See also: IRC log, previous 2007-10-23
PROPOSE to accept the minutes of the F2F with
...amendments from
Antoine
...http://lists.w3.org/Archives/Public/public-swd-wg/2007Oct/0107.html
...http://lists.w3.org/Archives/Public/public-swd-wg/2007Oct/0109.html
... Additional amendment from Alistair
... We make no statement either way about whether skos:Concept
is disjoint or not disjoint with owl:Class
... and from Tom
...http://lists.w3.org/Archives/Public/public-swd-wg/2007Oct/0137.html
RESOLVED to accept the minutes with the above amendments
<scribe> ACTION: Ralph to update face to face minutes to reflect the amendment and capture attendees[recorded in http://www.w3.org/2007/10/30-swd-minutes.html#action01]
PROPOSED to accept minutes of the Oct 23 telecon:
http://www.w3.org/2007/10/23-swd-minutes.html
RESOLVED
Guus: Nothing to track here for
this telecon
... comments have gone to SWEO group. For the moment,
... we can be silent on this.
Ralph: We're done with that task. Can remove from agenda.
<Ralph> http://www.w3.org/2007/10/23-swd-minutes.html#action14
<Antoine> http://lists.w3.org/Archives/Public/public-swd-wg/2007Oct/0139.html
Guus: The minutes capture the interpretation.
<aliman> s/transitive/functional/
Ralph: We're happy that it's not functional?
Guus: Yes, there may be examples where it's not.
Ralph: Context of that discussion
were w.r.t the semantics of inScheme
... recently had a discussion in a separate group about
semantics of isDefinedBy
... would be nice to have even stronger semantics for
isDefinedBy
<aliman> I remember TimBL saying something like that about isDefinedBy in the context of the tabulator
Ralph: is this group interested in going further with this?
Guus: Can't do this -- we'd need to define our own property
<aliman> TimBL wanted to avoid having to make unecessary HTTP requests
Ralph: Would be acceptable for
this group to suggest what the semantics of
... isDefinedBy are.
... if they wished to
<aliman> +1 possibly useful interpretation, especially for large vocabularies -- typical for SKOS
Guus: If this isn't inappropriate, it might be useful
ACTION: Ralph to reconstruct proposal for semantics of isDefinedBy [recorded in http://www.w3.org/2007/10/30-swd-minutes.html#action02]
Alistair: Two separate
issues
... stronger semantics for isDefinedBy could be useful to
reduce HTTP call overhead
... also whether or not we think of inScheme as functional. Is
it reasonable to think of concepts in
... more that one scheme. If so, forks discussion as inScheme
and isDefinedBy go in different directions
Guus: propose we put this on the agenda once we have Ralph's message
Jon: Disagree with some of the
decisions we're making.
... But don't have supporting evidence, but things "feel
wrong".
Guus: We haven't actually made decisions yet.
Alistair: Jon refers to e.g. decision to deprecate inScheme and use isDefinedBy instead
Jon: Yes
Alistair: Justification is that copncepts get used in different schemes.
Guus: Discussion in Amsterdam are that current semantics of isDefinedBy are the same as inScheme
Alistair: What do we mean by
Ralph's interpretation?
... do we simply mean functional?
Guus: Rationale in going from
inScheme to isDefinedBy was that semantics of
... inScheme seemed to be the same as isDefinedBy. So go for a
general non-SKOS solution
Ralph: circular argument. isDefinedBy has such loose semantics, so such a claim seems unsatisfying
Quote was: isDefinedBy may be used to indicate an rdf vocab in which a resource is described
Jon: Is a concept scheme the same as an rdf vocabulary?
Guus: If we have evidence from use cases, we simply reopen the issue.
Jon: It's not whether or not it works, it's about concept scheme == rdf vocabulary
Alistair: Made a statement about
SKOS concept schemes and OWL ontologies, but no
... statement about "RDF Vocabularies".
Guus: Is this different because it would allow linking to an OWL vocabluary rather than a concept scheme
Jon: Wording of isDefinedBy talks about rdf vocabularies.
Antoine: Can see a problem.
Original guide says that inScheme is a link between concept and
concept scheme
... scheme it's part of. No
requirement that it has to give a meaning. isDefinedBy we'd
expect
... some *definition*
<aliman> From http://www.w3.org/2006/07/SWD/wiki/SKOS/Reference/ConceptSchemes ...
<aliman> "The choice to leave the formal semantics of skos:ConceptScheme undefined has been made to allow different design patterns to be explored for using SKOS in combination with more formal languages such as OWL, and for using SKOS with query languages such as SPARQL.
<aliman> For example, interpreting skos:ConceptScheme as a sub-class of owl:Ontology would be consistent with the SKOS semantics. This would also be consistent with using owl:imports to make logical import statements between SKOS Concept Schemes.
<aliman> Interpreting skos:ConceptScheme as a sub-class of the class of named RDF graphs would also be consistent with the SKOS semantics. This would also be consistent with using the name (URI) of a SKOS Concept Scheme in SPARQL queries as the name of an RDF graph, to establish, for example, the provenance of a semantic relationship between two SKOS Conceptual Resources."
<aliman> Except for the semantic condition stated above, the formal semantics of skos:ConceptScheme are undefined.
Alistair: talks through
definition
... this allows use of owl:imports rather than defining a new
import mechanism
... wanted to find solutions to problems without trying to
redefine or invent our own solutions
Guus: We should reopen the issue
regarding inScheme, but not w.r.t. containment,
... rather the choice of skos:inScheme and rdfs:isDefinedBy
Alistair: Several different flavours of inScheme.
Guus: Solution might be that someone proposes that we leave inScheme as a subproperty of isDefinedBy
Ralph: Don't see how that helps.
What's the range?
... understand the range of isDefinedBy to be RDF vocabulary
(whatever that is)
Guus: And range of inScheme is conceptScheme?
Ralph: Concept scheme doesn't necessarily describe concepts. It's a grouping of concepts.
Alistair: We never really got
into this much detail.
... didn't get into the whether things were defined. Used
language like concepts "participate" ina scheme
Ralph: Where, as a user, should I find a definition for a concept?
Alistair: derefence the URI
Sean: confused now about derefencing URI and isDefinedBy
<Elisa> If it's at all helpful, the "formal" definition of a "concept system" from ISO 1087 is "a set of concepts structured according to the relations among them". This doesn't necessarily say that a concept system is an ontology, but is somewhat stronger than a simple collection of concepts.
Alistair: If you're using slash
namespace, get lots and lots of redirects.
... if you have an isDefinedBy then you don't need to get it
every time.
Ralph: That's an operational
argument. Given a property URI, how do you find the namespace
within which that
... property is defined.
Alistair: Why can't you dereference a property URI?
Ralph: You can. Not clear where
in a chain of redirects you encounter the namespace
... can make some guesses.
Jon: Can somebody explain why
we're not defining a concept scheme as a collection?
... then use rdfs:member
Ralph: Does that satisfy the use cases?
Jon: Not certain.
Alistair: Reason why we might not
go there is that some people think of statements to be part of
a concept scheme.
... included that in original wording.
Guus: We have to round up this
discussion.
... For OWL ontologies, we never explicitly defined
containment.
... Reluctant to open a can of worms. However, we do need
to
... continue this discussion. Will someone reopen the issue?
Initially
... on inScheme vs isDefinedBy.
<aliman> It would be great Jon could point to some use cases re. potential issues with inScheme
ACTION: Antoine to summarise inScheme vs isDefinedBy and decide whether or not to reopen the issue. [recorded in http://www.w3.org/2007/10/30-swd-minutes.html#action03]
Guus: For next week would like to discuss Concept Schemes vs OWL Ontologies.
<Ralph> [likely regrets from me for next week, due to W3C big meeting week]
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]
Elisa: Vit and Elisa working on editors draft.
Elisa: trying to get front matter
in and organise things.
... Have posted an editors draft that pulls together everything
from the wiki
... but didn't pull in stuff from Best Practies WG.
... Have now posted a first editors draft that captures what
was in the wiki, boiler plate things,
... italicised sections for comments.
... Want to go back to original idea for saying what we're
going to talk about here
... is identification of versions rather than version
management.
<Ralph> http://www.w3.org/2006/07/SWD/Vocab/principles
<Ralph> Vocab Management editors' draft
<Ralph> [I will add a citation to that on the WG home page]
Elisa: Additional food for
thought in research items. Bottom line is that there is
something up there
... fills in some of the blanks. Worth looking at the outline
to see if there are topics that should be
... addressed but are not, e.g. provenance.
... schedule is to iterate one more time over next couple of
weeks and integrate feedback.
... After that we'll be ready for internal review.
... Need another couple of weeks before internal review.
Guus: telecon next week, no telecon in two weeks.
Meeting adjourned