Multilingual Support intergration

I work across multiple languages and I imagine many of us do. Using non-Roman scripts often means having to offer translations/transcriptions (as per Chicago or MLA style, and many other styles). There's virtually no support for this kind of thing across citation software, except the Juris-M branch of Zotero, which makes me think Zotero might be the best platform to accommodate this kind of feature.

Is there any interest/possibility among Zotero developers to work on porting/creating add-on/plugin support for multilingual Zotero? I suppose it would mean bringing in some of the Juris-M solutions into it; is Frank Bennett still active with the Zotero community? His solution is still what is widely recommended by East Asia libraries in the US, for instance.

At any rate, I really hope more robust multilingual support can be integrated into mainline Zotero 6, which seems to have some amazing features with markdown support etc. It just seems to me, that given the globalizing intellectual environment, multilingual support really should be a priority!

Of course, I understand this is not necessarily easy or straight forward to do, esp. given staffing/time/priorities/funding... but I wonder: what would it take to make it happen? Grants? institutional support? additional team member? support from a library? faculty involvement?
  • Also PS:
    I notice that somewhere else in the comments it is said [Juris-M] is no longer being updated and Zotero 5 and 6 databases will no longer sync at some point. this is very worrying!
  • How complex are your needs?
    We can likely accommodate adding a translation to foreign-language titles within the current Zotero framework, but this would require separate citation styles (though only lightly modified & could arguably in the main style for things like MLA & Chicago) and be nowhere close to as flexible as Juris-M's multilingual integration.

    I don't have anything on the other questions, though generally speaking, adding massive new features is not just a money question: it's necessarily going to take up a good chunk of core development attention so it would need to be a development priority (unless it's possible as an add-on, which I'm skeptical about, but otoh, the Better BibTeX add-on does show that you can push those quite far)
  • So i don't know how Juris-M works under the hood, but so far the separate citation styles work fairly easily for Juris-M; I've been able to use Open CSL to create citations for it.

    From what I can tell, Juris M creates "extra" fields (all listed in extra in the vanilla Zotero database) that are tagged as alternatives to titles, authors, publisher etc. They are also tagged according to language codes (en for English, ja for Japanese etc) And there is an extra functionality that allows you to choose which one of these "alternatives" get exported. At least for my purposes:
    1. the ability to have alternative language fields for bibliographic information
    2. the ability to toggle which fields are used (or not used) when creating citations/bibliography
    3. the ability to mark what language these fields are in

    Would really all that would be necessary; the default Zotero citation styles seem to work perfectly fine with Juris-M in that case.

    It seems Frank has already developed the basic code for this-- is there a particular reason why this can't be an add-on or plugin?
  • as for development priority-- what determines what becomes a development priority? how much money are we talking? This is exploratory of course, but given the advantages of Zotero, I'm thinking it might really be worth some institutional investment. I work in Asian humanities, and one of the core reasons this thriving field has NOT adopted Zotero in the mainstream is precisely this multilingual issue. Can you believe many, if not most, of my colleagues do their bibliographies completely by hand? A sustainable solution to this problem would potentially benefit a lot of people.
  • +1 for this feathure
  • Thank you for @sw2090 for starting this conversation about multilingual support. Juris-M is helping me a lot for writing articles in various European languages, because I often need to localize several fields in citations: Author names (EN: Aristotle, DE: Aristoteles, FR: Aristote ...), publication places (Rome, Rom, Roma ...), and so on. I sometimes need transliterated titles (Greek/Latin).

    Because of the recent great improvements of Zotero (PDF annotations ...), which are seemingly not adopted by Juris-M in the near future, I am considering to move back to Zotero. If Zotero integrated such multilingual support, I would be extremely grateful!

    I do not need all the complicated legal support of Juris-M. Maybe these two projects (multilingual / legal support) could be discussed and implemented independent from each other?
  • This is a big +1 to @sw2090's suggestion. I have picked up and abandoned Zotero a few times just because it doesn't work well with non-Latin scripts. (I need to cite in Chinese and Arabic scripts, often in the same citation, and I know for a fact that my colleagues in Tibetan studies have negative experiences with Zotero's support for Tibetan and Indic scripts.) It's tough to cite a non-Western premodern book correctly, as well.

    This might be something to pursue through an NEH grant in collaboration with a non-Western book historian.
  • Another +1 for this. As much as I appreciate Frank Bennett's work in JURIS-M, I think the whole academic committee can benefit if we have a more stable support structure with using multiple languages, and not having to rely on one person's effort.
  • +1 for this feature
  • +1

    @sw2090 is right that this is a major impediment to Zotero adoption.

    I was an early adopted of the multilingual Zotero fork back in the day, and I loved it. I was sorry to see it go.

    And while I am genuinely amazed at the work of my former colleague, Frank Bennett, Juris-M will never be Zotero unless something enormous changes.

    For now, my workaround is to have separate entries in each language. I'm fortunate that the only Asian language I regularly work with is Japanese, so this is a relatively minor issue for me. However, I'm definitely in a shrinking minority as a younger generation of polyglots wrestle with 5, 6, and sometimes more non-alphabetic scripts to produce richer and better sourced scholarship than ever before. It would be fantastic if Zotero could support them.
  • Just wanted to add that Jurism still works for me for multilingual/transliteration support.

    I used Zotero throughout my doctoral program and towards the end in 2017, I started using Juris-M/Multingual Zotero.

    Even now, I still use the "Jurism" version of Zotero for reference management.

    I just did a test of creating a bibliography from a multilingual item and it still works fine for me.

    Everything except the special multilingual fields sync to a vanilla Zotero library in the cloud. When I log and view my library online (via the Zotero site), all my sources with multilingual fields are there like normal entries, but the special multilingual fields from Jurism are listed in the "Extra" section.

    So, as far as I can tell, Jurism is still a good option in the meantime.
  • yes Juris-M still works for me, but what spurred me to post about this was the recent move to Zotero 6 and the concern with backwards compatibility, since Juris-M relies on Zotero to sync. We can't assume that as Zotero versions keep emerging that this compatibility will remain. I think on the forums it was stated somewhere that backwards compatibility with Zotero 5 clients WILL end at some point in the future. Hence the urgency.
  • Yes, I would suggest that the multilingual parts be moved to a plugin, maintained by someone who (like Frank) works in multiple scripts.

    The wide variety of existing plugins shows that there's no reason someone should need to maintain a separate fork of Zotero for this.

    We're happy to provide whatever technical support we can to make it easier to hook into the necessary places in Zotero.
  • so maybe we can try to concretize this more:
    @dstillman when you say "we" do you mean the Zotero team?
    unfortunately, I'm not personally equipped to do this kind of work, but it is possible for me to try to scrounge up the necessary resources to make it possible. i.e. university support (I'm at UCLA), apply for grants etc....
    Is the this forum the best place to launch this discussion?
  • edited 10 days ago
    Honestly I don't think you can force this — someone sufficiently motivated and with sufficient domain and technical knowledge will have to come along and commit to developing and maintaining this indefinitely, and they would likely have to be the ones to seek out grant funding or university support if needed/desired. Basically, everyone Frank was and did, but to create a plugin instead of a fork.

    Even if you could get funding without a developer on board, which seems unlikely, I think it would be short-sighted to line up development funding without a longer-term maintenance plan. Someone has to want this and want to do this, just as with every plugin on the plugins page.
  • that makes a lot of sense. at any rate, without a developer it's unlikely anyone would be willing to fund such a grant.

    But I guess I'll just leave this point up here; if there's someone out in the void who would be interested in/capable of this undertaking, there's enough people out there who would work to make sure it is supported.
  • so the plugin/fork would both require long-term maintenance, but I assume Frank must have chosen to do a fork over a plugin for some set of technical reasons. Is there an advantage to making it a plugin? Easier to work with in the long term? or not?
  • edited 10 days ago
    Don't worry about Jurism. This should be a plugin (and there are countless more examples for how to create advanced plugins than there were 12+ years ago when MLZ/Jurism was first created).
  • okay! Well, I have my quest. I will have to find such a developer...
  • I would strongly support this as well. I would also add that in addition to working with sources in multiple languages, an issue for many scholars is the need to *write* in multiple languages as well, and to localize citations according to the target language. This requires the citation manager to use appropriate versions of the field--for example, when citing an East Asian-language title in an English-language publication, depending on the format being used the romanization, original script, and/or a translation may all be needed (or only one of these). Citations of the same source in an East Asian-language publication will use only the original script.

    Without the capacity to handle all these variations, Zotero is not just useful enough for scholars who work with/in these languages, who as noted above are more and more numerous. Juris-M is/was. If this functionality can be replicated with a plugin, ideally one that can also read/convert old Juris-M fields, I would certainly support it--and I expect that some institutions might as well.
Sign In or Register to comment.