Multilingual Zotero - how to keep both the original author name and its translation in Word

I was working with Multilingual Zotero in MS Word and the problem I confronted with is that when I generate references (Chicago style - full note) Zotero gives only two options: either to keep the author's name in its original language or to translate it.
I was wondering whether it is possible to have both authors' names (in original and translated) similarly to titles (in case of which Zotero keeps the original title and adds translation in "[ ]").
Hope for your assistance.
  • note that multilingual Zotero is experimental pre-alpha. Once this is actually fully implemented the type of things you want will be possible, but that's still a bit to go.
  • Thanks, and what about editing the citation styles? It is possible to resolve the issue by changing code in the CSL somehow?
  • See http://gsl-nagoya-u.net/http/pub/citeproc-doc.html#id36 for the way this is currently determined. Judging from that page, I think that displaying multiple versions of author names is currently not supported. I think that the range of transliteration and translation options will eventually be supported for all fields, but for now there's not much you can do.

    Is there a specific style guide that requests multiple versions of author names in this way?
  • Judging from the examples given in the citeproc-js manual and the multilingual Zotero page (http://gsl-nagoya-u.net/http/pub/zotero-multilingual-overview.html#multilingual-citations), it looks like this should be possible already.

    Do you have checkboxes ticked in the Name Transliteration column of the multilingual pane of the Zotero preferences?
  • Multiple versions of personal names is not yet supported, but I plan to add it. One of the hard science publishers (I forget which) has (sensibly) adopted a policy of allowing transliterated author names to be supplemented with the full name in the original script, to avoid information loss. It will take a bit of work, and will need to be carefully tested, but it's certainly on the list.

    Before we get there, the data model (the way multilingual metadata is stored in the database) needs to be reworked. That will break syncing of multilingual metadata in the short term, but it's the right thing to do, and should be the last major change to the multilingual version, giving us a stable foundation from which to tidy up details like this one.

    (Apart from that, I'll reiterate what adamsmith says above, that this is an experimental version, that should not be relied upon for production work or the storage of one's main repository of Zotero items.)
  • Thank you for your answers!
    Dear, ajlyon, I tried to have checkboxes with the Name Transliteration ticked and vice versa, however, in any case the original name in original language is not kept.
  • Gentlemen, thanks for all of your help. We did/do realize that this is an experimental version, but it was the only we could at least partially execute what we wanted to do. And we were (mostly) successful. Please do keep this work up. We will undoubtedly be seeing much more multilingual work being done and we will have to be able to properly source this. I presume the standard styles will have to adjust to this too!

    -Stephan (also on behalf of Nadya and all other members of our team)
  • Yes, the styles will need to be changed some, once the syntax for specifying this kind of behavior is settled upon by the CSL community.

    I think that you should be OK if you use multilingual branch to store data for now-- perhaps Frank can clarify whether he anticipates it being possible to cleanly migrate data from the experimental version to the final release. As has been noted, migrating through RDF export should work regardless.
  • I'm starting to work on the revisions to multilingual data storage. The surgery involved is fairly drastic, and transparent migration of data to the new format via Zotero upgrade probably won't be supported. RDF export and re-import will (definitely work and) will probably be the only migration path, at least in the short term. Migration in that way will entail hand-re-linking of references in any documents that were created with the experimental version. We'll see how things go, but that should probably be taken as the baseline assumption.
  • Stephan, et al.

    I've put in some work moving the multilingual branch code to the new data model, and things seem to be working out pretty well with it. Having been through the code one more time, I think the prospects of transparently converting databases created by the current XPI version to the new data model are fairly good. It will be another week or three before I'll be able to say anything definite, but in principle it should be possible, and I'd like to see it through to implementation.

    (Caveats aside, I'm very glad to hear that people are already attracted to the multilingual version, and I'd like to do what I can to make the experience a good one.)
  • Stephan (sdspieg) et al.

    The XPI installer for an upgrade to the multilingual version is now available. The new version (like the old one) should be treated with care and caution, but it does include an upgrade step to convert the SQL data from the old to the new data model, in a way that should preserve links to existing documents. If you try that out on a copy of your data (being careful to back up both your database and any documents exposed to the new version), do post a note back to let us know how it works for you.
  • Frank,

    First off - thanks for your efforts. We are having another go at this now and will report back.

    -Stephan
  • Hello,

    I am rather new to Zotero as well as the Multilingual function. Could someone go through how to input data with the different language tabs and how to make a correct citation for this? I've tried to do it following the example on the nagoya-u.net site but can't seem to make my citations look right. The names keep coming out in the wrong order and I also can't seem to figure out how to insert the kanji as well as translated English title without italics. Sorry for the rookie question. Thanks very much!
  • No worries, and very glad to help.

    Can you put up a screenshot of what you have at the moment (the Preferences -> Languages menu), and a view of an item? Not sure where is good -- if you want to use twitter with twitpic.com, my twitter handle is @fgbjr. Follow me, I'll follow back.
  • Thanks for the quick response. And many thanks for creating this function for Zotero! I put up the screenshots on twitpic as suggested. Thanks again!
  • It's pretty astonishing to see new material appearing in screen layouts that I only saw myself a couple of days ago, but that all looks correct. The italics are a product of the style, and are applied across the whole data segment (so if you have italics applied to the title, you get italics for both the main title and the translation). If that's not correct, we should probably spend a bit of time collecting examples and learning about publishing conventions before settling on exactly how to address it; but it can certainly be fixed, if it's broken.

    For sorting, what issues are you finding? In the configuration shown in your screenshots, the CSL processor should sort things on either English, if that's available, and otherwise use the headline field content.
  • One thing that is easy to miss is setting the language of headline entries-- you can left-click on the label in the info pane to set the language of the main entry. It's probably good to do this when adding variants in the several languages you have set up.

    Speaking of the recently changed interface, it's great. Except that you seem to have disabled scrolling, which causes a problem for people like me: http://twitpic.com/3uhu4v
  • Oh, that's not good. I'll take a look.
  • @ajlyon,

    Should be better in the latest xpi, just uploaded.
  • I'm still seeing m2809 there -- shouldn't it be m7897?
  • Yes, I was lazy. A bug report against the processor has just come in so it might be awhile, but I plan to set up proper automatic updates for the multilingual xpi. When I set that up, I'll make the version numbering update automatically as well.
  • Actually, the interface isn't fixed, even after I install the XPI. Therefore I thought maybe it's the XPI.
  • Hmm. I've bumped the version number and pushed it up again. This will get better when I've built the update/publishing script.
  • Thanks for the clarification. I see what I did wrong with the italics. But how do you change the name order for citations? I'm not sure what the "headline field content" is. Sorry again if this seems obvious and I'm just missing it.
  • In your sample, you have the authors written in romanized form. If that's all that's available for an item, it will be used for sorting. For Japanese sources, you would write the author names in kanji (or katakana), and then add the romanized form as a ja-alalc97 variant, in the same way you did for the title field. Then when you have ja-alalc97 selected as a preferred sort form, it will be used to generate sort keys.

    That's what's intended; if it doesn't work, let me know and I'll take a look.
  • Thanks very much! That solved the problem. Sorry for the confusion and thanks again.
  • Very glad to help, I'm glad to hear it's working for you!
Sign In or Register to comment.