Hacker News Reader

Comment by anigbrowl | Hacker News Reader

Quibble: I downloaded the pdf, which defaults to a file called 'hott-online.pdf'. Why not Homotopy Type Theory - 1st Edition.pdf'? I download a lot of pdfs (mainly because of being a law nerd), and I am sick to the back teeth of having to rename virtually everything that I download so that I'll be able to find the filename again later. I do use Mendeley to keep my documents more-or-less organized, but what have people got against human-readable filenames? Really I ought to be able to get semantic metadata for everything I download.

As for the work itself, great stuff, I look forward to reading it, or at least dipping into it (not a mathematician).

View on HN

I doubt I'll ever willingly include spaces in files names and I avoid punctuation too if I can.

If you've got a directory system that goes many levels deep though those long file/folder names won't fit on your screen and so you can't see at glance where things at.

I do sympathize with your plight though, I guess I've just accepted that if I want to find something I download again I have to rename or move it to where it ought to belong.

I would love if all filenames were human readable but it seems like too many people have a great desire to use bad scripts that can't handle spaces.

Zotero has a feature which will attempt to read metadata from a PDF and create a bibliographic reference for the PDF.

It also has a feature which allows you to rename the PDF based on the bibliographic reference it creates.

Using those two together has been a really easy way to earn brownie points with my professor boss.

Thanks for the tip.

A similar application exists for Gnome, called Referencer (https://launchpad.net/referencer). It keeps a library of PDF files and can extract metadata such as title and DOI.

Follow-up: now that I've had time to check, at least there's proper metadata in the .pdf properties. Meantime, my search for the Netscape Navigator of the Semantic Web continues :)

Sounds like a reasonable reason to open an issue.

Reply on news.ycombinator.com