Re: [tlug] Unicode/ICU question about joining lines

[Darren Cook <darren@example.com>]

> It is, it just doesn't get much traffic because all linguistic
> problems are straightforward. ;-)

:-) The problem described by Travis is one that comes up in various
places, but I struggle with it most in PDF files. Whenever I tell people
that PDF to Text extraction is a hard problem, research-level problem,
and there are even academic conferences for it, they look at me like
obviously I'm an idiot and haven't tried applying Algorithms to it.
(https://xkcd.com/1831/)

Another key place it needs to be dealt with is in OCR. And for pulling
out paragraphs from emails. And now I'll add restoring git commits to
the list!

Darren