Creating a list of all my PDFs that auto updates?



Zotero is free, and seems to do exactly what you want, unless I’m missing something.


It’s probably the better approach. I just wanted to nerd around in the shell to see how to recursively pull metadata from files…

Calibre is also a great tool. You can download metadata, edit it, rename the PDFs with proper names and export the whole thing as BibTeX or whatever.


I suppose it depends on what information is important. Some people in some disciplines seem to really find the journal to be important. For me the journal is largely irrelevant so I go

Author(s) YEAR - Title

I usually shorten the title if it’s lengthy, choosing whatever portion of the title is the most descriptive. If there’s more than two authors I do et al.

I don’t use any metadata. I use DEVONthink mostly for managing, so I can search file names or contents of files and usually get what I need without being concerned with formalized metadata. Houdaspot would do the same job if you prefer to use Finder for managing your articles.

Looks like maybe you’ve found a good solution to enhance the terminal output, I might have to dig into @Lars’ suggestion myself just for fun!


I do more or less the same…actually BibDesk autofikss it for me using that convention, into folders by letter.


@dfay I’d love to take a look at the database.

@JohnAtl I haven’t tried Zotero in about 5 years. I discovered Papers and have been using that for the last 5 years. I’m toying with Bookends and it’s integration with Devonthink. I guess part of my hesitation of using one of these citation managers is that I’d have to teach and reteach this to my students (since they’re undergrads they usually stick around a year at most).

@dfay that’s what I use too. I was curious to see if you had some naming convention that helped Devonthink.


I’ve been mulling over why I chose to approach this problem in such an unproductive way. I was stuck with trying to organize what I have (project folder/lit review folder/pdfs) rather than restructuring my system. It seems obvious now, but I blind to this solution. Even when members tried to help point it out as the practical solution.

I’ll probably go with a free citation manager or comparable app (e.g. Zotero, Mendeley, or Air Table with box links to the articles). I’m playing with Mendeley and tagging projects (so that students we can quickly look up which articles relate to which projects).

Lars, I’m still going to try what you suggested! Just for the fun of it… hah! I didn’t have time until now because I was taking care of my daughter alone for the past 2 days.


If this is the case, the easiest solution may be to build something on the back of an existing course management solution if there’s one at your institution.

I haven’t used Mendeley beyond basic testing - if it works well do report back.

This is the kind of hidden service work that will probably count nothing for your tenure / promotion file but which can eat up a lot of time.

In re the FM database - I can post it but it will take a bit of cleanup and documentation. Bear in mind it was created to replace an abandoned mobile app (PocketBib), isn’t oriented towards sharing / multi-user scenarios, and is a mirror not a sync solution, i.e. the original data gets copied in but can’t changed (well it can but won’t sync back).


Good point. I can see this taking a huge amount of my time. I think I’ll stick with the existing options for now. So far Mendeley has had issues with syncing just one citation in a group I created (the pdf was missing) on two of my computers with the same account and Zotero synced right away.


Unless your institution provides Mendeley for free I’d recommend Zotero. It works well, and those who don’t own a citation manager get to try a good one for free. Plus you can set it up to reference files stored in a WebDAV location meaning it’s completely free to use.


To add to @RosemaryOrchard’s very good points, if you happen to have a student or collaborator that is already familiar with Mendeley, then Zotero is no real jump – they’re functionally very similar, so knowing one gets you very close to knowing another, and for the most part, references are pretty portable, so was to export from one and import to another application.