Re: [W3C] Terminology module: next telco (July)

Dear Fahad, dear Thierry, dear all,

thanks for this discussion. In any case, the only way to keep closed
ISO/TC-internal discussions synchronized with open standardization
processes is to define a stable interface, preferrably not the ISO standard
itself (because it is proprietary, and drafts are not to be circulated),
but accompanying documentation (say, XML schemata as under
https://www.tbxinfo.net/tbx-downloads/iso-24633-resources/), along with an
open source reference converter to generate the agreed-upon RDF
representation (e.g., as Turtle or JSON-LD serialization) from TBX XML
data. This reference converter (and the data structures it produces) can
then be developed in a joint fashion. In this way, ISO rights are preserved
(neither converter nor schema need to quote from the standard), and ISO
definitions remain untouched (and can just be applied to the different RDF
serializations).

On the other hand, in my understanding, any attempt of RDF modelling from
scratch *within ISO/TC 37/SC 3* would effectively rule out the
participation of the wider community because of ISO regulations on
information sharing. What could work instead -- if this is the goal, indeed
-- would be to develop an RDF model from scratch outside ISO/TC 37/SC 3
(but, optionally, under inclusion of ISO/TC 37/SC 3 members, be it as
proposers, contributors or observers), and then to acknowledge this as
(part of) an ISO standard in a subsequent step. The difference is that the
initiative does not formally lie within ISO TCs. That was the way of
earlier ISO standards (e.g., https://www.w3.org/2011/07/wspas-pr.html or
https://www.oasis-open.org/2017/02/21/iso-iec-jtc-1-approves-oasis-odata-standard-for-open-data-exchange/),
and unless we hear otherwise from ISO, that seems to be the only way for an
open community to formally contribute to or collaborate with this TBX-RDF
initiative.

Another way out would be a formal agreement between ISO and, say, W3C on
joint standard development, but as far as I know, W3C legal form doesn't
meet ISO requirements for potential contract partners. (W3C has been
undergoing some restructuring lately, so I am not 100% sure on that point,
but that was the situation in the past. But even then, it is not clear what
this means for W3C Community Groups, because these are part of the W3C
ecosystem, but not the W3C as an institution.)

BTW: As there are several TBX-to-RDF converters around, already, I think
the community and downstream applications would greatly benefit from not
starting RDF modelling from scratch all over again, or, at least, not
behind closed doors, but to involve current providers and users of such
data and converters to maximize impact and maintain interoperability. If
this is really the goal, it can be much more easily achieved in an open
community where developers and users can just lurk and contribute as
needed, rather than in a closed setting in which individual experts need to
be identified first, and then be formally invited on demand.

Just my 2ct,
Christian

Am Mo., 19. Juni 2023 um 21:29 Uhr schrieb Fahad Khan <fahad.khan@ilc.cnr.it
>:

> Dear Thierry,
> Thanks for the clarification. You are right that "serialisation" is a
> sloppy use of language here which I'm effectively using as a shorthand for
> something more like a crosswalk (the serialisation would technically be to
> RDF/XML or turtle, etc,), if you like we can go with representation from
> now on. Regarding your second point, I feel that the Terminology module
> should offer the possibility of as faithful a conversion or representation
> of TBX data into RDF as possible, and even that this should be one of the
> main selling points of the module.
> Cheers
> Fahad
>
> Il giorno lun 19 giu 2023 alle ore 21:02 Thierry Declerck <
> declerck@dfki.de> ha scritto:
>
>> Hi Fahad, All,
>>
>> Thanks for the mail! From me just a "terminological" comment. The ISO
>> work is on "representing" the core of TBX in RDF, but not on"serializing"
>> TBX in RDF (as RDF is a modeling language). I think that we have to be
>> careful here (and the ISO committee, the link you are pointing to) is using
>> the right term (Representation).
>>
>> Besides this, I think that an Ontolex-driven representation of
>> terminology data should go beyond the "mere" reproduction of TBX and should
>> make full use of RDF-based modeling. Some of the points Patricia mentioned
>> in the document refered to by Jorge are related to this point.
>>
>> Well just some thoughts.
>>
>> Cheers
>>
>> Thierry
>> Am 19.06.2023 um 19:29 schrieb Fahad Khan:
>>
>> Dear All,
>> As I mentioned in the W3C Day preparation meeting today, I was in
>> Brussels last week for the annual ISO/TC 37 meeting where I became aware of
>> work being carried out within the aegis of the terminological committee
>> ISO/TC 37/SC 3 with the aim of defining a serialisation of TBX into RDF.
>> <https://www.iso.org/standard/84361.html> Subsequently I got into
>> contact with Hanne Smaadahl, the chairperson of the committee, who was
>> extremely enthusiastic about the prospect of the two groups coordinating
>> their respective efforts in a case where there was a very strong chance of
>> overlap ( as I recall one important use case for the new OntoLex module is
>> the conversion of TBX datasets into RDF). I think it would be worth
>> discussing this at the next terminology focused telco to try and understand
>> what such collaboration might look like, especially given the closed nature
>> of ISO standards, something which I guess is kind of anathema to
>> our community (though I also think it makes no sense to have two rival RDF
>> seralisations of TBX floating around).
>> Cheers
>> Fahad
>>
>> Il giorno lun 19 giu 2023 alle ore 19:10 Jorge Gracia del Río <
>> jogracia@unizar.es> ha scritto:
>>
>>> Dear W3C ontolex members,
>>>
>>> As you know, we recently started a discussion on a potential Terminology
>>> module for Ontolex. Our previous telco took place on 10th May, and it's
>>> time to schedule a new one. Please, those interested in participating fill
>>> in this doodle poll with some options in the first two weeks of July by the
>>> end of this week: https://doodle.com/meeting/participate/id/dwr0RRzb
>>>
>>> As a reminder, the main decision from the last telco was to start
>>> documenting the issues that arise when working with lemon-ontolex to model
>>> terminologies. Patricia enlisted some of them here:
>>> https://www.w3.org/community/ontolex/wiki/Terminology#Open_Issues, they
>>> will be filled with more details before the next meeting. Also, any other
>>> member with experience in modelling terminologies in lemon, feel free to
>>> document your own issues using the same template.
>>>
>>> Best regards,
>>>
>>> Jorge
>>>
>>> --
>>> Jorge Gracia, PhD
>>> Department of Computer Science and Systems Engineering /
>>> Aragon Institute of Engineering Research (I3A)
>>> University of Zaragoza
>>> http://jogracia.url.ph/web/
>>>
>> --
>> Thierry Declerck
>> Senior Consultant at DFKI GmbH, Multilinguality and Language Technology
>> Stuhlsatzenhausweg, 3
>> D-66123 Saarbruecken
>> Phone: +49 681 / 857 75-53 58
>> Fax: +49 681 / 857 75-53 38
>> email: declerck@dfki.de
>>
>> -------------------------------------------------------------
>> Deutsches Forschungszentrum für Künstliche Intelligenz GmbH
>> Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany
>>
>> Geschäftsführung:
>> Prof. Dr. Antonio Krüger (Vorsitzender)
>> Helmut Ditzer
>>
>> Vorsitzender des Aufsichtsrats:
>> Dr. Ferri Abolhassan
>> Amtsgericht Kaiserslautern, HRB 2313
>> -------------------------------------------------------------
>>
>>