Can I add paper links to Zotero without ever downloading the PDF?

skosch
Hi,

I'm looking for an easy way to bookmark arXiv papers that I read in Firefox alongside their metadata. I don't want to have Zotero running in the background, nor do I ever want to download any PDF to my hard drive (nor does it need to be stored in Zotero's cloud), as a simple link is always enough to get me back to the original arXiv pdf. I also don't care about pdf versions or annotations.

I have already authorized "Save to Zotero.org" and unchecked "Automatically download associated PDFs and other files when saving items". Yet my Firefox connector only offers "Save to Zotero (PDF)" and that fails.

Is there anything else I need to do? Thank you!

Sebastian
  • dstillman
    edited March 2, 2021
    If it's showing "Save to Zotero (PDF)", it sounds like you're on the PDF, not the article page. You should always save from the article page when possible, and certainly if you don't want to save the PDF.

    If you uncheck the PDF option in the Zotero Connector preferences and save from the article page with Zotero closed, you should get only the metadata in your online library.
  • skosch
    Wow, thanks so much for the super fast reply. Saving from the article page worked indeed, exactly what I was looking for!

    The process is really not ideal though – I don't know if I want to bookmark an article before I've at least skimmed it, and by that time I don't have the article page anymore because I've clicked through it to get to the PDF.

    I suppose the solution is to add to the browser extensions the ability to save PDFs without actually attaching the files – just the metadata, plus a direct link to the PDF, would be enough. Is that feasible/worth opening a Github issue about? (I'm surprised that this isn't what most people want to do by default, is my workflow that unusual?)
  • adamsmith
    Both the fact that you're not interested in saving PDFs and that you're not saving from the abstract page are quite unusual, yes.

    Where possible (which is typically where the PDF's URL includes a useable unique identifier), we actually do allow saving just the item from the PDF page and that's actually the case for arXiv, but the Firefox add-on framework doesn't allow for any of this on PDFs, so that only works on Edge/Chrome/Chromium. In other words, this is already supported to the extent possible. No need for a spearate github issue.
  • dstillman
    @adamsmith: It doesn't look like we're doing translator saving on arXiv PDF URLs (though arguably we could) — that's why @skosch is seeing "Save to Zotero (PDF)", not "Save to Zotero (arXiv)".

    @skosch:
    by that time I don't have the article page anymore because I've clicked through it to get to the PDF
    I mean, you just click Back.

    If you save from the PDF, you're always going to get the PDF, and that's not going to change — Zotero needs the PDF to retrieve the metadata when you don't save from the article page. The PDF could theoretically be deleted after metadata retrieval, but that's much more inefficient (have to download the PDF, have to run a whole separate metadata retrieval step which may or may not work) than just clicking Back and clicking on the save button from the place where the metadata exists.

    And PDF metadata retrieval is a function of the Zotero desktop app, not the website, so that wouldn't work saving directly to zotero.org anyway.

    To echo adamsmith, what you're asking for is unusual, yes. Zotero is designed to work primarily with the desktop app open, and it's designed to save primarily from article pages, which contain the actual metadata. You can save directly to zotero.org and you can save directly from PDFs, but those aren't primary workflows.
  • skosch
    You guys are incredible, thank you both for replying so quickly. The Chrome tip is super helpful, and so is the background about the desktop app/website separation.

    I'll take to heart what you said about my workflow being unusual :) I will try and figure out a way to make things work for myself. Thanks again for your support.
  • adamsmith
    It doesn't look like we're doing translator saving on arXiv PDF URLs (though arguably we could) — that's why @skosch is seeing "Save to Zotero (PDF)", not "Save to Zotero (arXiv)".
    Yes we are (see https://github.com/zotero/translators/blob/master/arXiv.org.js#L118 ), I tested that this works in Edge before posting
  • dstillman
    Oh, weird, I seem to actually be getting different PDF URLs in different browsers.

    The PDF link in this article is loading https://arxiv.org/pdf/2004.13001.pdf (which the translator works on) for me in Chrome and but is redirecting to https://arxiv.org/ftp/arxiv/papers/2004/2004.13001.pdf (which it doesn't work on) in Firefox.
  • skosch
    Interesting, the translator thing is just what I had in mind originally. But I only see the PDF option in both Firefox and Chrome (can't test on Edge as I'm on Linux)
  • adamsmith
    edited March 2, 2021
    Chrome and Edge should be identical (Edge is really "Edgium" now). I'm seeing unpredicable redirects to the ftp/arxiv/papers URLs for PDFs -- e.g. the above redirects for me on all browsers, other PDFs on none of them.

    In any case, I think we should be able to get the ID from the ftp URL format, too. Will post here when that's the case.
Sign In or Register to comment.