Advice on multilingual libraries

For a big project we are starting, we'd once again like to use JURIS-M, which - I take it - is the successor to MLZ (Multilingual Zotero), especially now that it has also been updated to the 5.0 version. The main reason for that is that we want to be able to quote the original Russian source we will be using; and that the main trunk of Zotero doesn't - alas - offer that functionality. [I have noted before how I find it absolutely irresponsible that the international scholarly community continues to tolerate NON-multilingual citations, but I'll leave that rant aside]  And BTW - the reason why we still want to keep using the regular Zotero, is because we'd like to be able to use various incredibly useful add-ons like Zotfile, which presumably will not work on JM.

I'd appreciate it if people could provide us with some advice on a few points:


  • is it possible to run both programs on the same computer with both of them using the same Zotero data folder? I guess not, because of the different fields used in both programs, etc. But I still wanted to ask the question, because Frank mentioned somewhere that these different data schemes do somehow get normalized in the syncing process - so I don't see why that same process could not be done locally between the two 'trunks' of Zotero;

  • are there any remaining issues on using Zotero/JM in documents in collaborative teams in which some members are using different trunk? (see also the discussion here);

  • where do things stand now with (preferably automatically) transliterated fields in JM? (see a previous discussion on this here);

  • any other caveats or do's/don'ts?

It's great to see that JM continues to 'play nice' with Zotero. I have asked many times before why Zotero itself could not include some of the very useful (at least for us - and we're not lawyers...) additional functionalities that JM offers. Both Dan and Frank claim that it's technically too hard. In my, infinitely less informed, view, it seems like Frank keeps proving that it IS possible, because his trunks DO work. But anyway, since:


  • there are two different programs out there now;

  • they both offer some disctinct functionalities that at least some of us find useful; and

  • since they still 'play nice' with each other -

it would still be extremely useful to get some concrete pointers on how those of us who want the best of both worlds should proceed. 

Thanks!

 

  • I'll note in starting that as far as I know ZotFile, Zutilo, and Better BibTex, the most widely used Zotero add-ons, all work with Juris-m and if there are other add-ons that do not, they typically only require the addition of juris-m to their install script and no other code changes.
    is it possible to run both programs on the same computer with both of them using the same Zotero data folder?
    Not possible, no. Zotero won't start up once you've used a database with Juris-m once because the database doesn't conform to the Zotero schema anymore.

    I don't have answers to the other questions. @fbennett would be in the best position to chime in.

    The problem btw. is not so much that it's technically "too hard" to integrate juris-m into Zotero but that it's massively ambitious to reintegrate many, many thousands of lines of code back into regular Zotero and there are just no resources to do that.



  • Juris-M and Zotero will not run simultaneously, but they can use the same data folder. Juris-M holds metadata in a database jurism.sqlite, while Zotero data is held in zotero.sqlite. If both are synced to the same Zotero account, they will contain the same data. Juris-M records will show some machine-speak code in the Extra field when viewed in Zotero, and you shouldn't touch that, but that should be the only limitation.
  • I haven't done anything about automatic transliteration of fields in Juris-M, and don't have any plans to do so. If you have code for it, it shouldn't be terribly hard to cast a plugin that supplies that functionality for one or more languages. In the spirit of not doing things twice, though, it would probably make sense to wait on it until Zotero migrates away from Firefox infrastructure.
  • edited February 28, 2018
    As @adamsmith says, most Zotero plugins should work fine with Juris-M, so long as compatibility flags are set in their installers. (There is one exception in the other direction, at the moment: the Abbreviation Filter currently works with JM 5.0, but not with Z 5.0, due to some tweaks needed to give the citation processor access to abbreviations pulled out of database storage.)
  • Regarding merger of JM functionality into Zotero, that may well happen at some point down the road, although it's a tall order. It would require a lot of developer time, and it would also have an impact on the landscape of support. If it were me, I would also look very carefully before jumping there. Meanwhile, though, JM has made the jump to 5.0, and I'm working lately to set up over-the-wire updates, a move that is long overdue.
  • Thanks Sebastian and Frank! Very useful. So if those plugins do work on JM (as long as the compatibility flags are set in their installers), we might still decide to go with that option. We'll still think about the transliteration bit, but we're really too small a player in all of this to really make a difference :(
    And as to a potential future merger - I still hope somebody will provide some funding for this. If there's ever a potential funder, please let me know, and I will write a glowing (and substantiated!) recommendation...
  • I think we can say pretty conclusively at this point that any sort of large-scale merger isn't going to happen. Frank laid out some of what would be involved in a previous thread of yours. Even beyond the monumental effort such a merge would require, we just don't have the domain expertise to maintain this functionality, and I think it would quickly become a serious impediment to ongoing Zotero development. Zotero's open-source nature will allow Juris-M to continue to exist and be updated by Frank and others who understand the requirements and can put in the time to maintain it, and I don't think there's anything wrong with that.

    As Frank notes there, support for additional fields might go a little ways to address some of this within Zotero itself, perhaps paired with some externally maintained data files. But I don't think it's realistic to expect much beyond that.
  • The BBT status for Juris-M5 is currently unknown because I haven't been able to start Juris-M in my test environment; someone is working to make that happen (https://github.com/Juris-M/zotero/issues/34) but I don't have the time to dive into why Juris-M doesn't run where Zotero does in the same env.
  • Pretty pressed for time at this end also, but I've pinged Duncan to sound out the urgency level.
  • BBT passes tests on Juris-M, new release out.
Sign In or Register to comment.