Re: S36 implementation details

On 15 Jan 2016, at 15:42, Antoine Isaac <aisaac@few.vu.nl> wrote:

> That's exact.

If you look at the source [1] you will see that S36, along with a number of other conditions are simply represented using comments. Note that there is also a “prune” of the SKOS vocab [2] which has further axioms removed in order to provide an ontology that fits within the OWL1 DL sublanguage and can then be used by applications that wanted to live within OWL DL. As Antoine points out, this was all in the context of OWL 1.

Cheers,

	Sean

[1] http://www.w3.org/TR/skos-reference/skos.rdf
[2] http://www.w3.org/TR/skos-reference/skos-owl1-dl.rdf


> 
> Antoine
> 
> On 1/15/16 4:39 PM, Frederik Baumgardt wrote:
>> Thank you, Antoine!
>> 
>> Did I understand correctly, that S36 is a convention and not implemented in the source file?
>> 
>> Best,
>> Frederik
>> 
>> PS: Sorry for replying directly, didn’t realize it won’t go to the mailing list.
>> 
>>> Am 15.01.2016 um 10:20 schrieb Antoine Isaac <aisaac@few.vu.nl>:
>>> 
>>> Dear Frederik,
>>> 
>>> It's mostly a timing issue.
>>> At the time SKOS was made OWL2 property chains were still a fresh notion. We were aware of their development but it seemed a bit premature to base the SKOS standard on them. Hence we didn't consider using them.
>>> As a matter of fact the official OWL specification of SKOS is still an OWL1 one.
>>> 
>>> Best,
>>> 
>>> Antoine
>>> 
>>> On 1/15/16 3:53 PM, Frederik Baumgardt wrote:
>>>> Hi all,
>>>> 
>>>> I’m wondering about this detail in SKOS whereby elements in a memberList of an OrderedCollection are also members of the same OC. It appears in this comment <https://github.com/AKSW/Xturtle/blob/develop/vocabularies/xturtle.skos/skos.ttl#L180> and in S36 <https://www.w3.org/TR/skos-reference/#collections> of the reference document, but I’m not able to find its implementation.
>>>> 
>>>> On a sidenote, I am inclined to think of it as an owl:propertyChainAxiom, but that would require a property linking a list with its members without the rdf:List/rdf:rest{}/rdf:first indirection. Which has me wondering if a transitive property in place of rdf:rest positioned inside an owl:propertyChainAxiom — like so: ex:member owl:propertyChainAxiom (ex:rest ref:first) would achieve the desired effect.
>>>> 
>>>> Sorry if I’m being ignorant of important skos/owl/rdf spec-details here. New to the field and still learning about the finer distinctions.
>>>> 
>>>> Best,
>>>> Frederik Baumgardt
>>>> 
>>>> --
>>>> Sr. Developer @ Perseids Project
>>>> Perseus Digital Library
>>>> Tufts University
>>> 
>> 
>> --
>> Sr. Developer @ Perseids Project
>> Perseus Digital Library
>> Tufts University
>> 
> 

--
Sean Bechhofer
Senior Lecturer
School of Computer Science
University of Manchester
sean.bechhofer@manchester.ac.uk
http://staff.cs.manchester.ac.uk/~seanb
http://orcid.org/0000-0002-5872-3573

Received on Friday, 15 January 2016 16:37:49 UTC