Re: Translations of classical RDF vocabularies

Hi all,

I like this initiative. That's indeed something useful for the community
and where everyone could participate.

I would see a direct value to link this with LOV as it could gives you
notification whenever a vocabulary has changed which would required the
translation to be updated. On the other way, translations could be pushed
to LOV and added to the aggregator (SPARQL Endpoint + nquads dump) so
everyone could already uses the LOV search based on origianl language and
translations...

Is there any tooling to make community based translation? I guess even
developing it on LOV would not take long (2-3 months using spare time) ...
I know already some people who would be keen to help on that.

Would that be a good idea?

Pierre-Yves.


Pierre-Yves Vandenbussche.


On Mon, Jun 23, 2014 at 11:41 AM, Bernard Vatant <bernard.vatant@mondeca.com
> wrote:

> Hello Jean-Marc
>
> 2014-06-23 12:10 GMT+02:00 Jean-Marc Vanel <jeanmarc.vanel@gmail.com>:
>
> Bernard,
>> Fine to see my translation here !
>>
>> I would rather see
>>         lingvoj:originalResource    <http://xmlns.com/foaf/spec/> ;
>> than
>>         lingvoj:originalResource    <http://xmlns.com/foaf/spec/index.rdf>
>> ;
>>
>> not sure !
>>
>
> Well indeed what you've done is a translation of the formal RDF
> vocabulary. Given that there is a content negotiation on the namespace URI,
> the former declaration could be interpreted as the translation including
> also http://xmlns.com/foaf/spec/index.html which is not the case. I
> prefer the object of lingvoj:originalResource to be as close as possible to
> some frbr:Manifestation. When there is a conneg in place for some URI, I
> tend to think that the resource represented is some frbr:Expression, and
> each available file a frbr:Manifestation. Note that I differ in this from
> http://vocab.org/frbr/core.html#translation which enforces translation to
> link two frbr:Expression, and from BIBFRAME which defines translation as
> linking two Works : http://bibframe.org/vocab/translationOf.html. Given
> this lack of consensus I have not put any formal constraints on rdfs:range
> of lingvoj:originalResource and lingvoj:translatedResource which are left
> open. Translating vocabularies supporting conneg is a good test bed for
> those models anyway.
>
>
>> But I'm definitely sure that I prefer to see:
>>         lingvoj:translator <http://jmvanel.free.fr/jmv.rdf#me> .
>> ...
>>
>
> Sure enough :)
> I've fixed it, I was just not aware of your URI :)
>
>  Phil and Bernard, If you want to be informed of new translations, you
>> can follow ("watch") the github project :
>> https://github.com/jmvanel/rdf-i18n/
>>
>> I'll provide some  lingvoj triples for translations there, say on :
>> https://github.com/jmvanel/rdf-i18n/raw/master/translations_list.ttl
>>
>>
>> Bernard, will you update soon to "official" URI for languages like
>> http://id.loc.gov/vocabulary/iso639-1/fr
>> ?
>>
>
> http://www.lingvoj.org/translations/translations-2014.rdf already uses
> such URIs, do you speak of other places they should be updated?
>
>
> 2014-06-23 11:36 GMT+02:00 Bernard Vatant <bernard.vatant@mondeca.com>:
>>
>> Hi Phil
>>>
>>> I updated for the occasion the examples of representation of
>>> translations using the lingvoj ontology, since Jean-Marc was asking about
>>> how to declare translations in RDF.
>>> http://www.lingvoj.org/translations/translations-2014.rdf
>>>
>>>
>>> 2014-06-23 11:18 GMT+02:00 Phil Archer <phila@w3.org>:
>>>
>>>  Just a quick response to this.
>>>>
>>>> I'm delighted to see this happening, thanks Jean-Marc. I've been
>>>> pushing for more of this to happen for the vocabs that W3C hosts and am
>>>> pleased to be able to point to:
>>>>
>>>> DCAT [1] (English, Spanish, Arabic, Greek, Japanese, French),
>>>>
>>>> ORG [2] (English, French, Italian, Spanish)
>>>>
>>>> Labels and definitions in more languages are always welcome (pls ping
>>>> me directly). The multilingual LD folks are unanimous in their belief that
>>>> this is really helpful. See http://lider-project.eu/
>>>>
>>>> Thanks
>>>>
>>>> Phil.
>>>>
>>>>
>>>> [1] http://www.w3.org/ns/dcat.ttl
>>>> [2] http://www.w3.org/ns/org
>>>>
>>>> --
>>>>
>>>>
>>>> Phil Archer
>>>> W3C Data Activity Lead
>>>> http://www.w3.org/2013/data/
>>>>
>>>> http://philarcher.org
>>>> +44 (0)7887 767755
>>>> @philarcher1
>>>>
>>>>
>>>>
>>>> On 21/06/2014 17:26, Jean-Marc Vanel wrote:
>>>>
>>>>> I would be nice, for those vocabularies and ontologies that have only
>>>>> rdfs:label and rdfs:comment in english, to have translations in other
>>>>> languages.
>>>>> I started by FOAF in french:
>>>>>
>>>>> http://svn.code.sf.net/p/eulergui/code/trunk/eulergui/examples/foaf_fr.n3
>>>>>
>>>>> You may ask: "why bother do this?".
>>>>> I see at least 2 use cases:
>>>>> - for input form and display generated from the vocabulary, I18N
>>>>> labels and tooltips are necessary
>>>>> - for faceted search, I18N labels are useful
>>>>>
>>>>> I used Google translate as a starting point.
>>>>> When doing this, it's better to use the N-triples syntax; the G
>>>>> translator would spoil the abbreviated URI's in Turtle.
>>>>>
>>>>> If you want to contribute and translate other vocabularies ( not only
>>>>> in french !),
>>>>> we can set a github project;
>>>>> and the most useful ones would be: Good Relations and schema.org,
>>>>> SIOC, DOAP, Dublin Core.
>>>>>
>>>>>
>>>> Déductions SARL - Consulting, services, training,
>>>> Rule-based programming, Semantic Web
>>>> http://deductions-software.com/
>>>> +33 (0)6 89 16 29 52
>>>> Twitter: @jmvanel , @jmvanel_fr ; chat: irc://irc.freenode.net#eulergui
>>>>
>>>
>