MITMH 2020: 2019 Dec

Careful: I'm pretty blasé about spoilers for this year-old hunt. If you haven't already solved the puzzles you wanted to solve, maybe don't keep reading.

Every so often, fearless leader Corey sent out a "State of the Hunt" email. Here are some excerpts from December's email to let you know what the writing team was doing. (And to remind me, as I write this, what we were doing.)

Overall: We still have plenty of work hours needed, but we’re on a good trajectory. All but about 20 puzzles are more or less done, and nearly all of those 20 are making good progress. Pennypark.fun worked very well during BTS3, for players and HQ. We have scores of Left Outers planning to attend Hunt in person, and Hunt Ops planning is in full force. Our path is clear—we just need to keep up the effort heading along it!

Registration: We have 98 teams signed up, estimating 2900 attendees including local and remote. We’re no longer accepting team homeroom requests (and, we’re oversubscribed at that) but registration will remain open through the start of Hunt.

Server: pennypark.fun is feature complete and now is only fixing bugs or adding critical features. Beginning Dec. 30, we’ll have a staging server always running, with simulated teams playing the Hunt (so everyone can watch the Big Board, etc.).

Story: The end game design has been completed, and the big part there is that, after solving the pressed penny capstone, teams must complete 11 mini runarounds (one for each land) before solving the final gizmo. These were authored over the BTS3 weekend and tested this month. Anyone arriving to Hunt early in January can (please!) test them out as well.

Puzzles: Although BTS3 is behind us, we still have many puzzles that need additional testing (for revisions or changes as they went into production). Please do keep up your test solving!

Requests for help

Testing. Please test puzzles! Even if you attend any of the big test solves, please try solving a puzzle or two from the list (perhaps you didn’t see everything in the BTSs).

HD1. December 30th, we’re going to host Hunt Drill 1. The goal of a Hunt Drill is to test all the mechanisms of Hunt, except the specifics of solving puzzles. Eg, this’ll be a great time to check how the Big Board works and whether the Hint Ops interface is working well for our Hint team. An invite for HD1 (and details about optionally attending in person in Yar’s house) will go out shortly. See #HD1 for details.

Staffing survey. Linda kindly requested everyone (everyone) to respond to this survey about staffing Hunt 2020. Whether you’ll be traveling to Boston or can help only from the comfort of your home far away, please do reply to the survey. No matter your location or hours available, we’ll have a way for you to contribute!

As I said, there wasn't so much strain on the Production team leading up to BTS3 as there had been before BTS2. Most puzzles were already in pretty good shape; there weren't so many last-minute scrambles.

The days leading up to BTS3 were less stomach-churning, production-wise, than those leading up to BTS2 had been.

My brain, always looking for things to worry about, worried more about what would come after—puzzles still struggling to get through testing; puzzles that needed lots of revision.

Tech Czar Doug, working on the web server, did run into more last-minute stomach-churning situations because he was putting together materials from a few teams. Some of those materials came in last-minute. I don't know whether Doug's stomach was literally churning, but I bet he wasn't happy that we had so little time to test things together. E.g., one puzzle that used Javascript didn't work for BTS3: the puzzle's own JS conflicted with with the appears-on-every-page JS. Doug tried to fix a long-standing problem in the puzzle-page page layout and broke the layout of several puzzles which had hackish workarounds for the problem…

But anyhow, BTS3 came along, albeit not without hiccups.

Big Test Solve 3 took over Rich and Kiki's house for a weekend. Once again, I knew too many answers to do much testing. I was kind of useful—I was the only member of HQ who was physically at the house. Thus, I spent much of my time in tactical eavesdropping, letting other HQ folks know how the team was progressing on various puzzles.

I helped test the Character Breakfast event, playing Pictionary in the medium of pancake batter.

The rest of December was somewhat busy in production (puzzle-HTMLification). It wasn't busy full-time as it had been in the weeks leading up to BTS2. But whenever I'd cleared out the task-list, within a few hours some puzzle or other would have have some long-anticipated improvement made. And thus, there was something new to convert to HTML.

The to-do list didn't get long enough to activate the rest of the Production-Team. But it never stayed empty long.

J. K. Rowling, author of the Harry Potter books, tweeted something something transphobic. Our hunt had a lot of Harry Potter-themed content. One of our puzzle-rounds was wizard-themed. Heck, I had collaborated on a Harry Potter-themed puzzle. Would our puzzles cause misery, reminding non-transphobes of this hatred?

I tried to think of workarounds. It was December; the hunt was coming up in about a month. We probably couldn't change all those puzzles and get the new versions tested in time. Could we add a content warning? Would that be weird? Meanwhile, plenty of puzzle change-requests poured in.

After a few days in which I theoretically-tried to think of workarounds but really-failed to get 20 distraction-free minutes to think about workarounds, I gave up on finding time to come up with any workaround.

Nina had playtested almost all of the puzzles but was nevertheless eager to do things to boost puzzle quality. She went through the puzzles one by one; rather, she went through the puzzle solutions one by one, looking for mistakes. You might recall that was my naive thought of how fact-checkers would check puzzles. But instead, fact-checker Stribs was testsolving-extra-carefully.

You might think that Stribs' approach would catch all the problems. After all, if a puzzle designer writes up the solution OK but lets a typo into the puzzle, you want to catch that. However, Nina soon found plenty of mistakes by looking over solutions.

There must be some amount of fact-checking where you hit a point of diminishing returns. But two passes of fact-checking following two clean test-solves was not too much.

(My previous experience with smaller events made me think that a couple of clean testsolves and a look-over of the solution would be enough. But I think in this big event, puzzle authors weren't so careful. If you're contributing one or two puzzles, you probably go over them pretty carefully. But when I was authoring puzzles for Mystery Hunt, it was all too easy to overlook problems as I flitted from puzzle to puzzle.)

Our fact-checkers saved our butts so many times.

Fact-checkers found things to change. Puzzle authors found things to change. Lots of changes to make; it rarely let up for long.

But when the pace did let up, I looked for puzzles that were stuck in the pipeline because the authors were too busy to write up the solution. One by one, I wrote up draft solutions, asked the relevant puzzle author to look over each one. One by one, I skoonched those puzzles along the pipeline.

This solution-ghostwriting was mostly helpful, but also backfired. I wrote up a solution to one puzzle and showed that solution to one of the puzzle's authors. He spotted a problem and wanted to write up something better. That sounds fine but but this puzzle author was Ben de Bivort, who was also putting together the high-tech gizmo for the finale. We really needed that gizmo to work; we could have procrastinated on a solution to this late-in-the-hunt-and-thus-seen-by-few-teams puzzle. So…I was distracting. But I was mostly helpful.

Art Czar Natalie and Nick had their kid, Leo. That was a big deal. I was happy for them. I was darned happy that Natalie had set us up with an account on that clip art site ahead of time.

We did a hunt drill. Folks gathered at the Yar/Jan house. For many Team Left Out folks, this was their first chance to see the administration interface: how to give hints, how to manage classroom visits, and other behind-the-scenes computer stuff. I'd used the interface at our Big Test Solves, so for a little while, I was the know-it-all. But other folks figured out the interface pretty quickly without much help. Doug had made things pretty intuitive.

Lots of changes. (I guess the holidays were in here somewhere?) Lots of changes.

Jan 2020

| comment? | | home |