Post #1678487

Yesterday's pastpuzzle was (I thought*) completely solvable from just two rows of one person's results under the hashtag here. Today's was trickier -- almost no yellow squares (which in hindsight was a hint in itself). After browsing the hashtag for a while I had a handful of possibilities for the last pair of digits, and a handful of possibilities for the century -- but a handful times a handful is too much to carry -- and a bunch of notes like "d2 is not 2+d4", "d3 is not d4-1", which I couldn't quite organise in my head. So I got around to coding up a little thing, where I enter results rows from other people's results, and it tells me which dates are compatible. (*It turned out there were actually two possible options for yesterday's, based on the two rows I used, almost a century apart, I had mistakenly rejected one of them). I fed it quite a lot of results before it pinned down today's. Next I'm curious to try and pull out some more details about how much different results rows narrow down the solution, and how they interact most efficiently... #pastpuzzle 469 🟩🟩🟩🟩 (0) ▪️▪️▪️▪️ ▪️▪️▪️▪️ ▪️▪️▪️▪️ 1/4 🥇 https://www.pastpuzzle.de