Multilingual Plugin (MLZ) doesn't import Japanese Langauge parallel fields.

I have been playing with Multilingual Zotero and love it. But one problem I keep having is that it will not simultaneously import Japanese and Romaji information. It will populate the fields with Just the Japanese or English, but not both. Even though I can see in the MARC record that there are parallel fields linked together.

I can't figure out if this is because of the databases I've been using, which includes major ones like CiNii and the National Diet Library's search services. Or maybe I haven't configured MLZ correctly?

Any advice? Thanks all.
  • Ah, the correct term I was looking for was 'Populate', not Import. I'm trying to populate the Zotero fields (with the MLZ plugin installed of course). Despite the claim in this article:

    http://www.zotero.org/blog/new-release-multilingual-zotero-with-duplicates-detection/

    The article above also mentions 'By clicking the translator icon in your URL...' but I haven't any idea what they mean by 'translator icon' I'm assuming this is the icon that appears in the address bar when you are using a catalog of database and it has the proper metadata so that you can just click it and populate the fields in Zotero.
  • edited June 19, 2013
    'm assuming this is the icon that appears in the address bar when you are using a catalog of database
    Yes.
    But:
    « If all the multilingual data you need for a citation record is not automatically imported, you can always add more languages and data manually into Zotero. Fields such as Title, Author, and Publication are not only editable but can have multiple language metadata attached to a single element. »

    That should work with CiNii though (according to the blog post - which is, btw, outdated), can you provide an example URL?
  • Thanks for helping out Gracile,

    I actually used the exact same citation as in the blogpost, as seen in this screen capture:
    http://zotero.org/static/images/blog/multilingual_01.jpg

    Here is a link to that CiNii citation:
    http://ci.nii.ac.jp/naid/110007986263/ja/

    I've switched CiNii between its Japanese and English interfaces, and it will only import the respective language, one at a time. I wonder if I have done something wrong in my settings, which I've screen captured here:
    https://docs.google.com/file/d/0B93Jf_JfK6GNSWJ5SGpZTXZoWHM/edit?usp=sharing
  • You'll have to wait for fbennett...
  • edited June 20, 2013
    While we wait for Frank Bennett--

    I, too, have great interest in getting better metadata from CiNii.

    The main problem is with the CiNii metadata and not with Zotero. Although it is somewhat better than it was, the CiNii metadata is inconsistent across records, inconsistent across language views, incomplete, and often incorrect.

    Please do a search of the forum using "CiNii". There are several discussions about the metadata problems. Please also write to the people listed on the site contacts page or that are listed in Zotero forum posts.
  • The English/Japanese interface switch with "/en/" and "/ja/" stubs on the item URL seems to be something new. I've fixed the translator to strip that out of the URL used to fetch the RDF. If you update translators, it should work again. Give it a try, and post back here if there is still a problem.

    I wasn't able to reproduce a fetch of one language only: I just got an error when trying to download the item (because the URL was bad and yielded an error page).
  • Hi Frank, thanks for checking it out. And to everyone for their advice.

    I hit preferences and updated the translators as suggested. MLZ still only pulled just the English or just the Japanese metadata depending on which version of CiNii I was on. I rebooted my computer to see that would fix it, then I tried using MLZ on another computer just in case there was some kind of quirk.

    MLZ is still great, and it saves me a ton of time. As DWL suggested it may really be an issue with CiNii, not MLZ. Although the same issue comes up when using the CU Boulder Library catalog; I can pull the English metadata out, but the parallel MARC fields in Japanese don't make it in. Still, it's very easy to copy and paste the extra data.
  • (translators need to be specifically re-written to import multilingual data to MLZ because they obviously don't for Zotero. That's not been done for MARC)
  • As adamsmith says, multilingual import doesn't yet work with MARC records. It will be a big win when it's implemented, but the work will have to be done by someone who knows the MARC standard well.

    I think you may be seeing only one language because neither "English" nor "Japanese" have been added in the Preferences -> Languages menu. Try setting those up, and then revisit one of the entries you grabbed from CiNii.

    (If the multilingual fields wake up only after configuring languages, that would be a bug in MLZ, and I'll work on getting a fix out pretty soon.)

This is an old discussion that has not been active in a long time. Before commenting here, you should strongly consider starting a new discussion instead. If you think the content of this discussion is still relevant, you can link to it from your new discussion.

Sign In or Register to comment.