[hirenj]

I really love the idea of ditching reference managers. The utility of these three scripts (or something like them) really is demonstrated when on mobile.

RSS feeds, Twitter (or emails from colleagues) give me a raw list of papers to read. Using the url of whatever journal webpage as a basis it then gets converted to a doi by the Zotero translation server (hosted as a lambda on AWS). From there its the crossref API to generate a filename for the PDF, and then Elbakyan's resource to download the actual file (even though I'm usually on an academic network that has full access, the scihub api is cleaner). All this gets launched from the share sheet from the browser, and deposits files in a cloud drive.

A little bit off topic, but could anyone explain to me the actual utility of a reference manager if all your references have dois? I don't get what it provides over a folder of PDFs with annotations?

[btrettel]

I use Zotero for references but keep the files in a separate organized directory structure, where the filenames match the citation key in Zotero. Each approach has its own advantages and disadvantages.

The largest advantage of the directories is speed. I don't need to wait for Zotero to launch, and Zotero has gotten much clunkier over the years. (I think they need to do an entire rewrite at this point.) I also have some bash aliases to, for example, open a PDF file I have in my reference folder given only the citation key.

Zotero is better for annotations, and other information associated with a citation. I make very heavy use of the related documents pane. I also often put citations in multiple folders in Zotero. That's a bit more complicated in the directories as I need to add links and then maintain the links. (I have a few shell scripts to help with the creation and maintenance of the links.)

Zotero also makes generating bibliographies in different formats easy. I normally use BibLaTeX format, but a journal I'm submitting some things to requires BibTeX. The conversion was dead simple in Zotero. If you manually curate your bibliography file then this would be a pain.

Also, I have many citations in Zotero where I don't have a copy of the document. Not all documents have DOIs, and not all documents have been digitized. I use Zotero extensively when visiting the library so I can keep track of which documents to scan.

[SebiH]

> I use Zotero for references but keep the files in a separate organized directory structure, where the filenames match the citation key in Zotero. Each approach has its own advantages and disadvantages.

Why not install the Zotfile plugin [1]? You can configure it to do exactly that, and point it to a Dropbox folder to get better synchronisation between devices.

[1]: http://zotfile.com/

[btrettel]

I wasn't aware of Zotfile. I'll take a look. Thanks. Hopefully it has some way to automatically associate with ~10,000 files...

[ajot]

Well, I do use Zotero for automating the bibliography making (including re-submitting a paper to another journal which uses a different citation style) and it's availability for MS Word and LO Writer users, which let's me send my manuscripts to my PI without a bigger hassle.

[hirenj]

I'm trying to establish convention in our lab that bibliography generation is the last step before submission. It always ends up that the bibliography (ie field codes with ref info in Word) is the most fragile part of the document, and everyone that isn't the person that generated the bib is terrified of changing anything in it for fear of ruining the references.

It seems a lot more sane to use nicknames for references and then format references right at the end when you know it's not going to be edited any further. It still seems like it's an unsolved problem in the process.