Re: [SKOS] A revised proposal for ISSUE-39 ConceptualMappingLinks

Dear all,

> *[NEW]* *ACTION:* Antoine to send a msg proposing a resolution for 
> next week telecon on ISSUE-39 considering Alisatir's 3 subtopics 
> [recorded in]

I would propose for next telecon that the workgroup accepts [4] as a 
solution to ISSUE-39 ConceptualMappingLinks [3].
This proposal is introduced in the mail attached below (also accessible 
at [6])

With respect to the 3 sub-topics raised by Alistair [7], the proposals 
made in the wiki page [4] are:

> (ISSUE-39A) Should "grouping" constructs for mapping be included, and if so, what are their semantics?
-> This sub-issue is postponed until until a solution has been found to 
ISSUE-40 ConceptCoordination [8] and ISSUE-45 
NaryLinksBetweenDescriptorsAndNonDescriptors [9], which are about very 
similar problems.

> (ISSUE-39B) Is it necessary to have parallel vocabulary (skos:broader // skos:broadMatch etc.)? If not, how do you differentiate between intra-scheme vs. inter-scheme semantic links?
-> This sub-issue is dealt with: the SKOS mapping relations are 
introduced as parallel to the existing SKOS semantic relations 
(skos:broader, etc)

> (ISSUE-39C) What's the difference between "related" and "overlapping"? Is there enough precedent to justify a new property for "overlapping"?
-> This sub-issue is dealt with: only skos:relatedMatch is kept in the 
proposal, and a discussion item gives motivation for it.

Of course I invite the WG to consider this proposal considering any new 
argument that could have been raised on the SWD and SKOS lists between 
now and next week. The current proposal tries to take into account all 
the feedback we got on the subject over the past months, but the topic 
was very active over the past days (over 20 mails in two weeks!).





> Dear all,
> Following last week's teleconference [1], and, especially, the 
> insightful comments got from the SKOS mailing lists [2] (thanks again 
> to all of you who contributed to this important discussion!), I have 
> revised my proposal for ISSUE-39 conceptualMappingLinks [3]. The 
> result is accessible at [4]
> The main differences with the previous version [5] are:
> - skos:overlappingMatch is not kept.
> - there is a formal proposal for postponing resolution on: part of 
> RDFS and OWL semantics, semantic conditions, inconsistent examples, 
> entailment rules and syntactic constraints. I asked for feedback last 
> week, and so far there has been only one comment on the axioms of [5]. 
> I guess none of us has the time for this between now and the end of 
> the year :-(
> - there is a formal proposal to raise an issue on mapping between 
> conceptual entities (e.g. groupings) that are not of type skos:Concept
> - the discussion section has been revised: it now especially includes 
> a paragraph on owl:sameAs vs skos:exactMatch and a paragraph on 
> allowing mapping statements withing one concept scheme
> Best,
> Antoine
> [1]
> [2] 
> [3]
> [4] 
> [5] 

Received on Tuesday, 11 December 2007 19:11:34 UTC