Generate Better PDF Titles #32
Description
Currently, we just use the filename as the PDF title. However, it seems that many of them are just random numbers without a lot of semantic content.
However, there are some tools which can make better guesses at PDF titles automatically (e.g. https://github.com/metebalci/pdftitle). I think they basically look at 1) metadata and 2) large font text. This would make the search interface much nicer because at the moment you basically just have to click through the search results to see what's in them.