SemWeb Best Practices & Deployment WG -- 16 Jun 2005

<DavidW> IRC only: Benjamin N

<BenjaminNguyen> I'll be phoning in if I have anything long to say David.

-> previous meeting 2005-05-19

Admin

RESOLVED to accept http://www.w3.org/2005/05/19-swbp-minutes as the minutes of the 19 May telecon

ACTION: Guus to start a straw poll on new meeting day; Monday, Tuesday, or Wednesday -- all at 1700 UTC [DONE]

-> straw poll on telecon day and results

<aliman> monday 1700 UTC fine for me

Mike: I'm happy with Monday 1700 UTC

Elisa: works for me too

<BenjaminNguyen> I didn't vote, all is fine

RESOLVED to move the WG telecon time to Mondays 1700 UTC (1800 UTC in the non-DST period)

ACTION: Ralph post telecon date resolution to the list

RESOLVED next telecons: Monday 27 June 1700 UTC and continue bi-weekly as usual

ACTION: Ralph to start a poll on Thu/Fri 3-4 Nov vs. Fri/Sat 4-5 Nov vs. Fri/Sat 11-12 Nov. (noting the 11-12 dates conflict with OWL workshop) [DONE]

-> Galway f2f dates poll and results

DavidW: 15 responses; shows weak preference for the Fri&Sat before ISWC

<aliman> i don't mind about f2f dates

Elisa: Evan and I have a workshop on the 6th and Evan preferred to have a day break in between

RESOLVED: next face-to-face in Galway Fri-Sat 4-5 November 2005

Jeff: there is an ODBase conference on ontologies 31 Oct to 4 Nov in Cypress
... not sure if it affects me directly, though we submitted some papers

Liaison

2.1 Proposed resolution httpRange-14

ACTION: DavidW to identify the 4 httpRange-14 options [DONE]

-> httpRange-14 Options

ACTION: Chairs to discuss the httpRange-14 issue at the coordination level [CONTINUES]

Alistair: recall Dan Brickley's note "Some Things That Hashless HTTP URIs Can Name"

Ralph: the response thus far from the TAG is that they can't move on this issue without further technical input. Path of least resistance would be to endorse Tim's solution, then decide what to do about well-known vocabs that do not follow that.

DavidW: Need to address the issue of server-side processing at the same time though, as many see it as intertwined.

Phil: is there a deadline on the httpRange-14 issue?

David: we've been talking about httpRange-14 for a while, no pressing deadline

[post-meeting note: the TAG found a resolution to httpRange-14 at its 15 June meeting. Roy Fielding announced this resolution on 18 June.]

2.2 XML Schema Last Call

ACTION: Jeff to review XML schema LC draft [DONE]

Jeff: I've read the draft and don't have any additional comments to post. The XML Schema WG might not be aware of our requirements for schema datatypes. Jeremy has been talking with them about this

Ralph: If we want the spec to change materially during the Candidate Recommendation phase, we should point to an implementation that fails to do something important because it is missing a feature. Otherwise unlikely to get changes enacted.

Jeff: Is there time for us to make more noise on this?

Ralph: Last Call ended in April, so either they're dealing with a long list of issues or they think they're done. I don't know which.

ACTION: DavidW ask about the XML Schema Component Designators LC status at the SemWeb CG meeting

2.2 OMG: ODM review

Elisa: I'll send a pointer to the latest document revision
... we've incorporated most of the feedback received
... more feedback is still welcome
... expect one more round on the document between now and August
... expect to use MOF Query View Transformation to represent all the mappings
... look forward to an OMG vote in December
... current draft does incorporate NIST feedback

[post-meeting note: Elisa sent mail regarding latest ODM specification]

TF Updates

3.1 PORT

Alistair: comments on SKOS Core have started to trickle in;
... we now have two open proposals; (1) bug fix, (2) some changes in documentation properties
... re: documentation properties, it makes sense to have a single root property and use dcterms:audience
... continuing to discuss DanBri's idea to have a property to relate SKOS concepts to OWL individuals
... we've been asked about how to extend SKOS Core; we expected to describe this at a later date, either in a separate note or in the Core Guide itself. My current thought is a separate note
... also questions on how to relate SKOS Core to XML Applications; in particular, how to write an XSD for SKOS Core
... OWL has an alternate XML syntax; does it get used?

David: yes, I've seen the alternate OWL XML syntax used

Alistair: NewsML is revising their schema
... will be hard to use the RDF/XML syntax in these applications
... I may start to float some ideas

DBooth: regarding a new XML syntax, are you thinking of one that is already in use or design a new one?

Alistair: I'm thinking of designing a new syntax that is XML Schema constrained but GRDDL-able

DBooth: do you expect an XLST transform to be part of the design?

Alistair: yes

Ralph: I'm think that design of a new syntax that is as extensible as RDF/XML yet XML Schema validatable is a lot of work

Alistair: I'm encouraged by the direction of RDF/A

3.2 OEP

Chris: Specified Values became a WG Note on 17 May
... still working on n-ary relations document [editor's draft]
... talking about publishing a vocabulary to support n-ary relations note
... so people who want interoperability between RDF and higher-arity systems have a way to do automatic translation
... simple part-whole note still pending
... Jerry Hobbs is now participating and we're talking about turning the OWL Time document into a Note
... may have something by the end of the month
... Evan says he has some initial work not yet posted on units of measure
... "argument number" property was included in the n-ary relations vocabulary to support translation to other formats
... some people felt this vocabulary to support translations to other languages was out of scope for the note

DavidW: can you satisfy both camps by using this vocabulary in a non-normative example?

Chris: anyone wanting to translate RDF to a higher-arity system would need such a vocabulary

Phil: I would like to see this included, as it's important to many

Chris: the question is about scope; the OEP Note is about particular patterns, not about translation to other languages

DavidW: Best Practices gets to be practical about dealing with issues that are important

Phil: ordered n-ary patterns are an identifiable pattern

Natasha: I felt this was out of scope because the Note is not about mapping
... argument numbers won't satisfy UML mapping
... Developers would need an additional vocabulary for mapping to languages that represent n-ary relations in other ways [than argument numbers]

Ralph: It's useful to have examples that help explain this. Need to have a vocabulary for those examples. The boundary here is whether the WG proposes a specific vocabulary or only provides examples.

Natasha: there were two parts to the vocabulary and I felt the specific parts dealing with the mapping were out of scope

Phil: ordered n-ary relations are a valid specialization

Chris: if we decided it is out of scope for this Note, we could write a separate Note
... so the question is whether to expand the N-ary relations Note or write a separate Note

Mike: would it make sense to use the vocabulary in the examples of the first note and later write a separate note about the vocabulary?

Natasha: I had two concerns; whether a mapping vocabulary belonged in the N-ary relations note and the systems not covered by the current vocabulary
... we might want to tackle other things such as association classes as well

Mike: I recommend including [only] things that are needed to explain the current N-ary relations note

Ralph: keep it simple

Chris: the vocabulary is there for n-ary relations. Separating it into another document may make it harder to find.
... saying how to represent higher-arity relations in RDF is part of this Note

Mike: could also write an appendix of what this might look like and the appendix might eventually expand to a new document

Chris: I'd rather not take that approach

Mike: if the material exists, it makes sense to have it all in one place

Chris: I wouldn't want a half thought-out appendix suggesting a fuller version would follow later

Ralph: but the partial solution is partial only in that it doesn't cover other cases, as Natasha cites, but you're comfortable for the cases it does cover, right?

Chris: yes

Phil: I feel a separate note is better

3.3 WordNet

ACTION: Aldo to propose an update the Wordnet TF description [CONTINUES]

3.6 RDF-in-XHTML

ACTION: Gavin find out from his community and contacts if they have use cases [CONTINUES]

-> meeting record 2005-06-14 XHTML telecon

Ralph: new XHTML2 WD clarified more about their WG process
... good news is there is a WD
... In my opinion not sufficient progress from last spec; I preferred the more explicit wording of the October RDF/A whitepaper
... no TF telecon for a while, telecon attendance low, though I missed the previous telecon

DavidW: should we get SWPB volunteers to review draft?

Ralph: we need help, sure
... problem with b-nodes. Thought the authors would have additional attributes for b-nodes
... something analogous to nodeID attribute in rdf/xml
... WG seems to prefer xptr scheme for bnode
... which scheme will win -which is more natural for HTML authors
... no new input on GRDDL question
... should WG take up GRDDL as a task or endorse it still an open question
... do have a document from XHTML WG that is close (modulo bnode issue)
... waiting for JJC to come back
... language may not be precise enough

Phil: why did the XHTML wg change its stance

Ralph: (around bnodes) not really a change, wasn't sufficiently addressed in previous (October) note (ie it was a bug)
... four solutions on table and need to pick one

DavidW will Ralph review new WD?

Ralph: Yes
... the TF considers reviewing this draft a high priority
... need Jeremy for this too

3.10 SE TF

-> Minutes of SETF Telecon 07-06-05

Phil: we'd like reviewers for Composite IFPs draft in 3-4 months
... using more than one resource to identify a resource
... we'd like reviewers for current Ontology Driven Architectures and Potential Uses of the Semantic Web in Software Engineering now

DavidW: recommend asking on the list, since attendance here is light
... I may be willing to be one reviewer

<BenjaminNguyen> I am interested also, depending on review deadline

Chris: I'm willing to review but I'm listed as a contributor

Phil: your contributions were very early on

Phil: Benjamin's offer accepted, there's no particular urgency so 4+ weeks to review OK

<BenjaminNguyen> fine

RESOLVED: SE draft reviewers are Chris and Benjamin with DavidW likely to comment

Summary of Action Items

[NEW] ACTION: DavidW ask about the XML Schema Component Designators LC status at the SemWeb CG meeting
[NEW] ACTION: Ralph post telecon date resolution to the list

ACTION: Aldo to propose an update the Wordnet TF description
ACTION: Chairs to discuss the httpRange-14 issue at the coordination level
ACTION: Gavin find out from his community and contacts if they have

[DONE] ACTION: DavidW to identify the 4 httpRange-14 options
[DONE] ACTION: Guus to start a straw poll on new meeting day; Monday, Tuesday, or Wednesday -- all at 1700 UTC
[DONE] ACTION: Jeff to review XML schema LC draft
[DONE] ACTION: Ralph to start a poll on Thu/Fri 3-4 Nov vs. Fri/Sat 4-5 Nov vs. Fri/Sat 11-12 Nov. (noting the 11-12 dates conflict with OWL workshop)

SemWeb Best Practices & Deployment WG

16 Jun 2005

Attendees

Contents