Has anyone witnessed anything remotely like a puzzle hunt at Burning Man?
I am considering this rather quixotic endeavor. If you find the concept interesting/upsetting/lulzy or have relevant experience, drop a note here in comments or to me directly, and I can tell you more and hear your thoughts. (Ideally, you’d talk me out of it.)
Nobody talked me out of it. Instead, a few Burners and puzzle hunt enthusiasts joined up: Myself (Dan Egnor), Sean Gugler, Jesse Morris, Ana Ulin, and Glenn Willen. We had all been to Burning Man and most of us had run puzzle hunts, but none of us had ever worked together.
It was quite an adventure.
Puzzle hunts are community-oriented mind-bending adventures known for puzzles where figuring out what to do is half the challenge. Burning Man is a dusty festival known for grand art, wild parties, and general exuberance. None of us were sure these things would mix well, but we hoped they would.
Try it yourself
Before I spoil the details, you might want to try solving the hunt.
Each puzzle had a “guide sheet” acting as introduction and directions. During the hunt, we used electronic keypads to check answers; here, you’ll have to email us if you want confirmation (or hints).
Amazing Sideshows: Guide, 1, 2, 3 (solves to a word).
Creative Differences: Guide, 1, 2, 3 (7 digits; was BUTTERY on the keypad).
Mirror, Mirror can’t be solved at home (but solved to BONK).
Target Practice: Guide, puzzle (solves to a short phrase).
Unholy Union: Guide, 1, 2, 3 (solves to a word).
The meta: Guide, puzzle (solves to a location in Black Rock City).
Read on for the full story. Spoilers ahead!
The basic concept
Schedules are loose at Burning Man, and we wanted to capture the attention of people who wouldn’t go out of their way for a “puzzle hunt”. So, instead of advertising an event at a specific time, we made installation puzzles designed to intrigue wandering visitors. Each installation included a lock-box; solving the puzzle allowed visitors to open the box and claim a prize. That way people could have a good time even if they only saw one puzzle.
Four of our six puzzles followed the same physical format, which is a triangular billboard-type sign. (The other two had puzzle-specific construction.) Puzzle content was printed on 4-by-8-foot vinyl banners and attached to 4-by-8-foot sheets of plywood.
For the final build, we trucked the lumber to TechShop for marking, cutting, and drilling, then loaded it back in the truck for the trip to the desert.
Prize lockbox design
For simplicity, we used an ordinary electronic safe bolted to a small table. Every puzzle solved to a word or short phrase; each keypad was relabeled to include all the letters in the answer with some random other letters.
Inside the safe was a light, a bucket of prizes (small pin buttons), and a log book where solvers could leave a note if they wanted.
On top of the safe was a general information sheet for the hunt.
In front of the safe was a small solar spotlight and that puzzle’s “guide sheet”. Guide sheets were designed to help solvers gain a foothold on the puzzle while preserving the intrigue of a “mystery puzzle”. Banners and other installation materials had to be ready weeks ahead of time, but guide sheets could be modified at any time — we re-printed one during the event to remove an unintentional red herring.
I reprogrammed the safe’s controller to avoid the 5-minute lockout after three incorrect tries, to allow longer combinations, to open the safe immediately after the correct word is entered (rather than requiring an explicit “open” key), and to play happy sounds when the safe opens. For one puzzle, the chip was modified to accept two variations on the answer.
Puzzle design and playtesting
After brainstorming and testing among ourselves, we settled on six puzzles to develop and test for the hunt.
We conducted two major playtest events.
The first event was held at home, and we invited Bay Area puzzle community members. We set players loose on mini versions of the installations scattered around the house. While neither a representative audience nor a realistic solving environment, this test was very helpful. We found major sticking points in “Creative Differences”, “Mirror, Mirror” and “Carnival Zodiac”, and smaller glitches in almost every puzzle.
The second event was held at a “Backyard Burning Man” event in the Boston area. We used slightly fancier mini versions of the installations. This test was also very informative, especially since the participant mix was closer to our target audience. We were able to validate changes made after the first playtest, and discovered more issues with “Carnival Zodiac”.
As we revised the puzzles, we sent PDF versions to friends and community members for test solving. Their quick feedback was invaluable.
Our greatest thanks go out to everyone who helped test these puzzles!
We were finally ready to head to the desert and run the hunt.
“Amazing Sideshows” puzzle
The art theme of Burning Man 2015 was “Carnival of Mirrors”; many of our puzzles followed this theme. “Amazing Sideshows” (by Ana Ulin) had three signs intended to resemble vintage carnival sideshow advertisements. Each sign used a different technique to highlight letters, in a way that was clued by the content, leading to a final extraction directive yielding the answer.
Due to its relative simplicity and visible location not far from the city, “Amazing Sideshows” was one of our most solved puzzles.
“Creative Differences” puzzle
“Creative Differences” (by Sean Gugler) was based on the classic genre of spot-the-differences puzzles, with a twist. There were three signs, with differences between all of them, but due to the geometry of the sign only two can be seen at once. This encouraged solving by groups of people.
This puzzle was troubled during initial testing; players knew to find differences, but couldn’t figure out what to do next. During our first major playtest, a large group of solvers spent over half an hour stuck. We made substantial changes to the guide sheet and keypad as a result.
The updates must have helped, as this was the most-solved of all our puzzles! Players liked the familiarity and were able to solve the puzzle. We frequently witnessed pick-up groups of solvers helping each other by circling the sign and calling out differences visible from different points.
“Mirror, Mirror” puzzle
As you can see, “Mirror, Mirror” (by Jesse Morris) did not follow the triangle-sign format!
Instead, eight smaller backlit glyphs surrounded an area. At the center was a table with half-silvered mirrors available. Using the half-silvered mirrors, solvers could combine pairs of glyphs (by matching colors, or symbols) to form letter shapes, as hinted by the guide sheet.
This puzzle was entirely solvable during the day (and many people did so), but was best at night, when the ring of colored glyphs created a delightful ambience of mysterious calm.
“Mirror, Mirror” was hard to test in its intended configuration, but Jesse packed the important parts and created most of the installation at the Backyard Burning Man event.
“Target Practice” puzzle
“Target Practice” (by Sean Gugler) also had a unique physical build. Designed to evoke a carnival shooting gallery, it had rows of waterjet-cut metal ducks and stars with letters cut in them. Reading the ducks in the right way provided a series of instructions which yield the answer.
Depending on who you ask, “Target Practice” was either the easiest puzzle to solve, or one of the hardest (after Unholy Union, and the meta).
“Unholy Union” puzzle
“Unholy Union” (by Dan Egnor) used the conventional triangle-sign format. This puzzle features a set of animal combinations (a lion with butterfly wings, a grasshopper with a peacock’s tail, etc) which form portmanteaus (such as BUTTON or PEPPER) that match illustrations at the bottom of the sign. Using that sequence to order highlighted letters reads out words which repeat the initial mechanism to yield the answer.
This was almost universally cited as the hardest puzzle. Many people — as many as half the solvers — never fully understood the basic mechanism and solved the puzzle as an anagram.
“Carnival Zodiac” metapuzzle
“Carnival Zodiac” (by Sean Gugler) was the final “metapuzzle”.
Players needed to associate each answer word with one of the “signs” on the billboard. The horoscopes associated with each sign identified part of the city by geography. Collectively, taking the intersection of all the regions yielded a single spot. Teams needed to visit that spot to get the final answer, then return to unlock the safe.
This safe’s keypad was unusual; instead of letters or numbers, it had cryptic symbols. The symbols were each two halves of letters in a stylized alphabet. The final answer, attached to a street sign at the designated spot, was written in that same stylized alphabet, in a way that indicates the sequence of buttons to press.
This was our most troubled puzzle during testing, even for players familiar with the concept of a metapuzzle. Testers had trouble with almost every step. Eventually we made the guide sheet quite explicit, including a 5-way Venn diagram intended to convey the concept of intersection. Based on solve rates, we were successful! Solvers seemed to find it a satisfactory conclusion to the hunt.
The schematic map on the billboard is also a large scale map to the other puzzles. Because we had to print the vinyl banners ahead of time, but didn’t know puzzle placement until we arrived, we used vinyl stickers to add the puzzle locations on site.
Each of the feeder puzzles had a small (1" round) button as a prize; the metapuzzle had a larger (1.75" round) button declaring the owner a “Puzzle Hunt Champion”, as well as a laser-cut wooden jigsaw puzzle depicting the Burning Man Puzzle Hunt logo.
We had no idea how many Burners would engage with our nerdy little pastime. For each puzzle, we had 300–400 prize buttons made (depending on our estimate of puzzle difficulty). For the metapuzzle, we made 100 prize buttons and around 75 wooden jigsaws. We replenished the safes during daily inspection and maintenance rounds.
During the day on Saturday (the day the Man burns, and the day before we disassembled the structures) we ran out of prizes at every single site, including the meta. As directed by our signs, a stream of solvers began arriving at our camp asking for prizes. We collected the names and addresses of 70 additional meta solvers to mail them prizes. (Jesse is fulfilling these requests as I write this.)
For our expectations, we considered that very successful.
Some of the solvers were “puzzle people” — the first meta solver we met was an MIT Mystery Hunt veteran — but for most, it was their first encounter with anything like a puzzle hunt. Many solvers we met asked us to return next year with a similar event.
We were happy to see solvers offering each other hints in a way that was respectful of each other and the puzzles. Often, pick-up groups who met at one puzzle site ended up solving many or all of the puzzles together.
We had a handful of glitches during the week. The keypad on one safe failed; the “Target Practice” guide sheet had an unintended red herring; several safes reset or ran low on battery. We were able to repair all these issues, and the solvers we met seemed generally understanding.
Each site had a log book in the safe; by the end of the event, they were all full of signatures and notes, almost all very positive. (Anyone who didn’t solve the puzzle didn’t get to write in the book!)
We brought a puzzle hunt to an unusual and challenging environment, and it worked pretty well. We credit a combination of designing for the environment, targeting the audience, and extensive testing.
Thanks again to everyone who helped! Most especially our playtesters and the members of Cirque du Squid who were very helpful and lovely during construction, teardown, and in between.