1998 MIT Mystery Hunt - What is Known

The 1998 MIT Mystery Hunt is mostly missing. The page where it would normally be really only has the errata. There is also an earlier version of this page within the web site which does not have the erratum for puzzle 18.000 in Junior Year.

We also, from stories that have been recorded about the endgame, know that the coin was found in an elevator shaft in building 4, at floor e.

From the errata page, we know that the hunt was based on taking fictional MIT classes, divided into four rounds corresponding to the traditional undergraduate years. There are some other details we can infer from the errata, listed below.

We do, however, have two puzzles, or parts thereof. And thank you, Thomas Mack, for your recollections! But first, those two puzzles:

19.154-62-3522 Seminar in Enigmatology

This puzzle appears at the bottom of the page linked above, or at least part of it does. What we have is reproduced here, in case it disappears:

For those of you who overslept, here is the (Songs) puzzle from the seminar:

Each of these phrases is part of a song title. Specifically, each of the phrases appears in parentheses within a song title such that it may be considered a subtitle, alternate title, or optional part of the title. The full titles are:

* This song by the Greg Kihn Band is actually just called "Jeopardy" but it's clearly the one referred to. This could be an indication that they were trying to shoehorn in a J they needed for extraction, or they might have been influenced by Weird Al's version "(I Lost on) Jeopardy".

The songs are given in alphabetical order of the full titles (sometimes starting with the given words), which suggests we need some other ordering mechanism. Most likely, there was content provided to teams along with the other puzzles which are now missing.

Two ordering mechanisms that don't work are:

The puzzle number is unusually long and so it may be relevant to solving the puzzle. One idea is that a second step leads to nine songs or words, and the digits index into those titles or words to extract a three-word answer. (The 19 refers to Course 19; see below.)

9.ZZZ Abstract Dream Analysis

This puzzle appears, unlinked, within the MIT site at https://puzzles.mit.edu/1998/dream/ and within this Reddit thread a Hunter from that era confirmed having a copy of the puzzle saved on disk after all these years, so it was likely part of the hunt. Again, in case it disappears, the puzzle is reproduced here:

  1. 2------ --- -5----4
  2. --------1 ---- ------6--
  3. 4--- --- --1--
  4. --7--5- --- -- --- 6----
  5. --7----- 6-----5--7
  6. -1- --- 2-- --5 -- --- ---
  7. ---- -3-- --7 2-4--
  8. 3--7-, -6---- --4 ----
  9. -3--- -- --7-
  10. -5--- --- ---6-
  11. --- ---3 -- -45----
  12. ---6-5--- -1---
  13. 3----7- --- --- 5---- ---- -- ---2

In the source code of the page, at the bottom, is an HTML comment. In those days, it was common to provide hidden hints to a puzzle in this way. This one says: What are you doing reading this? Go watch the movie!

One theory we have on this one is that phrases from some dream-related movie fit the blanks a la Wheel of Fortune. The digits each hit the same letter consistently (and perhaps help to extract the phrases from the script of the movie by doing so) and assemble in order to spell the answer. A different variation of this is that they don't match, but the phrases should be a more obvious set, but the letters represented by each number go together in the order given to spell a seven-word clue, which would have enumeration 4 3 5 5 8 6 7.

To date nobody has identified a movie that works. Keep in mind this puzzle was released in January 1998, so The Matrix, Being John Malkovich, eXistenZ, and Inception had not been released yet.

Another idea is that the titles of dream-related movies go in the blanks. However, entry 8 matches few film titles and I could not find any for entry 13 and only one, a movie from Hong Kong released just months before the hunt, for entry 6, which would not reasonably have been findable for teams with the resources of the day.

The Other Puzzles

We only have the titles, numbers, and brief errata for other puzzles, but we can deduce a few things (in addition to the subjects and/or types of some of the puzzles).

In some cases the errata appear to be open hints rather than corrections.

In some cases we have only a number or only a title. It's not clear whether there were puzzles which had only one or the other, or if this was just due to haphazard construction of the errata list. The fact that some puzzle titles in junior and senior year are written in all lower case rather than in usual title case, with the corresponding hints all in lower case, leans toward the haphazard.

The Overall Hunt Structure and Metas

Thank you to Thomas Mack who sent in these recollections:

As noted above, the theme of the hunt was getting a degree in Enigmatology from MIT. There were four rounds corresponding to the years of a four-year degree.

The hunt authors had intended the meta answers to spell out IVY LEAGUE GOLF COURSES but switched the last two rounds.

Puzzle Numbers

The part of each number before the dot corresponds to MIT course numbers (what would be thought of as majors or departments at most universities), and they seem to correspond with appropriate ones. For example, 6.666 Arcane Programs would be part of course 6, Electrical Engineering and Computer Science, which is where programming courses would be found. A bunch of them are in Course 19, which doesn't exist. There used to be a Course 19, Meteorology, but after the 1983-84 school year it was folded into Course 12, Earth Sciences, and by the time of this hunt, it had already become part of the MIT lingo that Course 19 referred to MIT-style hacking, and the authors used it probably wherever another MIT course number didn't fit.

The part after the dot seems to have no obvious pattern. In some cases, as with 6.666, it seems thematic for the puzzle (suggesting the "arcane" puzzle has a link to Satan). Likewise, 4.4444 is a puzzle based on Tetris, so it has four fours here, referencing the 4 squares in each Tetris piece, and 9.ZZZ, a puzzle about dreams, has the letters used to represent a snoring sound rather than numbers.

In Senior Puzzles, 19.333 Chaotic Clues and 19.667 Grid Puzzles likely go together, seeing how their decimal parts are the approximations of 1/3 and 2/3, and their titles suggest the two parts of a crossword puzzle.

Rhodography's erratum says to note the corrected subject number (7.51896), suggesting it was important and not just for theme, unless it was the course number that was corrected. This and some of the other numbers seem unusually specific, suggesting they could have been used for extraction or in metapuzzles. This was a Freshman year puzzle, which clearly needed an ordering mechanism, so perhaps these numbers were used for ordering.

Other Details About Specific Puzzles

7.51896 Rhodography - Rhodography isn't a real subject, but rhodology is the study of roses, and the French word rhodographie means a treatise on roses, so this puzzle probably had something to do with roses. Course 7 is biology, which is appropriate. Thomas Mack explained that this was a jigsaw puzzle that solved to an image from a comic strip with part of the dialogue blanked out. It turned out to be too hard to find a copy of this strip in 1998, so the answer was released. Since it should be related to roses, it is likely the comic strip was Rose is Rose, and with modern tools, we can find what was too hard in 1998: The May 18, 1996 edition of Rose is Rose, which includes the term NUCLEAR REACTOR.

19.7 Cryptography - The third "cryptogram" is not solvable. This had to be on purpose, though without seeing the puzzle - and without knowing whether there were three or many cryptograms in the puzzle - it is hard to have a good idea why. Was it a metapuzzle for the group which didn't work like usual cryptograms? Was it a red herring meant to waste solvers' time?

6.666 Arcane Programs - The puzzle was replaced with a text file. In these days, HTML was quite new and some pages were coded by hand, by people who might not have been familiar with all the syntactic rules, so one possibility was there was a computer program in the page and its special characters were not all escaped properly, and rather than try to fix it, they simply pasted the program into a text file where it would not be troubled by those rules.

8.137 - The erratum for this puzzle mentions a thesis, which you don't need to read "a single word" of. Thomas Mack explains it was the thesis of one of the hunt authors. He encoded a binary message in the bound copy of the thesis kept on file at MIT by flipping the watermark on certain pages. Teams were directed to the thesis without any information of this sort, so the erratum was meant to close off many blind alleys.

19.1289 Decoding Advanced Cryptography - Was this a numeric code of some sort?

19.789 Color threory - Given that there were three colors and three font sizes, it's possible there were also three of another thing and the spelling of the title was a hint this was a ternary cipher.

18.123 - We don't have a title here, but the number 123 suggests "easy" or counting, and the fact there was a wrong number suggests another numeric code.

4.123 Coloring Within the Lines - If the given text is a hint rather than erratum, it is possible this is a nonogram. If it is an erratum, then maybe some other puzzle where you have to shade squares given some initially shaded ones. It is in Course 4, Architecture, so a puzzle where you are building something (or drawing a picture of a building, perhaps) would be appropriate.

18.617 Number Theory - Looks like some sort of numeric trivia. The one about the book might have wanted a call number.

4.444 Structures - Explicitly stated to be about Tetris. It sounds like the pieces had letters on them and you were spelling out a message upon assembling them. If one of the blocks had ERE'S then it probably meant the word THERE'S, WHERE'S, or HERE'S was in the message. (Thomas Mack confirmed solvers were given a bank of Tetris pieces and had to fit them in sequentially to spell out some sort of message.)

6.535 roamin' - The title is a reference to Roman numerals, in addition to travel. It was in Course 6, EE-CS, because it was an Athena cluster runaround. Thomas Mack explained that this puzzle started out with a long number. You had to convert it to letters, and then read the letters as Roman numerals. This gave a phone number, and when they called it, they were sent to an Athena cluster. When teams got there, they got a phone call from headquarters telling them to be at another cluster within a time limit. This was repeated with several other clusters with progressively shorter time limits, with the call in the last one telling the team they could find the answer on a bulletin board in the room.