Juris-M, the need is real!

Dear Develop Team,

After being failed to download and install Juris-M, there is a question jumping up my head: why not?

Zotero is great and I can't thank you all enough to develop and maintain it.
However, while Zotero is working well with a monolingual document which is normal with English users, the situation of users whose mother languages are not English is a little bit complicated.

There are many cases, such as theses or national journals, the bibliographies are needed to be sorted into groups of languages used.
I suppose that this issue is mentioned sometimes in this forum.

For example:

Language A:
1. Author 1 (Last - Middle - First name)...
2. Author 2 (Last - Middle - First name)...
(sorted by the First name)

Language B (mostly English)
3. Author 3 (First -Middle-Last name)...
4. Author 4 (First -Middle-Last name)...

(Sorted by the Last name)

It is a headache to sort like that at the same time.
Usually, I have to make sure that the English document list is correct, then amending the other list by hand which is time-consuming.

Juris-M is supposed to solve that but specifying in the law field.
I read in some discussions that Juris-M is not for the majority, that might be true.
Nevertheless, I suppose that the above requirement is not the minority because there are also many users like me who have to use more than one language for my manuscripts.
Hence, I really hope that feature could be implemented into the nearest update of Zotero.

Wish you all the best!

P/s: Any idea how to install Juris-M while the server is down, please?
  • There doesn’t seem to be much activity around Juris-M lately, it hasn’t been updated since 5.0.93 (Zotero is now on 5.0.96) and for the last few days downloading or updating the app has not been possible, not sure why.

    Personally I also don’t really understand why Juris-M isn’t just integrated into Zotero.
  • I agree that multicultural and, therefore, multi-language support are important, practically and ethically.
  • There are essentially three ways in which Juris-m diverges from Zotero:
    1. It has improved legal support with not just additional item types but also things like modular elements in styles that behave differently depending on jurisdiction, parallel citations as used in anglosaxon legal writing, etc. Zotero has made pretty clear that they don't think they have the right expertise to implement the full complexity of international legal citation, so that's just not going to happen.
    2. The original "Multilingual" part of Juris m, which allows the addition of translations and transliterations of titles and names to items. I don't think there's currently a statement from the Zotero devs about this, but while it's very useful, it's also quite messy GUI-wise, it's not at all rooted in the current data model, etc., so this would be an absolutely massive undertaking for a relatively small group of users: contrary to what some of you are implying, these aren't people working in non-English, non-Western, or multiple languages, but it's people who work in traditions that include a lot of going back and forth between translation/transliteration/original titles, a much smaller subset of the above groups.
    3. Juris-m uses its own version of CSL and the developer writes the citation processor himself, so he can make changes to CSL more quickly than Zotero which, like 30+ different tools using at least 5 different citation processors uses standard CSL. This means some of the more lightweight changes that have existed in Juris-m will definitely make it to Zotero. The ability to sort bibliographies by language as mentioned by the OP (which is already possible in Zotero using unofficial/nonvalidating styles btw., there are a number of existing threads here with directions) is one of those things, as is the inclusion of some additional legal item types (treaty e.g. will likely be in the next major version of Zotero) and very likely some lightweight additional support for multilingual writing such as the ability to include translated titles, but that one is further out.
  • With respect to multilingual bibliographies, much of the work for developing a streamlined version has been done for regular CSL and will likely be incorporated into CSL 1.1 or 1.2.
  • PubMed metadata provides non-English language article titles with an English-language title within square brackets. It also provides the title in the original language. At my university several faculty insist on both the original-language title and the English translation be included in a citation.

    Is it likely that the new CSL and Zotero's implementation will eventually facilitate this 2-title business? The original language is provided with the nbib export.
  • yes, but that's not with the next big update but 1-2 down the road.
  • edited June 15, 2021
    Thank you all for your comments here!
    I agree that Juris-M is overkill for most of users, but the official support for the ability of displaying bibliographies depending on language is really essential.

    @adamsmith : I am sorry if I overlooked this in the search results, but as far as I know, while sorting by language can be done easily by adding

    "<"key variable="language""/>" (I know it is wrong code, but not sure why it is hidden without adding ")

    there is more complex requirement in the displaying of authors' names and sorting by Last name vs First name depending on the language groups (like one in my example). Can it be done by editing CSL file?
  • edited June 15, 2021
    I'm not 100% sure on the sorting, but in general yes, separate bibliography rules by language are possible just like in Juris-M using the same syntax.

    Edit: For Chinese (and I think Japanese) in particular, when non transliterated, Zotero will already automatically handle name display and sorting correctly when names are entered in reverse order (so lastname: 毛 firstname: 泽东 -- think of this as "family name" and "given name" instead). That will display & sort 毛泽东 in the bibliography and shorten to 毛 with form="short"
  • edited June 15, 2021
    Yes, that is what I did till now as a workaround (my case is Vietnamese). In my example, I didn't mention other things related to language, that is:

    1. Family name - Given name (in full, without initials, i.e. Nguyen Quang Hung)

    3. Given name in initials form - Family name (i.e. J. Doe)

    Could you please tell me how to do that?
  • You can apply any of the CSLm extensions for language-specific name formatting in a style in Zotero: https://citeproc-js.readthedocs.io/en/latest/csl-m/
  • Juris-M is still being actively developed (though the download links are currently down). The latest production version is 5.0.93, but the beta is 5.0.96.
    Name sorting is an ultimately intractable problem because different authors prefer different orders, especially people with East Asian names publishing in non-Asian languages, but generally Juris-M gets it right.
  • edited June 30, 2021
    @bwiernik :

    Hi, I look at the link https://citeproc-js.readthedocs.io/en/latest/csl-m/
    and notice a language-name extension, but I don't understand how to use it.

    "language-name (extension)
    The ISO (or BCP 47) language code of the resource. The value is set to the target of a vector symbol (">" or "
    <script src="https://gist.github.com/quochung2310/f68b5eaee5fca2223daf43dd702b0a87.js">"</script>;
    (Line 17 to 38 macro author)

    I want the output will look like this:

    (Vietnamese documents)
    1. Nguyen Van A ...
    (Last - Middle - First name in real-life order, but I already typed in reversed order in the database, so Zotero understand that A is Last name, Nguyen Quang is first name, then sorting by A)

    2. Vo Van B
    3. Le Thi C
    (the sort order is A-B-C)

    (English document)
    4. Doe J.K...
    5. William G.Q....

    Could you please take a look at that?
    Appreciate your help!

    P/s: how to type code into this comment? I tried to type code, but it is hidden?
  • enclose word "code" in angle brackets to open and close with a closing tag containing the word code.

    <code> 1234 </code and end the close code with a greater than angle bracket
  • Hi. Can someone help me to start mastering adaptive lacalization of terms in Juris-M? I have created a test style (for journal articles only) - test.csl and attached three references. The references are identical at first sight, but one has the language listed as "ru", one as "en", one as "eng". I would be very grateful if you could show me an example (using my code) how I can change "et al" to "и соавт." for the language "ru", but keep "et al" for the language "eng". :'-/
  • I am afraid that this is not possible. I think you can only set the "et al" language for all the citations within a specific Word document. In "Set document preferences" from the Jurism Word processor plugin, you can choose "Russian" or "English" etc. in the language drop-down field below the style list.
    Regarding the Jurism-specific multilingual options, see: https://juris-m.github.io/posts/2020-11-08-0.html
Sign In or Register to comment.