{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiazzkrsbq7jpn25pn4i222llxlbilib2jrp32wyjjieqjpcmwcrjq",
    "uri": "at://did:plc:pi6woz4d47bkuws673w2il2r/app.bsky.feed.post/3mgxjxutewqp2"
  },
  "path": "/t/announcing-scrod-like-haddock-but-faster/13761#post_20",
  "publishedAt": "2026-03-13T18:20:32.000Z",
  "site": "https://discourse.haskell.org",
  "textContent": "this has got to be the more awesome part really.\notherwise your throwing away typechecking which is available, like, why?\noh, because it doesnt typecheck yet!\ndocumentation for part complete code is an awesome idea!\n\ni wonder if someone can get a partial compilation version of this working…\nit might serve to be a really informative graph of type mismatching or something.\n”this function is likely written incorrectly, since its type would imply infinite self referential combinatorics, but it uses the following functions from nearby scope allowing its correct type to be nearly resolved, and the most likely part which is wrong is where variable name reuse in a shaddowed block changed in a recent rewrite as compared to this other version of the documentation which was way less incorect”",
  "title": "Announcing Scrod, like Haddock but faster"
}