Multilingual controlled vocabulary

Help
Paul Boon
2011-09-12
2013-04-30
  • Paul Boon

    Paul Boon - 2011-09-12

    I am investigating the possibilities of tematres for use in our project (lets call it X).
    Now I did get tematres(1.4.Beta) to work and its fine, but I am strugling with building a multi-lingual controlled vocabulary.

    I have set op the multi-lingual approach "reference model with local vocabulary" as described by the manual [http://r020.com.ar/tematres/manual/trminos_de_vocabularios_de_referencia_locales.html]

    I started my vocabulary with the name X and added terms in English and also added relations which works fine.
    I created an X_nl and an X_de target vocabulary for Dutch and German using the "Administration - Vocabulary configuration" page.
    Then I added equivalent terms from my source (English) terms to the targets in Dutch and German.
    Those are shown on the terms page as "EQ", but when I export the vocabulary the Dutch and German terms are not in there no matter what the format is.
    Also I don't know how to "view" the vocabulary in Dutch or German showing the Dutch and German terms and all the relations that I see in the source vocabulary(NT/BT).
    Maybe I am doing something wrong or my expectations are incorrect?

    The alternative would be to set it up using the "federated" approach.
    But that doesn't sound natural for a controlled vocabulary.
    Also I am not to happy about managing several independend instances of tematres, one for each language with its own database, user accounts etc.

    Any suggestions or tips for setting up a multilingual controlled vocabulary with tematres are welcome.

     
  • diego

    diego - 2011-09-16

    Hi Paul I will try to response

    1) not appears Terms from target vocabulary  in export to alphabetic version on txt

    Upsss!!!! you have found a bug!! :/ A few minutes ago I published the 1.41 version of tematres with this bug solved. thank you!!  Please… if you can… check if now it right.

    2) Multiligual vocabulary using Federated model
    In our experience the federated way is better because for several reasons:

    a) In search operation only appear terms in the native language, each language have an isolated ambient for management and for search operations
    b) In your case, may be the english is the pivot language, but when your vocabulary growth and each language begin to gain scope and size … may be desirable  more autonomy (eg: in UF terms in each language)
    c) Some colaborative project have separated terminological teams for each language. If you have the same team… yo can use the same user and pass to all the vocabularies (is no the better solution… i know :((( )
    d) The federated model use terminological web services to operation between vocabularies… you can use the same way to create mapping relations between different vocabularies in the same domain (eg between GEMET and LTER).

    well … :)

    please let me know if we can help you and we are very grateful for your help to found bugs :)!

    best regards

    diego ferreyra
    tematres@r020.com.ar

     
  • Paul Boon

    Paul Boon - 2011-09-22

    HI Diego,

    The export of "Alphabetic list (txt)" is now indeed having the EQ terms in it, but the SKOS export does not have it.
    I would expect skos:exactMatch although in SKOS different languages for the same  concept are modeled using a prefLabel.

    Anyway, you have convinced me into using the Federated approach and I will have a look at that first.

     

Log in to post a comment.

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:





No, thanks