See also: IRC log, previous 2008-04-01
next telecon: 15 April 2008 1500 UTC
RESOLUTION: 01-swd-minutes accepted as minutes of 1 April telecon
ACTION: Chairs to draft charter extension proposal for SKOS until July 1st [recorded in http://www.w3.org/2008/02/26-swd-minutes.html#action01] [CONTINUES]
Guus: Tom and I agreed to do this after today (European proposal deadline today)
PROPOSED: to have a May 6-7 SKOS face-to-face in Washington
-> Face-to-face meeting poll results
Clay: we have a conference room for both days
at LoC
... the only issue has been determining if we can get network access for
everyone
... typically it's tough to get internet connectivity for visitors
... I'm hoping to be able to get 15 'blessed' connections
Guus: connectivity is more essential for us, being a Web group
Clay: I think it's mostly a matter of preparing; there's pretty high visibility for this
Ralph: do you mean to devote the entire agenda to SKOS?
Guus: yes, that's what I was thinking
... we could devote some time to the other tasks
... my main goal was to get major decisions taken to have a clear road to a
Last Call draft for SKOS
... however, if Jon and Diego want an hour for recipes I'd be happy to
schedule this
Jon: I'm inclined to think we don't need to put
recipes on the f2f agenda
... we're pretty close to a final draft
<aliman> what about vocab management?
Ralph: perhaps an hour on Vocab Mgmnt would be useful?
Guus: we were hoping to finish the other tasks
by 5 May
... but we could schedule a total of 2 hours for other topics than SKOS
RESOLUTION: Face-to-Face in Washington on 6 & 7 May
Sean: will there be dial-in facilities?
Clay: good question
<edsu> Ralph++
Clay: we can find a speakerphone
Ralph: no problem using W3C's bridge
Guus: thanks to Library of Congress for hosting
ACTION: Guus and Tom draft an agenda for the May f2f [recorded in http://www.w3.org/2008/04/08-swd-minutes.html#action02]
Guus: can we start at 9am on Tuesday?
Ralph: that's late for me :)
... I second starting no earlier than 8 and no later than 9 :)
Guus: expect to start at 0900 on Tuesday and finish by 1600 on Wednesday
ACTION: Alistair and Guus to check the text in the primer on relationship between Concept Schemes and OWL Ontologies. [recorded in http://www.w3.org/2008/01/29-swd-minutes.html#action13] [CONTINUES]
ACTION: Guus to write primer text re: broaderGeneric and equivalence w/r/t subclass [recorded in http://www.w3.org/2008/03/18-swd-minutes.html#action08] [CONTINUES]
Guus: what's critical for the Primer now?
Alistair: it's looking pretty good
... just some things I highlighted in my review, which also relate to the
mapping vocabulary issues
... the story we tell about concept schemes, ontologies, and levels; whether
we link broadergeneric to OWL or leave them more open; this will be the
difficult thing
-> [SKOS] Standalone definitions in natural language; previous links in headers [Tom 2008-03-25]
Guus: my advice was to have pointers to natural language descriptions
ACTION: [DONE] Sean to propose a way to handle deprecated properties (updating RDF schema) [recorded in [63]http://www.w3.org/2008/02/12-swd-minutes.html#action06] [recorded in http://www.w3.org/2008/04/08-swd-minutes.html#action05]
-> Deprecated SKOS Vocabulary [Sean 2008-04-07]
Sean: my proposal is to document the deprecated
vocabulary in the spec but omit it from the RDF schema
... would be nice to start with a clean schema
Guus: document the deprecated vocabulary in an appendix
Sean: we should discuss the namespace
<aliman> +1 to sean's proposal re deprecated properties
Tom: is there a W3C policy on support of schemas for spec that have not yet reached REC
Ralph: we can -- and should prior to CR -- have a namespace document
Tom: the previous SKOS drafts are a legacy specification
Ralph: is the question about use of the legacy namespace?
Sean: there will be some legacy vocabulary
lying around; the things we're going to deprecate
... is the W3C happy about there being legacy vocabulary around that is not
described in a machine readable form?
Tom: can you follow your nose?
Ralph: I think it would be considered
unfriendly to remove things from the namespace
... there is data deployed and tools such as Tabulator that will want to
continue to browse that data
... I think we'd get severe push-back both architecturally and from users of
the old specs if we remove things from the namespace
Alistair: I assumed we'd stick with the same
namespace just because there's a lot of deployed data
... the deployed data would be consistent with the current spec
Ralph: the things we're deprecating are not used?
Tom: Alistair's point is that there's a lot of
data the uses the existing namespace
... if there is a new namespace then a relationship between the two
namespaces would need to be declared
... would the old namespace disappear?
Ralph: if you are asking if the old namespace would dissappear the answer is no
Ralph: the old namespace definitely won't disappear
Alistair: there's been some effort to put data in SKOS form
Ralph: a lot of the data wouldn't need to change
aliman: a small amt of data might use some deprecated vocabulary would need to change, but the majority wouldn't have to change at all
Guus: i'm slightly worried about making a new
vocabulary that makes data on the web invalid
... i'm more in favor of using a version type link
Ralph: how expensive is it for us to retain the deprecated vocabulary in our namespace document? I respect the truth/beauty argument, but I wouldn't want to omit this possibility
aliman: i'm a bit confused: how do we deal with the deprecated vocabulary, and one that is what is the namespace for skos -- i thought we were talking about what the skos namespace should be
seanb: i think we're having both convos at the same time
Guus: if we keep the same namespace it would be
better to have the old vocabulary marked as deprecated
... if we move to a new namespace we stil keep the old one, but people who
want to use the new vocabulary have to change the namespace vocabulary
Ralph: the strongest reason for moving to a new namespace is to remove the clutter
aliman: by creating a new namespace you are deprecating the old one
Guus: not entirely true, people could still use the old one
Ralph: could provide owl:sameAs relationships
Antoine: i have a question regarding a question from Simon Spero on the list, about the semantics between the new/old vocabularies
Ralph: would it raise actual problems, or theoretical problems?
aliman: implementations might have to change
Guus: we cannot make assumptions about what people have done, if they have used transitive they will now be in trouble ... from a maintenance point of view there is a strong case for creating a new URI for the 2nd version
seanb: do we have a feel for how people are using the vocabulary?
<Ralph> old SKOS WD
Ralph: Alistair, what's your belief about how stable people felt the old definitions were
aliman: i think people have anticipated that things would change, because they've been asking when the REC would come out
aliman: i would also defer to Antoine, Jon and Ed on what people are doing
seanb: are they concerned with the vocabulary or just the namespace?
aliman: tools have been built, some are in the commercial space, it wouldn't hurt too badly to ask people to change their URIs
Ralph: i think we should document this and flag
it, and make it part of the Last Call
... from the point of view of the w3c effort, this work has been in working
draft for a few years, that would be the basis of keeping the current
namespace
... we should acknowledge existing users however
Guus: i would feel uncomfortable to force people to make a change
Ralph: the question is what do we force them to change, to change transitive implementations or URIs
Guus: there's nothing deprecated about the old namespace
Ralph: that means tools would need to recognize
both namespaces
... it's feeling like the least pain is if we change the definitions of the
terms in the namespace
Guus: whatever we do we should make it a Last Call issue
marghe: the advantage to having a new namespace is that we will keep in mind versioning, it may be difficult to keep track of the different changes over time, we wouldn't know when new properties are added, etc
Guus: this is something we can't decide on now, i think seanb's messsage raises some clear points, and we should cover this in the f2f
Ralph: we should adapt the VocabManagement document to suit our needs :-)
ACTION: Ralph to check whether the common interpretation of rdfs isDefinedBy fits the reasoning that was made in http://lists.w3.org/Archives/Public/public-swd-wg/2007Oct/0141.html [recorded in http://www.w3.org/2007/12/18-swd-minutes.html#action10] [CONTINUES]
-> 2008-03-24: questions about n-ary relations solution
Guus: let's come back to this if there's time later in the telecon
ACTION: Ben to prepare draft implementation report for RDFa (with assistance from Michael) [recorded in http://www.w3.org/2008/01/22-swd-minutes.html#action14] [CONTINUES]
Ralph: the TF last Thursday did make some
changes to the SPEC, 1 was to change instanceof to typeof, and the other is
to make a small change to the processing rules, to remove a side effect from
a change that this working group recommended -- we had removed some bnodes
that otherwise wouldn't be there and now wish to add them back, both have
implementation impact, relatively small, the TF is recommending we extend
Last Call by 3 weeks, to allow for testing of processing rules
... i asked for the processing rules to undergo a bit more testing before we
go to Candidate Recommendation
diego: need some time to look at the changes
Ralph: ideal path would've been to publish an
updated working draft
... that would add more than a 3 week delay
... we believe we have given notice to the active implementor community, we
may need to do this in a more visible way
... we don't have final new language on the change to show to diego, i don't
feel comfortable until we see the new language making any decision
Guus: i assume you are distributing a comment
to that effect to the various channels
... i think if you do that all is fine
Ralph: ok
Guus: it might effect our charter extension proposal, if after May 1 we might do more work that we expect
Ralph: we're still expecting some post deadline
comments, but don't expect them to have substantive impact, one of the last
ones is from Yahoo, and they are in favor of both of the proposed minor
changes
... we're not proposing to delay last call
ACTION: Ralph propose resolution to ISSUE-16 "Default behavior" [recorded in http://www.w3.org/2007/01/22-swd-minutes.html#action14] [CONTINUES]
ACTION: Ralph/Diego to work on Wordnet implementation [of Recipes implementations] [recorded in http://www.w3.org/2008/01/22-swd-minutes.html#action20] [CONTINUES]
ACTION: [DONE] Recipes editors to clarify and write some sentences for the title that spells out points Ed made for .htaccess and Apache [recorded in http://www.w3.org/2008/03/18-swd-minutes.html#action13]
Elisa: Ralph and Diego's comments were
helpful
... should be able to deal with these in short order
Diego: document seems to be in good shape; only
minor changes
... the way Recipes is cited
... my comments should be easy to tackle
<Guus> Diego's comments
Elisa: I agree; I may want to iterate with Diego to get the language describing the relationship with Recipes
Ralph: i think the work that's there now is good enough that it would be shame to abandon, if we publish now with @@ i think we can fill them in later
Guus: we need to try to keep a strict timeframe
Ralph: if we wrestle with the skos namespace, this is the document that should/can decide how we resolve the skos namespace, my hope has been that this issue of how to evolve namespaces would rest here, i would like to publish this as a framework where we can put our knowledge in
Guus: tom you have some review as well?
<Ralph> [VM] review of March 16 editor's draft [Diego]
Tom: i'll have them before next weeks call
<Ralph> [VM] comments on 16 March editor's draft [Ralph]
Elisa: I expect to be able to return a new version quickly once I get the comments from Tom and Mark
Guus: I'd like explicit emails to reach consensus
Elisa: OK
... I still hope to be able to attend the face-to-face
Guus: more SKOS discussion next week, and 2 more reviews of VM
[adjourned]
ACTION: Alistair to make a proposal for Issue 40 [recorded in http://www.w3.org/2008/04/01-swd-minutes.html#action06] [CONTINUES]
ACTION: Antoine will review Alistair's proposals w/r/t the relationship between the existing solution and the extension [recorded in http://www.w3.org/2008/03/18-swd-minutes.html#action06] [CONTINUES]
ACTION: Alistair to review Antoine and Guus' emails to move ISSUE-71 and -74 [recorded in http://www.w3.org/2008/03/18-swd-minutes.html#action09] [CONTINUES]