fullest. And right on time, too—it was now 13:20, and the site should have just opened to the crowds of waiting puzzlers.

Oh, if only it were that simple.

The Planets

The Bringer of War

When we finally got to our temporary²³ HQ room for Friday (10-250), things were… a little less than OK.

From what I could tell, nothing about the server has changed from teammate’s code. Half of HQ was in a frenzy trying to figure this out, the other half trying to endure a now-extended calm before the storm.

I checked my email out of curiosity. It was already starting to be filled with teams emailing the Puzzle Club email—which was decidedly not the advertised HQ email—wondering why their puzzles weren’t appearing. Some emails were… more diplomatic than others. I could only imagine what the actual hunt email situation looked like.

To this day we still don’t know what transpired here—Ethan brought in teammate to help and they couldn’t figure out what was wrong either. I’m not sure what precisely tech did to get the site working again, but (a) it was now vastly improved (as much as you could call an increase from 0 a “vast improvement”), (b) it still crashed and had to be “kicked” around every hour or so, and (c) it came at the cost of everything websocket-related, the most notable of which was solve sounds. So uh. Sorry about that?

In the meantime, though, the editors were meeting about actually getting puzzles to people while the tech issues persisted. While the Throne Room (Cerberus) was the first round, it also had one puzzle with an interaction component (Deep Conspiracy) that was liable to exacerbate our extant load. The decision was thus to release what would have originally been the third round (Judges) as a set of Google Docs, and set up an analog system to track answer submissions.

Which, yes, meant that we were temporarily resurrecting the old system of phone callbacks.

We got an area set up in one of the upper corners of 10-250, made a Google Form for submissions, and had factchecking go through all the Google Docs versions for accuracy. We were on the verge of sending it out when we found out that teams had gotten access to the first round.

We sent it out anyways—gave them stuff to work on if the server got turbulent again.

In the meantime, I peeked into the servers I had gotten permission to spy on. I was glad to see that they at least noticed the badge puzzle early, which tided them over for part of the downtime. But now it was actually off to the races.

The Bringer of Peace

With the server still being rather unreliable for the time being, we also set up an analog to the spoilr system used for tracking solves on the blackboard. Tracking the solves was a fun time that kept me occupied for the first few hours of hunt, right up until the first event was due to start. In the process I got to witness a bunch of teams blitz through 70% of the underworld and finally set my mind at ease with regards to Asphodel.

Oh yeah. Among the several other responsibilities of being The MIT Student: I have to supervise all the events. Which, given that I attended precisely zero events last year, I treated as more of a boon.

So I watched the first event go by—a nightclub event involving making coded bracelets, telephoning a dance routine, and making music. Overall I found the ordeal quite funny, particularly the dance section, and doubly particularly when I found out one of the dances was just the dance to Dschinghis Khan’s Moskau.

At the end of the event was a drink bar for six flavors of soda. I grabbed a bottle of the most familiar one (orange) to undry my throat. I then took the opportunity to walk around and visit some of the teams I knew people on, catching the TSBI Swarm coming back from the mid-hunt runaround and hitting ²⁴ at the tail end of theirs.

I got back to HQ past 10 PM and was immediately sent off again on courier duty. Then, 80% of the way to the destination, I was called back. In the meantime, spoilr had mostly stabilized ± the hourly reboot (meaning analog spoilr was now obsolete) and several leading teams had finished the Underworld, meaning the first two Overworld rounds (Minneapolis and Yellowstone) were now open to them.

Among the first puzzles in Yellowstone? Transcendental Algebra.

Nathaniel soon asked for my opinion on a unique type of callback for that puzzle’s task. After hearing his proposal, I simply had to agree… and immediately threw the answer at NES (the first team to complete the task) in every language I could think of.

Someone else then walks up to me and suggests toki pona as another language for the callback. At that instant, an idea works itself out in my head… and wouldn’t you know it, the intended mark (✈︎✈︎✈︎ Galactic Trendsetters ✈︎✈︎✈︎) is the second team to find the instruction.

I could not have asked for more perfect timing.

Too bad no one on the other end realized I was speaking in puflantu—I gave them the answer in a different langauge, then tried to visit them to explain and found that they were leaving for the night.

Eventually, the clock would tick past midnight, and I packed up my things to head back. All in all, a good day—the underworld target was hit on time, three teams had solved different overworld metas as their first essentially simultaneously²⁵, teams were now staring at the Hydra, and surely more puzzles would trickle in throughout the night.

The Winged Messenger

Saturday morning saw me print another several copies of the script for the Science Event. I needed to hurry, though—in some cruel joke my alarm failed to wake me up at a good time, meaning I was nearly late to the second event I had to supervise. Not to mention I still had more paper to print and drop off, because our email was sent into timeout mid-hunt and we needed to make as many of those interactions in analog as possible.

I dropped off the papers at the new HQ and bolted to the Student Center, barely making it in time for the host’s explanation of the “Puzzlympics”. Thankfully, I wasn’t needed for the event itself. Instead, I got to watch a lot of not-quite-running and drum out a taiko beat during finals for the cheerleading squad.

I grabbed lunch at Maseeh before heading back to HQ. Interestingly, nothing much had changed in the hours I was asleep for—the leaderboard for metas was relatively static, and the only thing I really missed was the opportunity to call ET Phone in Answer for their Transcendental Algebra submission. (I later DMed what I would have said to them.) We treated this as a sign that our unlock pace was running slow (likely related to how the Hydra was structured), and set some rounds to unlock earlier as a countermeasure.

After a bit of watching the big board and a bottle of orange soda, it was 1230, and thus time to set up for event number three. The one I’ve been fretting over all year: the Potions.

While the others were setting up the buffers and other things Ostensibly, we had safety equipment stored in W20-020. I went down to check. The room was still the same it always was. Hmm.

I reread the email, explicitly remembering agreeing on 020. Alec had sent an email since I last checked.

He dropped it off at 020A, an entirely different room. One that was locked.

Confused, I take an elevator upstairs to find a phone number to call. The person on the other end (who turned out to be the Meredith I was looking for yesterday) agreed to unlock the door for me. I pick up a bunch of tubs with some other things inside and walk back to the elevator. Meredith asks me if that’s all I need from the room; I say yes.

Cool, we now have waste management.

I was then called back to HQ to go deal with an urgent matter, because while I remembered that Wiesner was open to the public, I forgot to come up with a contingency for times when it wasn’t open. Access was important here.

Fortunately, we already had a solution, and I just needed to go print a QR code.

Unfortunately for me, the nearest black-and-white printer²⁶ is all the way out in Building 1.

So I took a walk, briefly poked a very curious eagle plush into Left Out’s room, grabbed the printout, handed it to Hosea to post at the entrance to Wiesner, and immediately headed back for the Student Center, since the event was about to start. We were almost done with setup, but Summer and Alina had a few questions for me about what I brought upstairs.

The organizers saw nothing to actually store the waste in. I pull up documentation to check my work, and thankfully there’s an inventory in one of the emails. Let’s just double-check…

Welp, time to call Meredith again. This time she offered to help with carrying everything up.

The event itself went off without a hitch—no chemical spills (though I remember at least one team spilling their turmeric supply) and no freak accidents. I was mostly in charge of making sure we had enough hot water to distill everything with, a task essentially equivalent to babysitting a GregTech boiler to ensure it had enough water. After the two hours were done, we took another hour to be meticulous in cleaning everything up and disposing of it properly, and we put everything back into W20-020a to be dealt with later.

We got back to HQ, and I could finally breathe a sigh of relief. Couldn’t relax just yet, though—the author for A Trip Down Memory Lane approached me about the on-site component for his puzzle having gone missing.

I go out to Building 46 to check… yep, it’s not there. I ask Hosea (who was the actual person in charge of doing that setup, since I was busy with kickoff then), who tells me to check “the other tree”.

And this is how I learn about this sculpture on the other side of Building 46. And yes, the poster was there—fortunately, moving it to the other side (after having someone double-check its location first) was an easy task.

Out of paranoia, I also went off to check on the Modern Architecture plants. The first one I checked (18) was slightly misplaced (the unusual font of the periodic table certainly didn’t help there), but the rest were fine.

With that done, I walk back to HQ. Progress on metas for the leaders has been slow—teams had solved Vegas during the potions event, but nothing further. In this lull, I check my email again.

Another email from Meredith.

F-ck.

I immediately head straight back to Lobdell. As it turns out, a few beakers (left unused) had made their way into the trash, and now two whole bags of trash needed to be considered chemical waste.

Welp. Guess they’re going in storage for Tuesday’s Adalbert to deal with.

With that done, I went off for dinner, returning to the Student Center for the fourth and final event—essentially Simon Says with a time loop.

The main lesson I learned from that? Y’all are dirty cheaters.

I spent a decent chunk of the event trying to prevent people from surreptitiously grabbing the envelope off Nero’s back. I also acted as the semaphore checker when that part came up, then proceeded to embarrass myself in front of everybody when I forgot which direction semaphore was actually read from.

And with that, all the events were done. Now to see how people were doing on the actual puzzles.

The Bringer of Jollity

Dealing with events and on-site stuff ate up the majority of my Saturday. As such, I hadn’t been keeping up with how the rest of hunt had been doing, and came back surprised to discover that only the Hydra had scored solves since I last checked.

Welp. Time for another bottle of orange soda. By this point it had already become my alcohol substitute of choice, and I definitely acted the part.

Our second tripwire was a team with seven overworld meta solves by 7 PM, and the leading team was ⛎ UNICODE EQUIVALENCE with 5. Our conclusion for this discrepancy was failing to predict solver behavior on the Hydra—we thought solvers would focus effort primarily on the 14 basal puzzles (which actually contributed to unlocking other rounds) rather than the 3 (er, 112) procedurally-generated ones (which didn’t for hopefully obvious reasons), resulting in a bunch of rounds unlocking later than expected. We started tweaking unlock values as soon as we realized teams were preferentially solving procedurally-generated puzzles, but clearly this wasn’t enough.

Our consensus was still that hunt would end on Sunday evening, but this was enough of a scare to invoke a change in hint policy: every non-meta puzzle that had been unlocked for six hours or more would be available for hints.

We sent the policy before tech had fully implemented the change, so the first few needed to be handled via email—notably, the hunt email was still in timeout jail, so we had to send it from our personal emails. I signed up to be part of the team responding to those emails. Even after tech implemented the changes, I preferentially claimed hints for leading teams.

Nothing changed in the leading team situation for quite a while after that. Saturday soon tipped into Sunday, meaning we needed to close up HQ. Right as we were about to head out for the night, the Persephone meta falls to UE. An hour later, Cardinality fells Scylla; meanwhile, Death & Mayhem (which was lagging a bit on meta solves) had unlocked puzzles in the last round in the hunt.

For a tripwire, seven hours of lead time… isn’t the worst case scenario, but it’s definitely not optimal. It’s around this time that we decide on the other major measure for cutting hunt time down: a “blessing” for each meta solve, to be given after an interaction (à la 2021), with the caveat that it must be used immediately.

We landed on this measure for several reasons:

• Teams tended to hoard free answers in 2023 (personal experience speaking here), and you can’t hoard if net intake is zero.
• It (hopefully) encouraged focusing on the meta in a way that didn’t explicitly rule out feeder puzzles from being looked into.
• We had a good theming for giving a free answer in an interaction.

By the time I woke up, the first part of Nashville had also fallen, but nothing further. When I got to HQ, we set up the blackboard to track meta solves for the leading teams.

I then see people starting to submit answers to the Hera meta. One of them soon grabbed the first solve… and they were not among the eleven teams we were tracking on the chalkboard.

I could not have hit my goal for the puzzle in a more satisfying way.

The Bringer of Old Age

At this point, though, it had become clear that hunt would drag slightly—we were still optimistic for a Sunday finish, though we really needed to make it so.

A lot of our time was spent on the big board, looking longingly at teams with rounds full of solved puzzles but unsolved metas. There were still seven left to topple, and if Saturday was anything to go by they would definitely take the rest of the day to solve. My time was spent mostly on the hints.

And at this point, from a hunt length perspective, we could only really focus on two teams: Death & Mayhem (with all the puzzles open) and Providence (who held the meta lead). UE and Cardinality, though they definitely wanted to get to the end of the hunt and definitely had the capability to, were very much unwilling to decide what the finishing time would be for us.

And so we watched with bated breath. I had run out of orange soda bottles to wave around by this point, so I went for water instead.

NYC’s meta is solved at 15. Oahu goes at 17. But at this point, something needs to change.

At around 1800, we decided to send out a feeler to the two prospective winners, both to check in on their general condition, and perhaps accelerate their progress a bit by giving them unrestricted hints on one (1) feeder puzzle.

We visited Providence first, since they were closer to HQ. They already had the main epiphany on New York, were slogging through Duet, and were starting to see things in Gaia. They were also relatively enthusiastic about potentially winning Hunt. Nathan gave hints for Building Blocks, I looked around the room at all the printouts, someone recognized me as that one solo solver, and we moved on.

Death & Mayhem were in Building 24 for the Hunt, which was only accessible from outdoors. I was surprised to find out it had started snowing—kind of like when the same thing happened last year just after we found the coin. Perhaps Gaia had also expected a Sunday evening finish.

Anyways, we got to D&M, who had just finished arranging the Gaia stars in the correct manner for the meta, were on the verge of a major breakthrough on the Minotaur, and… had fallen down a rabbithole on the first Hermes meta. (They got out of it later, but it took them a few hours.) Midway through the briefing, we were interrupted by a phone call… which was actually Sara calling back the answer for Radical Fishing Trip. Classic comedic timing.

Anyways, D&M asked for a hint on IO. Given that none of us knew enough about it to confidently answer every question, Bella went off to read the solution in private while I went off my working knowledge from having sent a hint on it earlier. We were essentially trying to guide them to what precisely the gates meant. Bella then came back, and I stepped out of the conversation a bit… before we both then had to leave the room to reread the solution. We eventually guided them to the step just before the final answer.

Anyways, return to HQ, more of the same nerve-wracking staring.

We get a Sedona meta solve at 19. This was the round we were worried about difficulty-wise, but it seems things went just fine.

Texas was solved at 20.

Washington at 22.

Nashville Part 2 at 23.

And just past the stroke of midnight on Monday, Nashville Part 3.

The Magician

Seems like a good thing, right? Every meta, solved at a relatively reasonable clip, which meant we had a good pace and were just off by seven hours somewhere in our calculation. Where’s the problem?

Well… the four leading teams solved the last few metas in as close to a Latin square arrangement as could be managed.

The good thing about this is that this affirmed the puzzles were doable cleanly, and that’s a very good thing to know for hunt. If one team can solve it, others will follow. The bad part is that we… kind of needed these solves to be concentrated on one team for hunt to end at a reasonable time. It also severely limited our options for nerfing, because we couldn’t just add “errata” on the puzzle and call it a day.

We pulled some more stops to try to get one of those teams over the finish line in time. Eventually, both Providence and D&M had every feeder answer in their unsolved rounds, making it a true meta race. And now we could but wait.

We see some movement. D&M solves Gaia. Moments later, Providence gets Texas. An hour later, Providence follows up with their own Gaia solve, and D&M gets Sedona.

1 AM came, and we had to vacate the Bush Room for our off-site HQ. This didn’t stop our eyes from being essentially glued to the guess log.

D&M gets Duet. They already thought that they knew about every puzzle in the hunt when they finished Mercury, so I can only imagine their reaction upon seeing yet another unanticipated puzzle in Nashville. Fortunately, it was their last.

Both teams are now down to two metas. The team’s sheets are shared in the editor’s channel. Great, another source of cortisol. Just what I needed.

In the meantime, some of us leave the HQ to go do the Texas free-answer interaction with Cardinality, who were headquartered in the ballroom on the other side of the floor. I go with them, if only to take my mind off the race, pecked at their bird plushes with mine, and then simply watched as they free-answered Modern Architecture to the author’s face. A very reasonable one to grab at this point, but it still stung a bit.

On the way back, I suddenly realize that the entire hallway has been plastered with “no birds” signs. While I knew it referred to Cardinality, I couldn’t help but feel offended.

Nathaniel then calls on me to respond to a Transcendental Algebra submission from Order of Random Puzzlers, who submitted a nine-language dialogue. A brief spot of levity in a very stressful situation, as I start talking in Standard German and am immediately met with an unintelligible language that… is also German.

Providence then gets a breakthrough on Sedona. They now had just one meta: Nashville’s third. D&M had just gotten here, and also had Aphrodite to contend with.

At this point, if we wanted to end hunt as soon as possible, we could have just directly made changes to A Night At The Opera without much problem—the only team that had gotten it at this point was ⛎ UNICODE EQUIVALENCE.

We asked teams if they wanted hints. D&M said yes. Providence said no. So we kept watching. Providence was now trying several different theories, none of which were the correct one.

But before Providence could say yes to hints, D&M suddenly switched stances. They had just broken through on A Night At The Opera, and both teams now had exactly one meta left. Different metas, meaning the option of just updating one meta was now out the window.

It was now 0400. Nerve-wracking.

We watched this for another half-hour. D&M was now piling onto Aphrodite, and got it unstuck from its position. We saw half of the meta answer pop up… and watched them not pursue the idea further. We soon realized they had completely forgotten about the existence of last names, and they probably wouldn’t pursue the theory again given half of their indices were busts. Providence was still on their theorycrafting train—they were considering the right idea (what they called the All White Theory), but probably didn’t have enough confidence to go all-in on it.

Meanwhile, the other teams were catching up. UE was now also waiting on one meta, with other teams not far behind. Teams with a very clear “we do not want a coin” stance were now within striking distance of the coin.

If we gave the leaders hints right now, it would all but force a D&M victory, since they were significantly closer to their meta answer. If we didn’t have the metaphorical Sword of Damocles to deal with at that point, I would have assumed someone else had their hand on the wheel here. But we had to make a move as soon as possible.

We sent out a message to the leaders, saying hints were coming. We made sure to time the hints to send simultaneously and ensure they were of the same quality (which essentially had to mean “enough to finish”, given where D&M now was).

It’s now 0500. We send the hints. Now we wait.

0503. Death & Mayhem solves Aphrodite and unlocks the capstone.

0513. Providence solves A Night At The Opera and unlocks the capstone.

Now we wait.

…

……

………………

0521. Death & Mayhem solves the capstone. Much elation/relief is had. We send the message about the coin.

0531. Providence solves the capstone, in a slightly shorter timeframe than Death and Mayhem.

We then realize that we didn’t successfully send the message about the coin, and thus likely caused several minutes of breath-holding anxiety for both teams. Whoops.

The Mystic

Regardless, it was over. Seven more hours until wrap-up, and then I could relax.

A bunch of us leave for the Bush Room at 0600, to prepare for the upcoming runaround at 0730. I stay behind for a bit to finish up my slides for wrap-up, planning to leave at 0700.

Unfortunately, I forgot that the Bush Room was now card entry only.

After unlocking the Bush Room and relaxing a bit, I was sent ahead to the starting room for the runaround to buy time for the gods to prepare. I arrived to… a less full room than expected, but I did have an idea in my mind for some ad lib.

For the duration of hunt, I carried around an eagle plush on one of my gloves, ostensibly for moral support, but also to make myself a little more identifiable. I started to make some small talk with the few D&M members in the room, acting as if the eagle was secretly Zeus (a god not present in the rest of the hunt) in disguise. This went alright, though I didn’t quite give that particular part of the subtext. Until now, that is.

Then the gods actually arrived, and I could simply watch the runaround unfold.

Nick took the framework and turned it into something extremely fun. Ciphered War for Ares was something we agreed on pretty early, but the other three were completely new to me. Puns aren’t my fare, but I can kind of make an exception for Demeter. Hephaestus was a nice homage back to the runarounds of old that tested knowledge of feeder answers. And Athena was probably the first time I reacted even slightly positively to a “deez nuts” joke²⁷. Also it was charades with bird calls. How can I not like that?

And finally—finally–we could pass on the mantle of hunt. Congratulations to Death and Mayhem—looking forward to 2025.

The next three hours was mostly passive idling—first laying down for a bit to recover, then meeting up with the A/V tech to set up for wrap-up, then trying to watch the other finishing teams go through their runarounds.

Then everyone filed in for wrap-up. I was moderately tired at this point, and my general oral clumsiness was much more prominent here than during kickoff where I had a script to read from.

I felt very spent afterwards, especially since I was officially running on 24 hours of waking time. But I had officially helped with running a Mystery Hunt.

---

After wrap-up, I met up with a bunch of Puzzle Club people at the student center. I was too tired to register any of it, and I’m pretty sure I fell asleep partway through. If I had to guess, more kibbutzing about the coming year.

Then I went back to HQ to clean up the last of my stuff. This was mostly about the box I toted most of the hunt paper around in, but I also made sure nothing else of import was left behind.

Afterwards, I got “lunch” with other TTBNL folks. I also hardly remember any of that conversation. I said my formal goodbyes to the rest of the team shortly afterwards and walked back to my dorm.

I fell asleep at 1700 that day, and my sleep schedule hasn’t recovered since.

Lascia ch’io pianga

This is the opinion segment. The opinions contained within may be inflammatory and controversial (read: different from yours). Click here to open, at your own peril. Don't say I didn't warn you.

I’m sick of The Discourse.

I’m not here to talk about how long a hunt should be, or how much the MIT affiliation actually matters, or whatever other thing Mystery Hunters like to argue about these days. I’m here to talk about how people express their opinions. If you came in expecting my opinions on the current hot topic for the Discourse, I’m not sorry for disappointing you, and I’m also not sorry for forcing you to listen to the Curb Your Enthusiasm theme for thinking I had anything substantive to say on your topic of choice. Yes, I’m calling you a clown. It’s the least you deserve for inflicting the worst scourge imaginable on eight generations of Mystery Hunt writers.

I know of several forums where people conduct The Discourse. I avoid The Discourse-y parts where I can, because observing past iterations of the Discourse has really cemented in my brain how unproductive a conversation it actually comes to be. Unfortunately, other people don’t, and I get exposed to these comments by proxy. Let’s just say that people aren’t too guarded about their word choice when it comes to being negative.

No, I’m not immune to being this pointed with my criticism, either—I have strong opinions on puzzle design and tend to excoriate puzzles that stray from them by a significant margin (see: last year). But The Discourse is never about bad puzzle design; it’s always about a bigger picture than individual puzzles. Things like logistics, where you have a million potential points of failure that are essentially impossible to triage or categorize in any meaningful way.

From what I’ve seen of The Discourse, it also has a rather vocal set of people who imagine everything that goes “wrong” with hunt (for whatever definition of wrong that happens to be) is an intentional decision on the part of the running team.

It’s not intentional. It functionally never is.

Even the most well-intentioned people can’t predict every possible effect of the decisions they make. But no one judges based on what was known in the heat of the moment. They always judge in retrospect, where the “objectively correct” decision is so clear the runners must have been blind (or worse, willfully ignorant) to not have done this.

I’ve experienced particularly grim reminders of this fact on discussions of hunt length, where the number 237 has essentially become weaponized into the equivalent number of rotten tomatoes. As a side note, a quick back-of-the-envelope calculation finds that the number of non-fish puzzles this year (187) is very close to the total number of puzzles in other years (notably, 2022 had 182), and we figured that adding a standard fish round on top wasn’t going to change that by more than a few hours (notably, 2022 ended some seven hours “ahead” of schedule). Not to mention that puzzle count is functionally meaningless in a hunt length calculation—that involves so many layers of integrals even Cauchy would have trouble with them all. Not to mention that summarizing a year’s Hunt into one number, whatever the metric, is horridly reductive on so many levels it puts the entire fucking standardized testing industry to shame. More and more I feel like this number is the only thing people remember this hunt for—it’s brought up in every context that vaguely reminds people of it, recency bias notwithstanding, and the only recourse I have for airing my displeasure with the subject is making an extremely visible point of walking out of the room (which you might notice is rather impossible in the online setting most of this takes place in).

I’m reminded of Shostakovich’s definition of a musicologist: a person who neither composes nor performs music, but talks about it nonetheless²⁸. It’s tempting to draw the more obvious parallel, but solving the puzzles does not suddenly mean you know enough to comment on the writing process, just as listening to the music does not grant you any direct insight into the composition process. A lot of The Discourse around the hunt process falls into this category, with results to match: they fail to be meaningful reviews of Hunt. They’re a denigration of the effort that gets put into it and an insult to the people who sunk an entire year into its creation. If the goal is to maintain the long-term health of hunt, as I’ve heard many people justify, The Discourse is the most actively counterproductive thing you can do with your time and words.

I’m not asking you to ditch your opinions. I’m asking you to ditch the nuclear football of a flamewar that comes with the opinions. This way of expressing them is steaming hot garbage. This shouldn’t be controversial—no one wins a flamewar. Everyone leaves unhappy. And by now I’ve lost faith in everyone’s ability to do so, and will probably continue to believe as much until those opinions stop existing. Of course everyone’s going to keep talking, because everyone wants The Hunt to fit their ideals.

And speaking of ideals, you may have noticed I put “objectively correct” in air quotes. That’s because “objectively correct” doesn’t meaningfully exist in philosophy, and especially not hunt philosophy—it’s entirely a manifestation of other people’s desires for hunt. And like any hobby, everyone is going to have wildly differing opinions. If the writers for one year decided to Increase The Fucking Thing out of spite as a reaction to the abysmal state of The Discourse, that’s just what they’ve decided to do, and we should respect that decision, minding the fact that this will likely influence the next writing team’s philosophy as well.

Hunt is a learning process: the writers build their philosophy incrementally with each hunt they experience. There’s no magic formula for what makes a hunt “good”, so the best we can do is identify heuristics. We saw 2023’s main difficulty as a high average puzzle difficulty (which is incidentally a length contributor) and generally being less forgiving to smaller/newer teams. We also had a few points based around our personal experience with hunts in general. We tailored philosophy to match, and judging by the results, I say we hit our goals pretty much on the mark. Whether or not D&M saw the number 238 and thought that raw puzzle count was a problem for 2024’s hunt, it’s now their prerogative to set the number for 2025, whatever that number happens to be. Not yours.

“But if Discourse is garbage, then how else can I stump about how I believe hunt should be run?”

You want a better way to express your ideals for hunt? Easy, it’s baked right into the ruleset—Win The Hunt, and write one with your philosophy at its core.

Gasp! An unpopular opinion? In an opinion piece? More likely than you think! And I will stand by this, given this is part of what I wanted out of writing for hunt.

My Mystery Hunt experience starts in 2021, where I hunted with a medium-power team local to MIT (though at that point less local due to a certain Big C that shall not be named). No notes, fun all around, writers absolutely nailed it. Having thus calibrated my expectations for the hunt, I chose to hunt with the same team for 2022. We spent Friday sweeping past the intro round, so the rest of the weekend seems bright, right?

Saturday was an Afghan quagmire, featuring us getting stuck on a small number of puzzles in the first two rounds. Relief rounds didn’t help the matter—progress became glacial, and there was nothing to be done until the early Sunday coin unlocked… two puzzles in each round of the rest of the hunt for us. (We still only solved a small number of those.)

Having now had two drastically different experiences of hunt, I could now properly define my personal metric for how fun a hunt is. At around the same time, I also grew to have a desire to write for the hunt—leave a personal mark on the history of the event responsible for the largest proportion of my good experiences in the last few years, and especially to do it while I was still at MIT. Both of these factors led to me seeking out a larger team for 2023. So I joined TTBNL, saw more of the hunt, won, and edited for 2024. Obviously we’d want to make the best hunt we can, and we made decisions accordingly. A year later, we had a hunt and were all decently happy with the result, even as we struggled with its (lack of) momentum through Monday morning.

If you think there’s a systemic problem with hunt that you would like the power to correct, Be The Change You Want To See. Win The Hunt, or caucus with a team that has the capability to do so. Make the eggs, instead of grousing over a taste that has never reached your tongue. Armchairing about it will get you nowhere. Think of it like legislating: protest and other forms of protected speech²⁹ might be a loud and visible way to express ideas, but in the end the lawmakers aren’t beholden to listen to it. The best way to enshrine your vision in the books is to be at the table yourself, writing the laws and engaging in the process.

And if you really don’t want to win the hunt because writing it is a “burden”, don’t come into the hunt expecting your sense of “objectively correct” to be in the winners’ philosophy just because you spilled enough ink to dye the Tigris River black³⁰. They’re the ones who earned the right to shape Hunt to their will. They should decide. If the things they decide on are incompatible with your views, c’est la vie.

As for me? I’m here to do puzzles and have fun doing puzzles. It’s why, of the three Mystery Hunts I’ve taken part in live, the 2022 Hunt (where the Coin was found on Sunday morning) actually ranks lower in my experience than the 2023 Hunt that everyone pans for its length and difficulty—I can’t remember a single puzzle I had strong feelings on³¹ before the coin was found, they unlocked all the rounds for everyone, and I could actually see Midterm of Unspeakable Chaos all the way out in city number 7. To what extent this is the result of bottlenecking or Palindrome’s writing style not meshing well with my brain, I dare not speculate. But this issue is explicitly why I linked up with a larger team last year, and less explicitly why I made a seemingly contradictory statement about liking to solve with medium-power teams in my first year. They were both about seeing more puzzles—volume, not percentage—because that’s a good heuristic for how much fun I have with a hunt.

Essentially, my request for future Hunt writers is “make sure I can see the fun puzzles you put effort into”, which is something I can already trust most hunt writers to do well. Besides that, go wild. We wrote the hunt we wanted to write, now write the hunt you want to write.

This is MIT. We thrive on free thinking and expression, and so does the Mystery Hunt.

If you think this is negotiable, you can strike puzzlehunts off your list of hobbies, and the world will be better for it.

That’s the most animated I’ve ever gotten about hunt, and you can tell because I failed to censor the literal two f-cks I gave to The Discourse. Let’s move off the topic. Preferably in perpetuity, but again, I know people don’t know when to stop.

Der Dichter Spricht

So how did I fare? Overall, great!

Puzzles

• Among my puzzles, the three feeders were consistently solved less than average for their round. In particular, both Transcendental Algebra and Modern Architecture were the least-solved puzzles within their rounds, though Asphodel is beaten by Do You Like Wordle?. They were also rather common targets for free answers.
• I heard a lot of praise for TransAlg specifically. Myriad Gamut also got some, and at one point I noted that the author of the Juno puzzles had changed his nickname in one of our shared Discord servers to the modification of his name I threw in as a callback. I was worried about stepping on his toes when I wrote that in; seeing that recognition was simultaneously relieving and elating.
• At least two people I know have gotten into Spiritfarer recently. Whether that’s a direct result of Asphodel, I cannot say.
• I knew going in that Modern Architecture would have the Playful Sounds problem of one minipuzzle being significantly meatier than the others. This was true in testsolving, and it definitely was the same in practice—as far as I can tell less than half the teams that got to it actually extracted the subanswer from JFK, which was about what I expected. I definitely could have expanded it into a larger puzzle, but decided not to because its subject material was already covered last year³² and there are only so many interesting things I could do with the retrograde analysis—I showcased as many as I could in the puzzle and added a few gimmes to fill the gaps.
• I’m still a bit sad about the one puzzle that didn’t make it, but any lingering feelings I have about it are water off my back at this point—I achieved most of the goals I had for it with TransAlg, and I wouldn’t have it any other way.
• Among the rest of the puzzles in the hunt, I’m pretty happy with the variety and quality we’ve achieved. I’ll leave suggestions for good puzzles to others, but here’s a sample platter of the ones I liked:
  • A More 6 ∪ 28 ∪ 496 ∪ … (probably the puzzle that produces the strongest sense of dread in the whole hunt.)
  • A Radical Fishing Trip (yay for non-butchered language puzzles!)
  • It’s Not A Well Written Quiz Bowl Packet But… (at least nothing went horribly wrong!)
  • The Linq (the obligatory recommendation)

Logistics

• Nothing went too catastrophically—though admittedly some things were very close to not happening. If you can make something like Circe happen, by all means go ahead, but be sure to do it right or all of MIT’s administration will give you hell. We’re still dealing with the fallout of that one, despite trying to be as thorough as possible with it—apparently we talked to the wrong division of EHS? Bureaucracy is hard.
• Running around and seeing teams (whatever the reason) was fun! And a good stress reliever from staring at the bigboard and fighting the standard logistics fires. Sorry I didn’t make visits more often—I had other things to worry about over the weekend. If there’s one thing I regret it’s not swinging by Simmons on the way back to my dorm to give some visits to comparatively far-flung Wafflehaüs. In defense of that action, though, I was prioritizing sleep.
• Events were fun! The events for the virtual years were likely heavily constrained by the medium, but I didn’t find them as engaging as the events I had heard about from past years. Attending all four of this year’s events reminded me why I found them so alluring.
• Generally, everything about running the actual event, seeing the culmination of a year of effort, was fun! Maybe I could have invested more effort into the team-facing side of things, but I liked what I had.

Sanity

• I survived!
• Would I write again? Probably! (Especially since I won’t necessarily be the one dealing with MIT this time, and having expertise makes this a little bit easier!) Just give me some time to recover. Definitely won’t aim for next year. Knowing Luck/TTBNL’s track record, expect 2031, though maybe I’ll be swinging around other teams in the meantime.

And in addition to all of this…

The Carnival of the Animals

…I got more puzzle furry friends to come to ANE after doing Mystery Hunt! Talking and interacting with them was a fun time, and a good note to end this puzzle-year³³ on.

Unfortunately, this is probably the last year in the near future that this will be possible, because next year’s ANE date lands right on Mystery Hunt’s weekend, and short of Mystery Hunt secretly hosting an event at ANE I don’t think I can feasibly be at both for another few years.

I’ll miss y’all. See you at hunt next year.

---

1. ‘We’ being the Team Formerly Known As the Team Formerly Known As the Team Formerly Known As the Team Formerly Known As the Team Formerly Known As the Team Formerly Known As³⁴ the Team To Be Named Later. I am abbreviating that to TTBNL for the rest of this, because Armadyl forbid I write that out a second time. ↩

2. Spoilers: we got exactly one corner and realized teammate took one jigsaw and scattered the pieces among all the boxes given out to teams. ↩

3. Anthro New England, the Boston-area furry convention, which was very conveniently timed to be one week after hunt both this year and last year. ↩

4. Apparently teammate picked their theme just as early, but this was still ahead of the curve. ↩

5. Fresh memories of my 2022 with ET Phone in Answer here. I’d say we nailed it, given the small team feedback I’ve heard thus far. ↩

6. Probably not for another six or seven years, I say, given Luck/TTBNL’s hunt record. ↩

7. There would have been a fourth class, but there was a schedule conflict and the intended replacement could not accept more students. Three classes with final projects is load enough in my eyes. :pain: ↩

8. The MIT Strategic Gaming Society, which holds meetings at Walker Memorial that I attend basically every Friday and Sunday night. What can I say, I love my board games, and leaving Walker from the north door you get a side-by-side view of three particularly relevant buildings. ↩

9. The way this puzzle worked meant it could only accept English puns that were expressible in Greek roots. The fact that I found what would become the final meta answer at all was a bit of a miracle—a Hera connection AND a number/letter connection? Jackpot! ↩

10. Pun intended. ↩

11. Hell, even “avid”’s selling him short. He was a bright mind the likes of which the world had never seen, and likely will never see again. No words could do that reputation justice, though if anyone could that anyone would be Jack. I could go on, but several other people have already stated it more eloquently than I could. ↩

12. Not thoughts of the suicide variety. ↩

13. MSPH sniped a truly annoying number of our meta ideas, and it wasn’t just because it had a Classics Round (Hydra and Hole to Hades both suffered from this). It’s a whole other level of insanity. ↩

14. intense judging stare at the Student Center Meta ↩

15. 215 does not include 112 procedurally-generated Hydra puzzles. 215 + 3 types of Hydrae + 19 metas gives the official figure of 237 used for wrapup, though notably this number does not include the capstone. ↩

16. Still did, judging by what hint requests I saw. ↩

17. The invoice was also missing half the hours I requested, meaning the number I got should have been doubled for all intents and purposes. ↩

18. This would have taken a miracle to time correctly, so I’m a bit thankful that we didn’t go with this idea in the end. ↩

19. Ironically, Zeus and Athena were supposed to have more important roles in the initial plot pitch, but Athena’s role of “humanity’s advocate” got passed to Hermes down the line, and Zeus’s role was the smiting, which got passed to Olympus as a collective. ↩

20. Both Modern Architecture testsolves primarily solved based on the three easier minipuzzles. The first avoided JFK like the plague. The second got it reasonably clean, and Bella then got a friend of hers to testsolve that segment. We also tested the second part of National Gallery on-site, since the remote variant was a little bit intractable with my level of camera quality (and we sped them through that part). ↩

21. Only sodium and lithium were there, by the way. Almost like they tried to have a whole periodic table and then gave up halfway through the alkali metals. ↩

22. Non-apology  ↩

23. For some reason, the Bush Room was booked for all of Friday until 8:30 PM or something. I have no idea what could have been going on in there. ↩

24. Seven zero-width spaces, written as zero zero-width spaces because I can’t be bothered to paste one in. ↩

25. Totally not foreshadowing for later. ↩

26. All three printers on the Infinite Corridor print in color, and yes I’m that stingy with my money. ↩

27. Thanks, Kayton. ↩

28. The actual definition uses eggs as an analogy for musical composition. ↩

29. Key word: protected. The author of this piece does not condone sedition or violence. ↩

30. If you have, by the way, shame on you for both kinds of pollution. ↩

31. Maybe I should postsolve it again. I barely remember anything from that hunt in comparison to the surrounding years, but that is a tall bar to clear given what the surrounding hunts were. ↩

32. Yet another reason why I hate Business Cards with a burning passion. A lot of my decisions on the JFK Library subpuzzle were made to showcase its wasted potential. ↩

33. The puzzle-year is an arbitrary division, mostly defined so my year ramps up towards Mystery Hunt instead of peaking in January. ↩

34. That’s six TTFKAs. ↩