Heh, I read the title and thought this was going to be about someone porting legacy publications into Latex code. Anything from Newton's gravity to Marie Curie radium.
Which would be nice - some seminal publications from not all that long ago are only available as terrible quality scans.
This is an excellent idea. I was curious about using DeepSeek OCR for exactly this purpose. But a tricky question is if we could do some sort of looping or something "energy based" and use classical search to find optimal parameters (LaTeX settings) to minimize the error (pixel difference). Me knowing I would get obsessed with the second half is what's keeping me from the first half. Maybe a vision JEPA would be good. If I had API credits to burn, I'd copy paste our two comments and see how far Fable gets.
I just asked GPT6-Astra to transcribe a page from Newton's Principia to LaTeX. It did an amazing job outputing a XeLaTeX doc with a mixture of text and TikZ. Looks really good.
Did you do a pixel diff against the original? There could be lots of little errors or inaccuracies that are hard to spot. (Or not, it's hard to be sure without comparing.)
Rather Claudish(?)-sounding README (and even the examples!) aside, this definitely has that early 20th/late 19th-century aesthetic. I think 99% of it is due to the choice of font.
12 comments
[ 0.21 ms ] story [ 2.7 ms ] threadI just wish Tex were a cleaner language.
Which would be nice - some seminal publications from not all that long ago are only available as terrible quality scans.
XeLaTeX: https://nullpaste.org/mcKAFOqJmbZv
PDF output: https://www.dropbox.com/scl/fi/rfaio8a6pgem98fy8cxzb/princip...