The Cruxwick File / Any Day In 2020

[kevandotorg]

I'm considering a puzzle book in the vein of the Sherlock Holmes Consulting Detective boardgame and Obra Dinn, where the reader is given a wealth of in-universe newspaper clippings, maps, directories and other documents, and a mystery can be solved by successfully cross-referencing information and eliminating possibilities.

The autogeneration idea would be to start with a single fact ("Mr Black was killed by Colonel Mustard") and then to recursively obfuscate it in a variety of ways ("Mr Black was killed with a skull-topped malacca cane" + "Colonel Mustard has a skull-topped malacca cane"), hiding all of these facts in various documents, until the case was complex enough to require some effort to solve. The script would generate as many non-overlapping cases as was needed to fill 50,000 words, along with a lot of randomly generated background filler to bulk out newspaper pages and address listings.

I'm imagining a standalone file of documents with a simple introduction and no further mechanics, rather than a Consulting Detective door-knocking system, but will see how it goes.

[verachell]

Wow your project sounds AMAZING! Great idea. Looking forward to reading the end result.

[enkiv2]

This sounds both extremely readable & extremely doable. I look forward to seeing it :)

[kevandotorg]

Sadly the generated mystery idea never quite hit critical mass, not getting much further than generating a simple (and trivially solvable) network of newspaper stories that described a crime and gave some suspects alibis based on them being in other towns that day. Small sample of output for The Cruxwick File:-

It failed partly because I didn't get a solid logic structure down early enough in the month, but it also disappeared into the general fog of 2020, and I realised I'd never get anything interesting finished in time.

So instead, and in that spirit, a quick end-of-month project that generates 50,000 words of personal diary for me for this year, starting on the day that England went into its first lockdown on the 23rd of March, and based on a rough idea of things I did during the 250 days between then and the 28th of November. It's felt like a very recursive and randomly ordered year for me, with any given day from earlier in the year being completely repeatable now, even including things which shouldn't make any sense in late November, like "hoping that the US election ends well" and "not knowing if there'll be a no-deal Brexit in January".

So if something happened once in that period, the diary assumes it will generally happen once every 250 days; if it happened ten times then every 25 days, and so on. It assumes no end to the restrictions and lockdowns, continuing until 50,000 words are written. First three entries of Any Day In 2020:-

Monday the 23rd of March 2020

Dreamed about walking for miles. Harvested some chard. Attempted to detect bats. Watched some short videos. Still no idea whether we'll end up with a no-deal Brexit during a pandemic. Launched a print-and-play board game for an online games festival.

---

Tuesday the 24th of March 2020

Slept badly last night. Did some playtesting of a word game online. Watched a science-fiction show. Listened to a podcast. Did some background reading for the new book. Read too much political news. Read an interesting article.

---

Wednesday the 25th of March 2020

Still no idea whether we'll end up with a no-deal Brexit during a pandemic. Watched some film I'd taped off the television. Met up with a friend in an open-world video game. Did some website maintenance. Got a new board game in the post. Did some writing on the new book.

Source code: https://github.com/kevandotorg/nanogenmo-2020
Full text: https://kevan.org/nanogenmo/2020/anyday.txt

[verachell]

I love the sample output for the Cruxwick File! Your final project was fun to read too (and totally understandable not getting time to do the full Cruxwick), but your sample output for Cruxwick so far was SO AWESOME!! Thanks for sharing it. It's fascinating that you generated different newspaper clippings with different fonts.

[dmorfin]

Just stumbled on this. The Cruxwick File sounds pretty neat. I think the key to making this an interesting game is that there needs to be another level of facts. There's the 1 "core secret" and the N "clues", but there needs to be another set of M "inferrable facts". The key to a good mystery is that the inferrable facts should be sufficient to derive the secret, but aren't ever directly mentioned. I think that's why the initial attempt was "trivially solvable". Then you adjust difficulty by adjusting how many of the inferrable facts are reasonably needed to solve the puzzle (e.g. at least 50% are needed), how many clues are related to each fact, and how many unrelated "clues" (red herrings) are present. Just my two cents, but it does sound like a really interesting idea.