Greg Wilson on Nostr: npub1upkp7…4tcwy cycles: TidyBlocks (think "Scratch for data science") we had to ...

npub1upkp7fd7rc3lrjg23r8gy0wc723vze7mxlx5984ut6zurjzpf5xss4tcwy (npub1upk…tcwy) cycles: TidyBlocks (think "Scratch for data science") we had to check there were no cycles when joining results (because joins wouldn't fit the click-blocks-together model, we had to use notifications). for shunting yard, we had some messy CSV files to parse, and treating it like nested expressions with precedence turned out to be the answer. unification was checking that HTML pages created by various SSGs conformed to rules. Bloom filters was bioinformatics: didn't write, had to debug.