Zotero & altmetrics — any update?

edited November 22, 2017
Reading this new arXiv paper on bibliometrics and altmetrics, I am reminded again of the initiative, announced a few years back, to open up Zotero bibliometric data to altmetrics APIs. This would provide enormously useful data about the relative impact and citation potential of publications, and given Zotero's diverse user base with a strong footing in the humanities, would likely provide a useful corrective to data currently used by applications like Altmetric and ImpactStory.

I'm a proud user of Zotero and its a tad bit disappointing to see that Mendeley has successfully cornered this field and that Zotero isn't (visibly) contributing anything here so far — Mendeley is a for profit tool and anyone building on their database faces the possibility that they'll want to monetise it at some point. Zotero on the other hand has been open from the start, and it would be fabulous if it could open up some of its bibliometric data (properly anonymised, yada yada) to the altmetrics community. A secondary benefit of this might be that it would demonstrate that Mendeley isn't the only game in town and that Zotero is worth checking out.
  • I just got the 5.0 update, hoping it would finally be there. but no ... nothing yet,. lets keep hoping on those integrations.
  • (bump)

    (sorry to be asking for info without being able to offer help — I have a feeling that
    maybe everybody wanted this to succeed but somehow it didn't materialise and has now slipped down the priority list.)

    Genuinely curious what happened to this and whether it's still on the cards, for reasons noted in my post above.
  • This is definitely still happening, but the public release is still a little ways off. We're hoping to have something out in the next few months. (We have a few other releases in the pipeline, and then we'll be shifting some dev time to finish this up.)
  • @dstillman Awesome news! Any further details on what it might look like?
  • Was reminded of this during the Mendeley local db encryption brouhaha...
  • Seems like there's still no progress on this front. Is it on the roadmap at all?
  • Wondering same
  • Hi everyone! Is there any update? Thanks
  • I've not read the paper in the original request here, but what data does Mendeley use (I take it https://www.altmetric.com/products/altmetric-api/, but maybe also others), and what results does it produce from it?
  • You can see via Altmetrics how many people have a given article in their Mendeley library. It'd be nice to have that available for Zotero as well.
  • Does altmetrics provide the backend for that then?
  • No, Mendeley provides usage data (via public API) to Altmetrics, which in turn bundles them with other metrics and provides them as a service.

    The Mendeley "reader count" variable is documented here: https://dev.mendeley.com/methods/#additional-catalog-document-attributes
  • Yeah, that's sort of what I meant with backend. There's no service for Zotero to run, Zotero could feed data into altmetrics, and get aggregate data back, right?

    I'll take a look at the API.
  • Do you need some kind of subscription to use altmetric? I try to look at the API docs but I get

    Email us at support@altmetric.com to find out if your organization has an Altmetric Explorer subscription.

    I'm pretty sure our organization doesn't have a subscription.
  • edited October 16, 2019
    The main use case (and the reason I opened this thread) is on the altmetrics side: the promise to source anonymous usage data from Zotero to feed into alternative metrics of scholarly impact. Here's the 2015 (!) blog post describing the basic idea: https://www.zotero.org/blog/studying-the-altmetrics-of-zotero-data/

    Mendeley is still the only partner doing this right now; the benefit for them I suppose is name exposure for a for-profit product. It would be lovely if Zotero could join the fray as it has a more diverse user base (I think) and in general altmetrics are lacking in robust measures of scholarly uses. For instance it's easy to find how often an article is tweeted about, but it would be at least as interesting (and a better measure of expected citation impact) to see how many folks have saved an article in their reference manager.

    The 2015 post describes a pilot which was then funded for wider implementation. A lot has changed since then on both sides so I'd be (pleasantly) surprised if it were still being worked on in that form.
  • Looking at that blog post, I don't know how useful it would be to have this in a plugin. If the interest is aggregate data, a plugin would get you really low (and potentially biased) coverage, because you'd only get data from people who a) knew about it, and b) took the effort to install it.

    Also you'd still need an altmetric account of sort to just look at the API, and I don't have one.
  • edited October 23, 2019
    I agree that a plugin is unlikely, unfortunately, to provide much help. My hope and that of others, based on the initial blog post and the example shown by Mendeley, was that Zotero would be able to make available and document a public API which parties like Altmetric, but also Dimensions.ai, ImpactStory, or SemanticScholar could tap into to get anonymised aggregate stats for relative frequency of scholarly outputs in libraries. It is still unclear how much progress was booked on this front and whether this is still on the cards in any way.
  • Any update on this?
Sign In or Register to comment.