Is Juris-M still active?

I will soon have much more of a need for "true" legal citations. I've until recently been primarily relying on the close-but-not-perfect citations generated by Zotero.

I can't tell if the Juris-M project is still active and if so, how far behind the true release it is.



As an aside, and I suspect this is an FAQ, why can't the codebase be merged eventually? Why does it need to be a fork?
  • It is still active, the developer is kindly still maintaining it and preparing several updates.
  • edited 8 days ago
    (I don't think it's a great idea to continue telling people it's being maintained and updates are coming until a new version is actually released. The last release was in 2023 and is based on Zotero 6.)
  • There is a Discord chat in which I can follow the progress of the development. As far as I can see, it had been paused for some time but was picked up recently.
  • Until there's a release or public code on GitHub, I think it's misleading to tell people that the software is maintained. As far as I can tell, there's been essentially one real commit on GitHub in the last several years. The software isn't maintained if the only code available to run is almost three years old, with no bug fixes or OS compatibility updates since then.

    As we've said, this functionality should be turned into a plugin (like the new multilingual plugin), not attempted as a fork of Zotero, an approach that stopped making sense many years ago.
  • edited 3 days ago
    Is Frank still on here? Maybe he can share some thoughts on this (Juris-M vs Cite Non-English (CNE)? Here's what Gemini 3 Pro tells me:

    [deleted — D.S.]
  • edited 2 days ago
    @sdspieg: Please don't post AI garbage here. If people want to ask an LLM a question, they can do so themselves, and then figure out whether what it's saying is true or not. Unverified point-in-time AI answers don't need to be immortalized here.

    E.g.,
    Juris-M: Because it is a fork, it sometimes lags behind the official Zotero updates.
    Jurism does not "sometimes lag behind" Zotero. It hasn't been updated in years. Until there is a new release, it should be considered abandoned. It's not helpful to tell people otherwise. (In a few days, Jurism will stop even being able to sync down items from official releases.)

    And no, Frank hasn't posted here in 5 years (to the day).
  • edited 2 days ago
    I'd be one of the last ones to claim that LLMs are 'perfect', BUT it seems to me that what I posted here was pretty accurate (or I wouldn't have posted it). So would you REALLY rather have NO info on this as opposed to a (human-curated - i.e. I DID feed it and had it analyze both full github repos, and I then still did a human 'small-check' before I posted) comment??!

    And btw - I've never understood this Zotero-'schism'. And why the main Zotero devs didn't try to keep Frank's (IMO quite valid) additions 'on board', because his use case (both the legal part AND the multilinguial part!!) IS so important the the global epistemic endeavor!!! I personally think Frank will clearly prove to be on the 'right' side of history on the need for better multilingual treatment in bibbliographic management. It is UNCONSCIONABLE how 'Western' scholars have ignored Chinese, Japanese. etc. etc. and yes - even -Russian scholarship. And Zotero COULD have bene on the forefront of this. As it HAS been on so many other issues. But it decided not to...

    But so I AM looking into the CNE repo, and I'll probaly use it; BUT from what I see, the 'interceptor' is quite fragile - I'm suprised you're ok with 'monkey patching' core Zotero functions like Zotero.Utilities.Item.itemToCSLJSON, as opposed to trying to implement this functionality more organically); not using caching may alos not be the best solution, etc, etc.

    But ok, I guess I'll abandon the Zotero forum on this then and will just move to Discord...
  • We explained many times over the years that we didn't feel we had the in-house subject-matter knowledge to implement either multilingual or legal citation functionality, and even if we did, Frank (as he would admit himself) was not a professional software developer whose code could reasonably be integrated into Zotero to be maintained in perpetuity by the core team.

    In any case, yes, we are much happier with people using a plugin than a fork, let alone an unmaintained fork. Monkey-patching isn't ideal, but that's why we've been adding more and more APIs for common integration points. (I would assume CNE is using the custom-field API we added specifically for developers to customize the item pane.) Plugin developers can request additional APIs to meet their needs.

    Even if we did try to build the functionality for either of these things ourselves, we would almost certainly implement it as a plugin rather than trying to add this incredibly complex functionality into the core product.
Sign In or Register to comment.