See also: IRC log
PROPOSED: Accept minutes of the November 27th telecon (http://www.w3.org/2007/11/27-swd-minutes.html)
RESOLVED: Minutes accepted
Next telecon: 11 December 1600 UTC
<scribe>ACTION: Quentin to review Editor's draft of SKOS Reference [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action02] [CONTINUES]
<scribe>ACTION: Vit to review Editor's draft of SKOS Reference [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action03] [CONTINUES]
<scribe>ACTION: Alistair and Guus write draft section in primer on relationship between SKOS concepts and OWL classes for OWL DL users [recorded in http://www.w3.org/2007/11/06-swd-minutes.html#action05] [CONTINUES]
Issue 39 Discussion
Antoine: The issue of mapping links, using existing relationships in SKOS to create mapping links was the original approach. This new proposal is much more conservative. It takes text from the existing mapping vocabulary, and introduces two new concepts: related match and overlapping match, and deprecates major and minor match elements from current mapping vocabulary. The proposal is to delegate status of these elements to "concept coordination" - SKOS mapping coordination constructs: and, or, not. The proposal attempts to establish some formal semantics for these constructs, including domain, range. The semantics are to be discussed, e.g., exact match is transitive. Section 3 covers other issues, includes example of animal ontologies, attempts to match them. It concludes with relation rules, intended semantics, issues leading to decisions in the proposal, for example replacing major, minor match with overlapping match. It also raises the issue of building mapping links between collections that are not SKOS vocabularies explicitly.
Guus: So, I would like to have discussion on a few points in the proposal. Why do we need exact match, and why can't we use owl:sameAs?
Antoine: The problem is when you use owl:sameAs, you essentially merge the two schemes, which the original author may not want to do.
Guus: The point is to have this rationale in the proposal for documentation purposes. You could also point to why you can't use OWL different class. Both points can be made here.
Guus: Is it also possible to summarize the main point of the follow-up email discussion...
Antoine: Issues raised - considered that there was a clear distinction between SKOS mapping relations, between different concept schemes can still use standard concept relations between concept schemes, but this would not be for mapping.
Guus: Is it inconsistent to use one of the mapping relationships in the same scheme?
Guus: This is a borderline case, we should cover how someone might want to use the mapping concepts to do this -- the main point would be to make a statement about this. This should hold for all of the mapping relations.
Guus: What is the typical example of ... I don't really understand the overlapping match.
Antoine: tThe overlapping match more or less replaces the major and minor match. This describes relations that are in a mapping - relationships that are common to two mappings. You have relations with significant semantic load; if concepts are related in this way, are there are other concepts that a user should consider? It isn't 100% clear where you might use these - Emma McCulloch will provide additional examples
Guus: Do you think there are convincing use cases?
Alistair: I haven't seen use cases.
Guus: This seems rather arbitrary
Antoine: The point came from Javier ... the general point was that the types of inexact or overlapping relations you might find are different in nature from associative relationships between thesauri, so there might be for example overlaps between two thesauri that cover a common time period.
Guus: There is no precedent for this in the thesaurus community, and this seems to be limited to time and place.
Guus: This seems a rather specific solution, not sufficient evidence to warrant includes...perhaps we could keep it as a separate option
<aliman> +1 on guus position regarding "overlapping" property
Guus: If we go for a separate proposal for mapping terminology
Antoine: We could keep this separate - I can be ok with that
Guus: Do you think, given the comments, this is sufficient to move this forward for next time?
Antoine: yes, but feedback under the specific semantics/axioms would be useful. If people think there are useful axioms to have there, that would be good feedback.
Guus: So we can leave this action, and move it forward for next week.
<scribe> ACTION: Antoine to come up with a revised proposal based on the comments received [recorded in http://www.w3.org/2007/12/04-swd-minutes.html#action01]
SKOS Extension Module - no action
<scribe> ACTION: Alistair and Guus to prepare material for next week on Concept Schemes vs OWL Ontologies [recorded in http://www.w3.org/2007/10/30-swd-minutes.html#action04] [CONTINUES]
<scribe> ACTION: Guus to write up the issue [of Label Resource] and add to the issue list. [recorded in http://www.w3.org/2007/10/09-swd-minutes.html#action01] [CONTINUES]
<scribe> ACTION: Ralph to reconstruct proposal for semantics of isDefinedBy [recorded in http://www.w3.org/2007/10/30-swd-minutes.html#action02] [CONTINUES]
Guus: Statement on issue 44 from Alistair ...
Alistair: Antoine raises concern about what might happen if we say nothing about the transitivity of SKOS broader. If we allow people to make either choice, people will make conflicting decisions, which is a valid concern. He proposes the notion of a locally transitive relationship, also see Alan Rector's comment in the SWBP note in simple part whole. This is a serious concern that needs to be addressed.
Guus: Ok. With respect to producing a reference, when can we expect something?
<TomB> [http://www.w3.org/2001/sw/BestPractices/OEP/SimplePartWhole/ Simple part-whole relations in OWL Ontologies]
Alistair: It depends on how far we want to get before putting out a first draft. If we want something to be reviewed before Christmas, publishing something as an editor's draft before Christmas...
Guus: Assuming that Antoine moves his proposal forward, it doesn't seem far fetched.
Alistair: We could, and then throw stones at it ... it certainly needs to go in there...
Guus: That would mean that we produce a new draft before Christmas, and use the Christmas period for review.
Alistair: Work on what we've got more or less right now.
Guus: yes, and go forward with that.
Alistair: We talked about a rules of thumb appendix, to show examples of how to do something sensible, which requires some of Sean's time, who is traveling.
Ben: The central issue is related on exactly how we do chaining and how we deal with instance of. We have agreement in the group that this is the last major issue to deal with. Assuming we don't hit any snags, and fix the use cases for the new proposal on chaining. I've update the schedule, we're pushing hard to get the syntax with this new proposal in over the next 10 days. It would be good to get people on the call to review the new syntax before Christmas.
Guus: Didn't we have reviewers for the first working draft?
Ben: yes, we're just tweaking it for the latest changes, hoping this will go to last call.
Guus: We should use the same reviewers.
Antoine did the primer...
Diego: I did a review, the other one was Ed.
Ben: Are you willing to review again, and if so, what would the schedule be?
Ed: yes I'd be willing to do this again.
Ben: In two weeks -- would you still have time to review this before the holidays?
Ed: yes, if it hasn't changed substantially
Ben: There is one notable change which we will explain
Ed: yes this would be fine
Diego: yes, ok
Ben: Then I will hand it over to you the 18th
Guus: Will this be a working draft?
Ben: This would be a working draft leading to last call ... how does this work? you publish a working draft and then say it is going to last call?
Guus: We agreed that the decision for last call would be taken by both groups. We could make a decision on the first call in January. We should check with Stephen if they can take a decision by the first call in January. You need to send a proposal to the WG to promote the doc to last call (to both groups, pointing to responses to comments ...document precisely what has happened with the comments), then you have to set a last call period. Normally the comment period is between 4 and 6 weeks.
Ben: The commenting period starts after the WG is published, correct?
Guus: It would be good to prepare a list that you want to send the last call to, also potentially track comments. Issue tracker would be fine. Ralph -- what would your proposal be for tracking comments?
Ralph: absolutely, the issue tracker
Ben: ok, you'll most likely hear from me as I work through the process
Guus: Are there any potential sources of delay?
Ben: The one potential issue is the one we're working on ... if we find things in the test process, then there are possibilities there. We have 4-5 implementations that will signal an issue quickly at this point.
Guus: then you should prepare an implementation report during the last call period
Ben: we are definitely prepared to do so
Guus: I think I saw progress on the recipies as well.
Jon: We've put up a new editor's draft - I think it covers all of the issues except that it's broken under certain circumstances
...which has to do with Apache redirect (1st action under recipes, which is continued)
<scribe> ACTION: Dan to ask apache about conditional redirects [recorded in [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action18] [CONTINUES]
Ralph: I think my action still needs to be addressed ...
Jon: I think we've dealt with that but of course we have a new one for you, which is action 24, the next one
<Ralph> Open Issues on Recipes
<scribe> ACTION: Diego to recast Recipe 6 [recorded in [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action19] [COMPLETED]
<scribe> ACTION: Guus/Tom to solicit reviewers for the Recipes document. [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action20] [CONTINUES]
<scribe> ACTION: Jon to add words that acknowledge the existence of RDFa as [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action21] [COMPLETED]
<scribe> ACTION: Jon to make changes as proposed [with regard to Issue-23] [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action22] [COMPLETED]
<scribe> ACTION: Ralph propose resolution to ISSUE-16 "Default behavior" [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action23] [DROPPED]
<scribe> ACTION: Ralph to come up with a URI for wiki page [for Recipes [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action24] [CONTINUES]
<scribe> ACTION: Ralph/Diego to work on Wordnet implementation [of Recipes] [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action25] [CONTINUES]
<scribe> ACTION: [COMPLETED] TF leaders to prepare a version of Recipes for review in [recorded in http://www.w3.org/2007/11/20-swd-minutes.html#action26]
<scribe> ACTION: Diego to update deliverables page w.r.t recipes document [recorded in http://www.w3.org/2007/11/27-swd-minutes.html#action20]
Guus: We still need to solicit reviews for this -- when would reviewers be needed?
Jon: Any time, I guess -- we're just missing that one piece, which I guess is fairly critical.
Guus: We should solicit reviews at this point, though.
Ralph: I'll certainly review it.
Guus: We are at the top of the hour, so let's adjourn.