![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
rhuA friend from shul called me a few weeks ago with a request. He was invited to give a presentation at the World Economic Forum, and he wanted to use a series of puzzles to engage his audience and demonstrate different forms of collaborative workflows. The way he presented his idea, it was clear to me that he actually had a good grasp on how puzzle solving works, both on the individual and team levels, and how metapuzzles could be used to illustrate his points. The idea was to have three rounds that could be solved between courses during dinner, and his talk would follow the dinner.
If you want to solve the puzzles before reading further, I'll direct you to this PDF which contains all three rounds of puzzles. (The page numbers will indicate which round you're in.)
This was an interesting challenge, because not only is the audience not made up of experienced puzzlers, they're not even necessarily going to say "Oh, a puzzle, how fun!" And while they all understand English enough to participate at the WEF, I can't assume any particular cultural background (so no indirect cluing -- and unlike the World Puzzle Championships, I can't just give them a Nurikabe and expect them to have any idea about how to start solving it.) Finally, while they might have smartphones, a puzzle that requires searching the Internet for the answer is also not appropriate given the time demands.
These puzzles needed to be solvable relatively quickly, with no guessing about "aha" moments, and requiring no information that wasn't explicitly present on the handouts.
On the other hand, the puzzles couldn't be so trivial that they'd take no time at all. As you'll see, one of the goals was that solvers *had* to organize into teams; you couldn't blaze through a round by yourself in the five minutes allotted for each round. What I decided was that the thought process should be straightforward and quick, but the mechanics could be designed to slow people down enough that everyone would need to form teams, but the teams would all be able to finish in time.
I wrote three rounds of puzzles. The first was a single puzzle with an answer-extraction mechanism, which could be solved by an individual. The second had three puzzles and a meta, and in order to complete it in the allotted time, you'd have to join up with others and work in parallel. The third had five puzzles and a meta, only in order to finish solving your puzzle you'd need some information from another puzzle, requiring not just parallel work but mid-stream coordination.
Well, that was the plan. Only I got horribly sick the day I was planning to write the round of five, and fell behind schedule. So what I delivered in time for test-solving (the Thursday before the MIT Mystery Hunt) was a polished round 1 and 2, and enough of round 3 to test-solve but without a lot of flavortext and without the polished formatting.
And a good, thing, too, because even though I put a lot of effort into making things as easy as I could, I failed. In fact, what we discovered was that my round 1 puzzle was best solved by a team working in parallel, and my round 2 puzzles were best solved by a group exchanging ideas as they worked. And round 3 could not be completed in less than half an hour.
So... I deleted round 3 (and saved the ideas for sometime when I'm less restricted, because there are some things I really like in there that were compromised by the limitations of the venue). I renumbered rounds 1 and 2 to be 2 and 3. I added explicit examples of "how this puzzle works" or a worksheet to walk solvers through the most efficient way to organize their partial solutions, so that no one would be misled down false trails or waste time on roundabout solutions. And I wrote a very very simple intro puzzle to serve as the new round 1.
I haven't heard back yet how it actually went. (I'm not even sure when it was supposed to run; perhaps it hasn't yet.)
It took longer to put together than I had expected, but it always does, even when you account for the fact that it always takes longer than expected. But I think the end result is fun, and illustrates my friend's talk nicely. And who knows? Maybe someone famous will be inspired to join a team for next year's MIT Mystery Hunt.
I hope that you enjoy it.
If you want to solve the puzzles before reading further, I'll direct you to this PDF which contains all three rounds of puzzles. (The page numbers will indicate which round you're in.)
This was an interesting challenge, because not only is the audience not made up of experienced puzzlers, they're not even necessarily going to say "Oh, a puzzle, how fun!" And while they all understand English enough to participate at the WEF, I can't assume any particular cultural background (so no indirect cluing -- and unlike the World Puzzle Championships, I can't just give them a Nurikabe and expect them to have any idea about how to start solving it.) Finally, while they might have smartphones, a puzzle that requires searching the Internet for the answer is also not appropriate given the time demands.
These puzzles needed to be solvable relatively quickly, with no guessing about "aha" moments, and requiring no information that wasn't explicitly present on the handouts.
On the other hand, the puzzles couldn't be so trivial that they'd take no time at all. As you'll see, one of the goals was that solvers *had* to organize into teams; you couldn't blaze through a round by yourself in the five minutes allotted for each round. What I decided was that the thought process should be straightforward and quick, but the mechanics could be designed to slow people down enough that everyone would need to form teams, but the teams would all be able to finish in time.
I wrote three rounds of puzzles. The first was a single puzzle with an answer-extraction mechanism, which could be solved by an individual. The second had three puzzles and a meta, and in order to complete it in the allotted time, you'd have to join up with others and work in parallel. The third had five puzzles and a meta, only in order to finish solving your puzzle you'd need some information from another puzzle, requiring not just parallel work but mid-stream coordination.
Well, that was the plan. Only I got horribly sick the day I was planning to write the round of five, and fell behind schedule. So what I delivered in time for test-solving (the Thursday before the MIT Mystery Hunt) was a polished round 1 and 2, and enough of round 3 to test-solve but without a lot of flavortext and without the polished formatting.
And a good, thing, too, because even though I put a lot of effort into making things as easy as I could, I failed. In fact, what we discovered was that my round 1 puzzle was best solved by a team working in parallel, and my round 2 puzzles were best solved by a group exchanging ideas as they worked. And round 3 could not be completed in less than half an hour.
So... I deleted round 3 (and saved the ideas for sometime when I'm less restricted, because there are some things I really like in there that were compromised by the limitations of the venue). I renumbered rounds 1 and 2 to be 2 and 3. I added explicit examples of "how this puzzle works" or a worksheet to walk solvers through the most efficient way to organize their partial solutions, so that no one would be misled down false trails or waste time on roundabout solutions. And I wrote a very very simple intro puzzle to serve as the new round 1.
I haven't heard back yet how it actually went. (I'm not even sure when it was supposed to run; perhaps it hasn't yet.)
It took longer to put together than I had expected, but it always does, even when you account for the fact that it always takes longer than expected. But I think the end result is fun, and illustrates my friend's talk nicely. And who knows? Maybe someone famous will be inspired to join a team for next year's MIT Mystery Hunt.
I hope that you enjoy it.