Log in

No account? Create an account
The world's puzzle hunt - Many a mickle maks a muckle

> Recent Entries
> Archive
> Friends
> Profile

June 12th, 2013

Previous Entry Share Next Entry
06:24 pm - The world's puzzle hunt
The US Puzzle Championship is taking place online on June 15th, which is this Saturday; the details will be posted at the USPC 2013 site imminently. US solvers interested in consideration for the perennially-top-contender national team at the World Puzzle Championship need to start solving at 10am Pacific / 1pm Eastern / 6pm UK time; last year, unofficial contestants had the freedom to choose when to start their participation, with the same 2½-hour time limit.

It's always a good contest and you'll be able to download the instructions for the types of puzzles in advance, so you can see whether this year's bundle suits your taste or not. (The event isn't being used to determine a place on the UK team this year; UK solvers only have one online contest to earn one of three spots on the team. Hey, I don't make the rules.)

After DASH a week and a bit ago, I've been in a puzzle-y mood pretty much non-stop. Accordingly, now is as good a time as any to make a post I've been planning to make for over four months, about this year's MIT Mystery Hunt.

The MIT Mystery Hunt has been an ongoing sideline fascination in my life for about as long as I've been interested in puzzles; I mentioned it first in a friends-locked entry back in 2003. Ten years on, particularly now I make the assumption that you probably aren't going to be reading this blog without an interest in games, it's almost possible for me to assume that anyone who reads this will know what I'm referring to by name. However, I shall just quote and update correct how I described it over a decade ago, more because I can than for any other reason:

The MIT Mystery Hunt is an annual event taking place at the end of the Independent Activity Period (possibly inaccurate translation: vacation) held by MIT which has been running for almost twenty-five thirty-five years. Very roughly, very large teams get together and solve extremely exceptionally difficult puzzles for a weekend. The puzzles test remarkably many facets of ability as opposed to always just logic; many of them require obscure parts of pop culture, others might involve feats of arts and craft, others might involve unusually challenging scavenger hunts around the MIT campus or the wider Boston, MA area. The hunt starts at noon on Friday and normally finishes (with some team finding a carefully-secreted coin) between 36 and 48 hours later, with outliers at both ends and a longer tail featuring a few 60+ hour finishes at the long end, usually with teams working shifts through the nights.

The idea is that the hunt is designed not just for hardcore pencil-and-paper puzzle enthusiasts - a whole gamut of skills should be tested. By and large, the puzzles are incredibly good - really off-the-wall concepts, almost always really well executed. Originally the event was at least nominally for MIT students alone, but as its fame spread, all sorts of people took part in it. There are no real limits on the quantity or types of research you can perform, so this works out as, effectively, "infinite phone-a-friend"s and extensive Internet use. I know a few folk in Boston who know folk and so last year in 2002 I helped out one team remotely. You can still read last year's 2002's puzzles and admire their ingenuity, or just read a non-technical general-interest article on the phenomenon with examples.
The definitive write-up of this year's Hunt probably comes from MIT student newspaper The Tech, which is as it should be; they also have a version of their story with incidental videos. Technical difficulties stopped the Hunt from starting until 2 p.m., rather than the habitual noon, on the Friday before Martin Luther King, Jr. Day; a winner was not declared until just before noon on Monday, and that declaration happened prematurely with the coin not actually being found until three hours later, with the actual hunt duration of 73 hours and 18 minutes being about five hours longer than the previous record. As a result, I fear that this hunt will go down in history, first and foremost, as being associated with the adjective "long".

Looking at people's write-ups of their Hunt weekends is possibly a good way to get a better feel of the MIT Mystery Hunt experience. I really enjoyed Eric Berlin's write-up that captured the emotions of a player on a competitive team really well; Andrew Greene's writeup is shorter, but conveys the sense of fun, and the always-lovely Clavis Cryptica wrote joyfully and comprehensively from a first-time player's perspective in 1, 2, 3, 4, 5 parts. Closer to home, in fewer sections but maybe with even more of the really enjoyable detail, rhysara wrote hers in two parts, and motris's own impassioned, honest and brilliantly penetrating must-read write-up is the one that seems to have attracted many comments from many players on several of the most competitive teams.

Traditionally, the winning team one year writes the Mystery Hunt the next year, and they are given the freedom to do so however they wish. The two previous years' hunts had been won in roughly 41 and 34 hours, with the 34-hour hunt's setters deliberately intending that many teams would get the fun of seeing the hunt all the way through. This year's setting team considered these hunts to be shorter than optimal and established a goal for themselves that the hunt would not be that short, even for the most competitive teams. As their hunt broke the existing hunt duration record by almost 10%, and had to be artificially shortened in order to do so (which, arguably, is a distinction that matters only to the purists, with a salute to them because that's no criticism) they certainly succeeded in their goal. The stoic Is It Hunt Yet?, normally a NO and a countdown clock until the next event, turned first to the joyous YES... and, eventually, to a slightly ill-tempered STILL.

However, the teams that played this year perhaps might not have been expecting it to be such a sort of Mystery Hunt, and this difference from their expectations might have stuck a little in some of the competitive teams' craws. The write-up in WIRED reflects how the strongest competitors may have felt straight away after going all-out for the duration (and, remember, many on some of the most competitive teams go without sleep - and this time for longer than the single missed night that they might have expected) and there is definitely some of this in Eric Berlin's write-up as well. Judging by comments left later, this piece definitely caught the mood of many of the solvers.

The reasoning for the length of the hunt has been brought into question by some solvers, particularly those who considered the hunt to have run too long. There is a post comparing the number of puzzles in previous MIT Mystery Hunts, typically narrowly into three figures over the past seven or so years. This year's hunt had 153 puzzles (source, 17 minutes through). The length of the hunt and the two-sleepless-nights-rather-than-one nature contributed to an unusually great degree of tiredness and led to, at best, some crossed wires (not just referenced in that post, but that entire conversation thread) which left a sour taste in some mouths.

Furthermore, some solvers criticised several of the puzzles in the hunt for being unusually long. One puzzle that came in for criticism in this regard was a jigsaw puzzle where teams were distributed a bag of jigsaw puzzle pieces; teams discovered that they did not form one single huge jigsaw, but 49 smaller jigsaws - each of which was a 7x7 jigsaw with one piece missing, with the missing pieces being used to determine the true answer to the puzzle. As an open question, might a 5x5x5x5 or a 6x6x6x6 version of the puzzle have struck a more satisfactory balance than the 7x7x7x7 version that was used in practice? (Clearly yes for some teams and no for others, but whether the balance would have been better is an open question.)

In addition, some solvers criticised several of the puzzles in the hunt for being the wrong sort of hard. I don't know if there is a meaningful sense in which a puzzle can be too hard for the MIT Mystery Hunt, though there is an argument that a puzzle constructor's job is to lose with grace and a puzzle that teams do not solve in practice is not fit for purpose. Probably more to the point, puzzles can be so hard that even the might and immense combined resources of the most fearsome puzzle teams on the planet do not enjoy solving them. (Cases in point this year: discussion of the fractal word search with an answer on the 86th level of iteration, and discussion of the Engima machine meta-puzzle.)

Yet the strength of the teams sets the barrier for "too hard to be fun" really high. One of the most famous anecdotes about one of the earliest Hunts runs Once I wrote a clue in Minoan Linear B, a totally obscure language that was used on clay tablets in ancient Crete. To make things tougher, I didn't tell them it was Linear B and I checked out the two library books on the subject. All the teams solved it anyway! Villainy indeed in the days when information on the Internet was so much more scarce than is the case now.

One theme I've heard in a few places (perhaps most notably from one of the team running the event in this recording of a talk at the Game Control Summit) is that the principal problem this year was one of editing. As it was the first hunt that the Manic Sages had run, many of the puzzle authors were very new and the interaction between new puzzle authors and their mentors may not have worked as well as it might have done; the ratio between effort spent on editing and the effort spent on authoring may not have worked so well.

Additionally, decisions taken about translating internal playtesting results into suitability for the live hunt came in for criticism. Quoting the hunt's art director:
We thought that the average Manic Sage was a below-average Mystery Hunt participant, and that the only reason we won last year was because of sheer manpower. So if a typical testsolving group only barely managed to make it through a puzzle, after a few attempts, it went in, with the thought that it would be easier for a “real” team. If a group of ordinary humans couldn’t solve the puzzle, but one of our best solvers could, it went in [...] Every puzzle we used was solved in testsolving. We just thought they were easy puzzles when they really, really weren’t. We now know that the average Manic Sage is a slightly-above-average puzzler, and our best puzzle-solvers are as good as anyone else’s. Had we believed this before, it would have improved our estimates quite a bit.
It has also been suggested that the puzzles that made it past playtesting were particularly well-suited to the strengths, and size, of their team. Manic Sages is a particularly large team; their winning effort in 2012 may have used around 150 actual solvers at some point or another.

A significant part of the instant reaction to this year's event was borne of tiredenss and frustration and many of the Manic Sages were, not at all unreasonably, disheartened by this reaction, with the art director's write-up also a very interesting read. I really liked Peter Sarrett's take on this. As befits someone with his experience and knowledge from both sides, he got to the nub of the issues but sympathised with the setters, having his own take on having run a hunt that was misconstrued by some of its participants. (It's not at all uncommon; I was similarly hard on a small hunt-like Quest that I co-ran myself at Nimbus ten years ago where I was unhappy with the way it turned out.)

And yet, and yet, there is definite reason why I think this should go down in legend as a long Hunt, rather than a more negative adjective. Every Hunt does a great many things not just right but brilliantly well, and the size and ambition of this Hunt is reflected by the great many innovations and high spots that people will remember and celebrate for years to come. Happily and appropriately, the positive and congratulatory feedback came just as quickly as the more exasperated feedback, with this comment thread just one of many with the high points as well as the low; see other comment threads on that post, among others, for further plaudits where they were due.

There were some glorious steps forward, some evolutionary, others revolutionary, in terms of infrastructure. The whole Hunt was themed around a heist to retrieve the coin of legend, with the end-of-game runaround requiring teams to physically pull the heist off by resolving six physical obstacles, which were frankly incredibly cool; you may have seen laser mazes before, but the picture half-way down the art director's report looks a different class. A standard complaint is that only the winning team or teams get to play in these incredibly intricately designed end-of-game runarounds; here, people got to practice on slightly simpler versions of one of the six obstacles, and thus have the fun of interacting with them, as rewards for completing each round. These were justifiably hugely popular and made the reward-to-effort ratio much more favourable.

The hunt also deliberately started with a relatively easy "round zero" made up of just six puzzles and an associated metapuzzle, so that even less experienced teams might get a flavour of the hunt. This was apparently greatly successful as a way to help everyone find their own depth. Apparently a team of five first-time solvers, who had only signed up on the day of the hunt, were still calling in answers to these Round 0 puzzles on Monday. They were clearly having a great time, and the running team were really heartened to hear it. I hope this is an innovation whose spirit can live on even if the precise method of implementing it is varied from time to time. Credit also for permitting teams to submit answers electronically, apparently for the first time, rather than having to say them (sometimes spell them) down a phone line; I suspect this will immediately be considered Best Practice.

Everyone who participated will have their own favourite puzzles. I was working the night shift on Friday, Saturday and Sunday night of the hunt this year so couldn't devote time to solving the puzzles, but I did enjoy looking at them as they were unlocked. British readers may be as taken as I was by this cricket-themed puzzle. It's one of the type of puzzles where a big part of the challenge is to work out how to solve it; I worked out how I would go about solving at least the first two-thirds of it, which was very satisfying. (Doing it would surely have been much harder than just knowing how to do it!)

The artifice behind A Walk Around Town is incredible. The concept is that you have a series of instructions about a fictional journey around Cambridge - Cambridge, MA, the home of MIT, which generates a message. However, this message reads "Start At Old Schools". The Old Schools are part of the University of Cambridge - the one in the United Kingdom - and it so turns out that the same instructions can be followed to describe a fictional journey around the UK version of Cambridge and generate a message to answer the puzzle. That's beautiful, a work of art.

Similarly, So Good They Named It Hull relies on the similarities in place names between England and New England. This delightful construction is possible because Manic Sages is a worldwide team with at least two separate clusters of solvers - and puzzle setters - in the UK (Cambridge and Manchester), and others still in Japan. (Huge kudos, as ever, to devjoe's amazing index of MIT Mystery Hunt puzzles, which goes to show the incredible range of topics that have been covered over the years.)

I think my favourite story from this year is one briefly mentioned in Eric Berlin's write-up:
Also grabbed from the unattached hunters list was a total puzzle hunt newbie, a lawyer (I think) from Denver named Jonathan. Except he didn’t join Palindrome for himself. He came to the Mystery Hunt with his son, Brandon, who is 11 years old. His father told me that Brandon was a puzzle nut who couldn’t wait to one day participate in the Mystery Hunt. The trip to Cambridge was Brandon’s Christmas present.

I was a little nervous for both Brandon and his father — the Mystery Hunt can be overwhelming and difficult even for people who believe they enjoy puzzles, and if the Hunt did not match Brandon’s expectations, it could have been a disappointing trip home for them. I needn’t have worried. Jonathan told me that his son had been through the Hunt archive and knew what to expect. It would have been fairer to say that Brandon had memorized the Hunt archive. He could speak of themes and specific puzzles with an astonishing fluency, and he fit in with the team like a veteran. He was always busy with a puzzle, attended a bunch of events, and was still bopping along with the rest of us in the wee hours of the morning.
I salute you, Brandon. I am also reminded of Kit Williams saying his seminal 1980 work that started the armchair treasure hunt genre, Masquerade, was "so simple, a ten year old could solve it". Could the same be said of the MIT Mystery Hunt puzzles? I'd have thought not, but perhaps Brandon might provide evidence to counter an only slightly weaker claim.

So the 2013 Mystery Hunt is in the can and we can hope that 2014 and future editions might go on to gain from the very considerable positives of this year's hunt, as well as the cautionary tales. Different people have different goals and choose to emphasise different aspects of the whole puzzle hunt universe, so it'll be interesting to see, in time, how this year's winners choose to interpret the traditions and how they grow upon their experience from running the hunt ten years previously.

It's also interesting to consider how the Hunt has changed over the years. Obviously, the biggest teams have got bigger, and the Hunt bigger with it. I also feel more comfortable talking about it as an outsider. At the end of the day, I am a disconnected no-mark with no connection to MIT, except that I visited the museum once back when Meg and I were long-distance and she lived in Somerville. However, I've seen it suggested by members of three of the bigger teams (Palindrome, Manic Sages and Codex) that they are all open membership, so there is a sense in which the event belongs more to the global puzzle community and less to the MIT Hunt community, relatively speaking, than ever before. There has long been an "unattached hunters" list, but the barrier to entry of needing to know existing participants is less influential than ever before. (Nevertheless, it is a delightful bonus!)

There's also the open question of whether the recent increases of size in the Hunt, its difficulty and its teams necessarily should be sacrosanct. Medium-size teams have beaten huge ones in the recent (though not immediate) past, where medium-size these days is considered to be 30-40, so fears of escalation are not entirely inevitable. I generally tend to support the lowering of barriers to entry, but my opinion is completely irrelevant here. I also tend to think that if the world has a single event where team sizes and puzzle difficulty are unlimited, which would be a fun thing to have once, then the MIT Mystery Hunt is at least as good a location as any to be it.

As I wrote in 2004, I can't remember when I first found out about the MIT Mystery Hunt, though I would guess it was in the late '90s. In 2000 and 2001, I visited Columbus, OH and got to know a number of gamers, many of whom were in the metro Boston area. Given that many of them had a puzzle and puzzle hunt connection anyway, it seemed likely that they would know more about the MIT Mystery Hunt. Happily, they did, and I helped one team out from home over the Internet in the 2002 and 2003 Mystery Hunts. It was a fairly lonely experience at times - very rudimentary communication between me and the rest of the team. I managed to solve two of the 2002 puzzles from start to finish by myself, though - and my 2003 entry suggests they were Drive-Time Shift and Ringside Seat, both of which could be considered to be showing their age these days. (Or perhaps that's the Impostor Syndrome speaking; if I solved 'em...)

In 2004, I got together with a couple of other solvers down in London and we formed a small UK cell of Codex. It was a great weekend. In theory I was on their mailing list for a few more years after that; in practice, I never did much to help, and I was certainly nothing to do with them long before they won. My shift pattern has long interfered with solving, and it'll probably only be the two years in five where a long break falls over the weekend where I can properly get involved. Nevertheless, I've been passively fascinated for years, at least mentioning the event in passing in Friends-locked posts in 2003, 2004, 2005, 2006 and 2011.

It's also the case that solving MIT Mystery Hunt puzzles on your own is a very remote activity. This is to some people's taste, but I have not yet developed the maturity to stick with these things; if I'm not making progress within about 20 minutes, I tend to get discouraged and leave it to someone else. There is part of me that has long been trying to get OK with the notion that I will probably never solve a MIT Mystery Hunt puzzle again, but if there were another cell of UK solvers trying to get together and give it another crack as part of a larger team then I think it would be irresistible. I've been saying that for years, but never actually got it together at the time. And yet getting people together for DASH in London was so much fun, and introduced me to other people who might potentially also be interested, so there comes a point at which it looks much more practicable...

The MIT Mystery Hunt is the biggest and by far the most well-known of its type, but every now and again I read about other universities doing similar things. Some of them get linked to at Puzzle Hunt Calendar, others don't - particularly if they really are restricted to their local communities.
  • Berkeley have a 10-hour event in July which caters for remote solvers, at least in part;
  • Harvard had seven, though I don't know if the tradition there has faded
  • Rutgers had one, ditto
  • South Carolina have had two, one this year, and explicitly welcome remote solvers. South Carolina! As the husband of a Georgia woman, I am honour-bound to point out that if there had to be one of the SEC colleges with its own hunt, it's not even one of the smart ones!
  • Carnegie-Mellon's admirable KGB have their own hunts from time to time;
  • the University of Chicago have a Scavenger Hunt rather than a puzzle hunt, but it's admirable and delightfully unhinged;
  • Microsoft Recruiting have a College Puzzle Challenge held in parallel in 14 campuses, sort of as an independent, unrelated reuse of the DASH federated style of a hunt;
  • and doubtless other of the great smart and geeky universities of the US have their own counterpart events. Oh, huh, a quick search reveals Stanford's own, and I'm sure there must be others still. Someone should compile a list of these, because there isn't one at Wikipedia or at the Puzzle Hunt Calendar.
This is not just a US tradition; the global puzzle community is well familiar with the counterpart events in Australia. The University of Melbourne's Maths and Stats society have been running their own hunt for ten years now. This takes a slightly different format, with the puzzles (and hints) released over several days, but the puzzles are as fearsome as any found elsewhere. The University of Sydney has done their version for five years now, and a branch of Canon have their own, partly as a recruiting tool. It's tempting to wonder how the world might be different if the Australian hunts had preceded the US ones, but it's a case of different nations developing their own traditions.

It's not just the English-speaking world, either. I am absolutely fascinated to have learnt, about half an hour or so ago, about the TMOU tradition in the Czech Republic, that appears to have remarkable similarities to the "The Game" tradition in the US. It appears to be an outdoor linear puzzle hunt ran overnight from Friday evening to Saturday morning, with online qualification for an in-person event. The differences are that teams can only travel by foot and using public transport, and the distance is of the order of 20 km. The temperature will probably not be much above freezing. Teams have 3, 4 or 5 members, and these days, the biggest games have a good two hundred teams, of which a handful might complete the hunt. TMOU does not appear to be a one-off; I reckon this is a Czech language puzzle hunt calendar. I'm in the wrong European country.

The Japanese tradition is the Real Escape Game, a commercial timed physical analogue to those escape-the-room Flash games, which run as one-off events and have run into Singapore and San Francisco. The first San Francisco ones were criticised for being rather less kinetic than advertised; it would be interesting to know if later editions have had bigger budgets and more spectacular stunts.

I've not yet played Hint Hunt in London, but it does appear to be a relatively similar sort of reinvention of the same principle. It also runs as a year-round attraction that requires booking. The price is at reasonably high-end, bordering on corporate entertainment, levels; perhaps this is a consequence of London land rates, but considering the margins to be made and the remarkable TripAdvisor reviews, perhaps there might be scope for imitators or franchises yet. Heck, the probably-not-all-that-puzzly Si5 Spy Missions is up to three UK locations now.

The UK tradition derives from the aforementioned 1980 Masquerade book, wherein all the puzzles are revealed at once, there are deliberately plenty of false leads and there are generally no partial confirmations between the original puzzle and the single solution - the buried treasure. While Masquerade was a single example of its type, the Armchair Treasure Hunt Club discusses the hunts that exist and often has their own. They had an online one in March that was cracked in a couple of days; the puzzle used quite a few of the encryption methods that might well be familiar to participants in other Hunts and the solution does look like it uses techniques familiar from other hunt traditions. A new online hunt will be posted in late June and it might well be a good place to start.

However, the state of the art in the UK is probably the Logica Armchair Treasure Hunt, originally intended for members of the Logica/CGI consultancy and friends. A 25-year archive of hunts and solutions has been posted, and if you enjoy reading puzzles and their solutions as works of creativity, these stand up on their own merits as brilliantly as anything from elsewhere around the planet. I particularly recommend the 1992 hunt, which eschewed the habitually question-driven format, among the encryption, for something artistic with remarkably low levels of text.

If this piece has a conclusion, it's that different cultures around the world often share a love of puzzles and riddles, but interpret this to come up with their own traditions, though there are a great many commonalities in techniques. There's no reason why one particular tradition might not spread to another country; there's no reason why a UK university might not start its own Mystery Hunt somewhere down the line, or that other countries still might have their own puzzle traditions that have not made it to the English-language world. If you've only been exposed to one tradition, take a look at others which might be even more to your liking. There's a lot out there - more than you may think - and much of it is ready and waiting to be solved. Get in there!

Please redirect any comments here, using OpenID or (identified, ideally) anonymous posting; there are comment count unavailable comments to the post already. Thank you!
Current Mood: enthralledenthralled


> Go to Top