2026-08-25 22:33 UTC
Like, handwriting recognition obviously works decently well and is pretty old. It is, however, *not* a theory of writing. Handwriting recognition models tell you *nothing* about how handwritten text grapholinguistically functions!
Replies (1)
-
@jpetazzo@hachyderm.io 2026-08-26 06:59
@kevin@fedi.kbg.sh "handwriting recognition obviously works decently well [citation needed]" I'm trying to transcribe a corpus of two million postcards. Handwriting recognition does not, in fact, work decently well. :(