W3C home > Mailing lists > Public > public-swd-wg@w3.org > February 2007

Re: [SKOS] Proposed Resolution for ISSUE 26: RelationshipBetweenLabels

From: Jon Phipps <jphipps@madcreek.com>
Date: Wed, 28 Feb 2007 09:33:05 -0500
Message-ID: <34b5049c0702280633w7e175770jdda572c948d75ada@mail.gmail.com>
To: "Antoine Isaac" <aisaac@few.vu.nl>
Cc: "SWD WG" <public-swd-wg@w3.org>

Hi Antoine,

On 2/28/07, Antoine Isaac <aisaac@few.vu.nl> wrote:
> >
> > Given the use cases and motivation cited in [2], most of the
> > relationships defined in the motivation seem to represent
> > relationships between concepts rather than between labels, so in that
> > case I'm inclined to agree.
>
> I'm quite puzzled here. For me things like synonymy and abbreviation are
> clearly at the lexical or terminological level, not at the conceptual
> one. If I have "car" and "automobile" I would like have just one concept
> for them, and not two that I would declare equivalent afterwards....

This is obviously a case where a quick glance (by me) is no substitute
for actual reading, and results in casual rather than thoughtful
remarks (also by me). I believe I must have seen 'antonym' and simply
moved on. You are quite correct.

>
> >
> > One other possible solution might be to recommend (not require) that
> > in the sole case of a need to maintain multilingual versions of a
> > concept and all of its literal properties, that concept schemes be
> > considered single language and provide for this purpose an
> > isTranslationOf relationship (or some other form of typed equivalence)
> > between concepts?
>
> This could do the trick, but I'm convinced we have enough multi lingual
> cases (and too small a number of explicit label translation cases) to
> say this would be counter-productive. All things considered, I actually
> would prefer to have to encode the translation links using whatever
> complex annotation alternative than to have to split concept schemes
> into more-or-less duplicate versions (and God knows I am not personally fond of
> this annotation solution ;-)
>

Expect some cluelessness from me here, linguistics and lexicography
are hardly my areas of expertise, especially as it relates to
thesauri, but forging on anyway...

I guess I'm thinking that relationships between lexical labels (terms)
in different languages may often be more complex than the notion that
they represent the same concept. It would seem to me that very
frequently concepts represented by labels expressed in language A may
have an "effective translation" relationship to labels representing a
similar, but semantically very different concept in language B. The
utility of this explicit relationship between lexical labels (as
representatives of their related concepts) would be to allow, for
instance, a search retrieval service to translate between languages by
inference without requiring that the labels in each language be
aligned to the same concept in the same scheme.

This is perhaps a bit tangential to the current discussion, but seems
related, since this would also appear to require that labels be
expressed as resources.

I think perhaps the issue of multilingual conceptual representation
might be best considered  separately from the issue of lexical
relationships between labels?

--Jon

>
> >
> > I'm also thinking here of my daughter, the soon-to-graduate Spanish
> > major, who tells me that there are concepts in Spanish that can't be
> > expressed in English but for which there are English equivalents that
> > are considered to be effective translations.
> >
> > Anyway, just a thought.
> >
> > --Jon
> >
> > [2]
> > http://www.w3.org/2006/07/SWD/wiki/SkosDesign/RelationshipsBetweenLabels
> >
> >
> > --Jon
> >
> >> > On 2/27/07, Guus Schreiber <schreiber@cs.vu.nl> wrote:
> >> >>
> >> >> ISSUE-26 [1]
> >> >> RelationshipsBetweenLabels
> >> >>
> >> >> Considering that:
> >> >> - representing lexical labels as classes would
> >> >> lead to an undesirable complication of SKOS in
> >> >> straightforward use cases for the application of SKOS,
> >> >> - representing relationships between labels is
> >> >> required in some use cases, and therefore an
> >> >> escape mechanism should preferably be available
> >> >> for such thesauri,
> >> >>
> >> >> I propose the WG opts for an amended version of
> >> >> the second solution proposed in [2]:
> >> >>
> >> >> RESOLUTION
> >> >>
> >> >> The WG resolves to add the following classes and
> >> >> properties to the SKOS specification [3]:
> >> >>
> >> >> - the class skos:LabelRelation
> >> >> - the properties skos:labelRelationSubject and
> >> >> skos:labelRelationObject with domain LabelRelation
> >> >> and range rdfs:literal
> >> >>
> >> >> In addition, the SKOS Guide should describe
> >> >> guidelines for SKOS users to define their label
> >> >> relations as specializations of LabelRelation and
> >> >> gives examples of its intended usage. The SKOS
> >> >> specification refrains for now to predefine
> >> >> specializations of LabelRelation.
> >> >>
> >> >> Contrary to the proposal in [2] the class
> >> >> LabelRelation is not defined as a subclass of
> >> >> skos:Annotation (which is in any case not yet part
> >> >> of the spec), as it is not an "annotation", but a
> >> >> lexical relationship.
> >> >>
> >> >>
> >> >> [1] http://www.w3.org/2006/07/SWD/track/issues/26
> >> >> [2]
> >> >>
> >> http://www.w3.org/2006/07/SWD/wiki/SkosDesign/RelationshipsBetweenLabels
> >> >> [3] http://www.w3.org/TR/swbp-skos-core-spec/
> >> >>
> >> >> --
> >> >> Vrije Universiteit Amsterdam, Computer Science
> >> >> De Boelelaan 1081a, 1081 HV Amsterdam, The Netherlands
> >> >> T: +31 20 598 7739/7718; F: +31 84 712 1446
> >> >> Home page: http://www.cs.vu.nl/~guus/
> >> >>
> >> >>
> >> >
> >>
> >> --
> >> Vrije Universiteit Amsterdam, Computer Science
> >> De Boelelaan 1081a, 1081 HV Amsterdam, The Netherlands
> >> T: +31 20 598 7739/7718; F: +31 84 712 1446
> >> Home page: http://www.cs.vu.nl/~guus/
> >>
> >
> >
>
>
>
>
Received on Wednesday, 28 February 2007 14:33:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:17:28 GMT