JiscMail Logo
Email discussion lists for the UK Education and Research communities

Help for DC-VOCABULARY Archives


DC-VOCABULARY Archives

DC-VOCABULARY Archives


DC-VOCABULARY@JISCMAIL.AC.UK


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Proportional Font

LISTSERV Archives

LISTSERV Archives

DC-VOCABULARY Home

DC-VOCABULARY Home

DC-VOCABULARY  April 2012

DC-VOCABULARY April 2012

Options

Subscribe or Unsubscribe

Subscribe or Unsubscribe

Log In

Log In

Get Password

Get Password

Subject:

Re: Moving forward on discovery and selection

From:

Antoine Isaac <[log in to unmask]>

Reply-To:

DCMI Vocabulary Management Community <[log in to unmask]>

Date:

Mon, 2 Apr 2012 16:21:49 +0200

Content-Type:

text/plain

Parts/Attachments:

Parts/Attachments

text/plain (103 lines)

Hi,


> @Antoine, are you on this list? Could you share some of your thougts on <sameas> links between (and within) LOD-vocabularies?


I'm not sure I have much to say here. I have never really defended owl:sameAs, which in practice is used in a quite harmful way (luckily no-one uses it for what it is supposed to be doing). Bernard is quite right.

I'd recommend though that this group focuses on well-defined cases to avoid owl:sameAs-like situations. 
I expect these cases as far as matching is involved, would rather address metadata element schemes like DC or FOAF, not matching of individuals in (bibliographic) databases, nor controlled vocabularies. This would allow to use patterns with clearer / less dangerous semantics, like the mapping axioms Bernard is hinting to, or owl:equivalentClass or rdfs:subPropertyOf. 

Best,

Antoine



> Hi Bernard,
> 
> Thanks for your response on my opinions and questions. I'm impressed to hear about the complexities concerning interconnecting vocabularies.
> 
> At Bibliotheek.nl we are currently working towards creating a connection between library-type authority files and the Dutch chapter of DBpedia. By doing this, we hope to open up library collections to the (semantic) web, while still being able to maintain the authority files.
> 
> @Antoine, are you on this list? Could you share some of your thougts on <sameas> links between (and within) LOD-vocabularies?
> 
> Kind regards,
> Johan
> 
> ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
--
> *Van:* DCMI Vocabulary Management Community [[log in to unmask]] namens Bernard Vatant [[log in to unmask]]
> *Verzonden:* maandag 2 april 2012 12:21
> *Aan:* [log in to unmask]
> *Onderwerp:* Re: Moving forward on discovery and selection
> 
> Hi Johan
> 
> With due respect to Antoine, the interlinking of vocabularies cannot be solved only by <sameas> assertions. Indeed I have chewed over this issue, sometimes /ad nauseum/ on the wheel and the hub, since 2004. My current view on this is that there is a whole spectrum of ways to say that two terms/names/URIs have more or less the same, or similar referents. Among those, owl:sameAs is the most radical expression of strict identity, and it has been so misused and abused that in the LOD cloud it has just lost its original meaning as defined by OWL specification. Links available at sameas.org <http://sameas.org> are, although precious, in some sense add to confusion, because many of them (but you can't figure which)
> SKOS provides more flexibility with various "matching" properties, but they are limited to skos:Concept. I have suggested also contextual rules, such as substitution rules (in such context, you can replace this URI by that one without risk).
> 
> In the latest version of geonames ontologies, I have for example shown that one can map concepts represented by owl:Class of rdfs:Class in some vocabularies to skos:Concept in another one, using OWL DL Restrictions
> See http://www.geonames.org/ontology/mappings_v3.01.rdf
> 
> Mike Bergman, ediror of UMBEL (among other hats) has also posted at AI3 a lot of useful posts on this issues, such as :
> http://www.mkbergman.com/935/the-nature-of-connectedness-on-the-web/
> http://www.mkbergman.com/938/what-is-a-reference-concept/
> http://www.mkbergman.com/941/making-connections-real/
> 
> That said, I agree (see my previous answer to Jonathan) this is an issue somehow orthogonal to the multilingual issues. Or, put it otherwise, the multilingual issue is everywhere, not only at the borders between so-called "natural languages", but inside them. Every formal vocabulary is a dialect emerging in a specific community, often addressing specific functional requirements. The issue of mapping different vocabularies in the same natural language is therefore an issue similar to a translation between dialects. And what is the difference between a dialect and a language is a question I prefer to keep agnostic about :)
> 
> Bernard
> 
> Le 26 mars 2012 08:59, Johan Stapel <[log in to unmask] <mailto:[log in to unmask]>> a ¨¦crit :
> 
>     Hi all,
> 
>     Yes, it resonates. I agree with Diane: excellent posts by Bernard, especially because he stresses the fact that it is not just about pushing our vocabularies out, but also about the maintenance and the interlinking that is needed on vocabularies once they are published as linked data.
> 
>     Last year Roy Tennant blogged [1] that we should establish a sustainable LOD infrastructure and aknowledge that "it's not the metadata, but the motion". In a collaborative past [2] Antoine Isaac has convinced me that we should be putting energy in creating <sameas> links between vocabularies [3], a topic Bernard has been blogging about since 2004 (!) [4].
> 
>     So, Jonathan, in my opinion the need to interlink LOD'd vocabularies does not only apply to multilingual vocabularies, but also to vocabulairies that contain similar topics, which is what I have tried to advocate in my pecha kucha talk at EMEARC in Birmingham [5].
> 
>     Kind regards,
>     Johan
> 
>     [1] http://blog.libraryjournal.com/tennantdigitallibraries/2011/07/18/its-not-the-metadata-its-the-motion/
>     [2] http://www.slideshare.net/antoineisaac/w3c-library-linked-data-incubator-group
>     [3] http://sameas.org/
>     [4] http://blog.hubjects.com/search/label/same
>     [5] http://www.slideshare.net/mobile/jstapel/pecha-kucha-oclc-emearc-20120228
> 
>     ________________________________________
>     Van: DCMI Vocabulary Management Community [[log in to unmask] <mailto:[log in to unmask]>] namens Jonathan Pool [[log in to unmask] <mailto:[log in to unmask]>]
>     Verzonden: maandag 26 maart 2012 2:35
>     Aan: [log in to unmask] <mailto:[log in to unmask]>
>     Onderwerp: Re: Moving forward on discovery and selection
> 
>      > Do the issues Bernard raises resonate with any of you out there?
> 
>     Thanks for referencing these two postings. The first of them seems at least in spirit relevant to the Language Commons project (http://languagecommons.org). The second posting describes, among other things, the unpredictability of the results of attempted vocabulary proposals, reminiscent of a common observation in the sociolinguistic research on "language planning" (i.e. the planning of lexical and grammatical change often fails). This suggests estimating the feasibility and benefits of independent vocabulary development, as opposed to vocabulary innovation incident to substantive creative activity.
> 
>     But the main question that the postings raise for me is what lessons have been learned about multilingual, as opposed to monolingual, vocabulary development, since the postings don't seem to deal explicitly with the multilingual case. Arguably, creators normally offer only monolingual lexical innovations as a byproduct, so multilingual vocabulary development depends on independent work by vocubulary specialists. But recently some organizations that sell worldwide have created massive sets of new terms and published multilingual versions of them.
> 
>     Mala Tabory once argued that international treaties are drafted more clearly when they have equally authentic versions in several languages, because multilingual parallel drafting forces drafters to resolve ambiguities that survive monolingual drafting. This suggests that multilingual vocabularies may make the conceptual schemes that they express not only accessible to a larger audience, but also more robust. Moreover, it is possible for a single individual to create a monolingual vocabulary and promote its use, but a massively multilingual vocabulary can't come into existence without the participation of many "gardeners", so Bernard's call for a more participatory regime of vocabulary development may apply more naturally to multilingual than to monolingual vocabulary work.
>     ¡¥
> 
> 
> 
> 
> -- 
> *Bernard Vatant
> *
> Vocabularies & Data Engineering
> Tel : + 33 (0)9 71 48 84 59
> Skype : bernard.vatant
> Linked Open Vocabularies <http://labs.mondeca.com/dataset/lov>
> 
> --------------------------------------------------------
> *Mondeca** ** *
> 3 cit¨¦ Nollez 75018 Paris, France
> www.mondeca.com <http://www.mondeca.com/>
> Follow us on Twitter : @mondecanews <http://twitter.com/#%21/mondecanews>
> 

Top of Message | Previous Page | Permalink

JISCMail Tools


RSS Feeds and Sharing


Advanced Options


Archives

April 2014
March 2014
February 2014
January 2014
December 2013
November 2013
October 2013
September 2013
August 2013
July 2013
June 2013
May 2013
April 2013
March 2013
February 2013
January 2013
December 2012
November 2012
October 2012
September 2012
August 2012
July 2012
June 2012
May 2012
April 2012
March 2012
February 2012
January 2012
December 2011
November 2011


WWW.JISCMAIL.AC.UK

Secured by F-Secure Anti-Virus CataList Email List Search Powered by the LISTSERV Email List Manager