SWBPD Vocab Managment TF -- 12 May 2005

<tbaker__> yes, but I'm having trouble with it

<tbaker__> got Ralph's voice saying this is the W3C... etc then it dropped...

<tbaker__> http://lists.w3.org/Archives/Public/public-swbp-wg/2005May/0071.html

<danbri> elisa: coming soon, new ODM draft!

-> http://lists.w3.org/Archives/Public/public-swbp-wg/2005May/0071.html Tom's talking points

TomB: the narrower scope feels much more manageable

<Zakim> danbri, you wanted to comment re rescoping

Alistair: there's an issue in Quick Guide to Publishing a Thesaurus on SemWeb that relates to VM

DanBri: the cut-down proposal focuses on RDF vocabularies. good to have this crisp focus
... lots of good text in the Wiki document that should not be lost

<tbaker__> Ralph: agrees on narrower scope.

<tbaker__> how do we get traction on drafting words that can become a document?

<tbaker__> applaud Alistair on starting

<tbaker__> was comfortable with most of wiki draft and would like to see some migrate to new one

<tbaker__> we noted that not everyone understands importance of maintaining stable URIs

<tbaker__> those kinds of concepts are important in initial draft

<tbaker__> my margin notes: technical issues we may or may not be able to get into

<tbaker__> how do we version namespaces? etc

<tbaker__> if it feels good to us to say that we're focusing on RDF vocabularies, then take subset from wiki - good next step

<tbaker__> Alistair: when I hacked note - try to look from potential user's point of view

<tbaker__> i want to know "what can I know from this"

<tbaker__> we say what we can say, and that's it

<tbaker__> say the things that we definitely can say - and that is our scope

<tbaker__> danbri:

<tbaker__> users of w3c tech end up feeling stupid because they cannot figure out

<tbaker__> e.g., xslt with rdf...

<tbaker__> so if there are tricky or unsolved problems

I agree that the WG should not invent "best practices"

<tbaker__> Ralph: we should document those things we consider BP

<tbaker__> strength: DC/FOAF/SKOS have faced these issues - extracting from that experience is really valuable

Tom: ok to scope on what we know, but would like to push a little beyond that
... e.g. evolving issues section
... broader scope than just what can be said with certainty
... I'd like to salvage some of the Wiki issues, perhaps in shorter form
... note design and policy decisions even when best practice is not fully baked

<tbaker__> Alistair: 99% of potential users

Alistair: I'd like the note to take a strong stance from the point of view of the users
... users are not interested in unsolved issues
... they want to know only what they need to know
... need to understand the user community
... I'd like never ever again to refer to 'hash vs. slash'
... this just draws them into a long historic debate

<danbri> ralph: i was going to ask about target readership...

<danbri> ...sympathise re not exposing students (?) to debates which only give them a sense of some deep topic which they'll not find an answer to

<danbri> ...at same time, people trying to use this stuff, often get perplexed by things for which there is no answer

<danbri> ...shouldn't be afraid to mention issues for which we've no answers

<danbri> "yes we know there's an issue, but we don't have an answer"

<aliman> sounds good ralph

<danbri> ...we perhaps don't need to give as much rationale for why something is best practice

<danbri> ...i sometimes go overboard on that

<aliman> most users say: tell me what to do?

<danbri> ...you suggest that the target readership may not care on that

<tbaker__> Danbri: agree with Ralph - while we shouldn't scare people with academic stuff

<tbaker__> there are often some practical questions - what mime type should I configure my server,

<tbaker__> can I put http doc there and content-negotiate, etc.

<aliman> +1 on both Ralph & danbri

<tbaker__> Ralph: we can recommend hash without even hinting at debate

<Zakim> Ralph, you wanted to comment on acknowledging issues

<Zakim> danbri, you wanted to [dis?]agree with al

<tbaker__> Danbri: what to put at URI - if reader clicks and can;t get something readable, etc

[use case for GRDDL -- publishing a text/html document at a namespace URI that contains the RDF Schema / OWL Ontology ]

Alistair: good to take a look at the issues from a very pratical perspective
... e.g. what mime types should be used where

<danbri> """Where the resources that are the members of an RDF vocabulary are denoted by HTTP URIs, an HTTP GET request with the header field 'accept=application/rdf+xml' against that URI should return an RDF/XML serialisation of an RDF graph that includes a description of the denoted resource."""

Alistair: can say your URIs should support application/rdf+xml without going into the hash vs. slash debate

Tom: see point 9 of my talking points
... Andy Powell has a nice paper addressing point 9d
... would be useful to summarize in 2-3 sentences and cite other documents like Andy's paper
... don't think we want to address Vocabulary Ownership
... propose VM telecons every other week
... e.g. between WG weeks

<tbaker__> Tuesdays 1300 UTC

<tbaker__> June 7

<tbaker__> June 21

Alistair: the Publishing Thesaurus Quick Guide has different URIs from what was originally published

Ralph: yes, I changed those UKAT URIs because W3C PubRules does not permit citing URIs that are 404

Alistair: the community does not understand the issue
... getting them to use URIs at all is a big achievement
... there is a downloadable RDF version of UKAT and GEMAT (sp?) that use URIs that do not dereference

Ralph: oops, I may have acted too hastily then in changing the Quick Guide

<danbri> Server: Apache/1.3.27 (Unix) PHP/4.3.10

<danbri> http://httpd.apache.org/security/vulnerabilities_13.html "Fixed in Apache httpd 1.3.33"

DanBri: I found a web server at example.org and example.com
... I am worried that there is an insecure Web server that could be logging accesses to example.{com,org}
... this could scare people away from using http:

<tbaker__> danbri, sounds like a good footnote

DanBri: e.g. URN crowd has taken advantage of concerns over http
... need to inform people of social policies around namespace documents
... e.g. changing your schema changes the interpretation of existing documents

next meeting: 7 June, 1300 UTC

<scribe> ACTION: Alistair post a starting draft in the VM area on www.w3.org [recorded in http://www.w3.org/2005/05/12-swbp-minutes.html#action01]

<tbaker__> i will post a brief report tomorrow

-> http://www.w3.org/2001/sw/BestPractices/VM/ the VM TF area

<DeborahM> hi

<ChrisW> agenda item 1: N-ary note

<ChrisW> issue of the first two patterns being different or not

<ChrisW> is this worth putting in a note

<ChrisW> http://smi-web.stanford.edu/people/noy/nAryRelations/n-aryRelations-2nd-WD.html

<ChrisW> latest editors draft

<ChrisW> Mike, Pat & Chris of the opinion that "pattern1" and "pattern2" are not different patterns, just different use cases

<ChrisW> Chris: called them different use cases removes the need to discuss why they are different so extensively,

<ChrisW> ... as the note seems to have gotten too long.

<ChrisW> ...suggest calling it one pattern, and removing some of the discussion as to why they are different

<ChrisW> MikeU: suggest talking about the naming conventions used after each pattern

<ChrisW> Natasha will talk to Alan about steps forward

<ChrisW> ACTION: Chris to send around n-ary vocabulary in OWL Full [recorded in http://www.w3.org/2005/05/12-swbp-minutes.html#action02]

<ChrisW> Chris: WG is very positive about the idea of publishing actual ontologies in w3c URI space

<ChrisW> ...many requests to have a "standard" way to annotate "Classes representing n-ary relations" in order to support interoperability

<ChrisW> Agenda item 2: simple part whole

<ChrisW> Discussion of whether the RDFS pattern is really needed. All it says is that partOf is a property.

<ChrisW> This doesn't appear to be very useful to have in note, the RDFS primer tells people how to declare properties in RDF

<ChrisW> Natasha: lots of experience from Protege users, they expect OWL to have "partOf" built in, and we

<ChrisW> ...get a lot of questions on how to model it

<ChrisW> ...saying that "it's just a relation" is actually a useful thing to say.

<ChrisW> ...for some reason a lot of people think it's different, so this can help

<ChrisW> Chris will work with new individual pattern and merge into the note.

<ChrisW> Agenda Item 3: OWL-Time

<ChrisW> Group welcome Jerry Hobbs to OEP at long last

<ChrisW> looking forward to having OWL-TIme a W3C note

<ChrisW> Chris: I have found - for purely aesthetic reasons - that having an "event" class in OWL-Time is impure

<ChrisW> Jerry: Does no harm, simply allows for events as well as time points to be order according to Allen relations

<ChrisW> Chris: Agreed is does no harm - this is purely aesthetic. Suggest separating "core" OWL time from this layer using imports

<ChrisW> ACTION: Chris to send around a version that has been split this way. [recorded in http://www.w3.org/2005/05/12-swbp-minutes.html#action03]

<ChrisW> Jerry: publishing OWL-Time in the W3C space a desireable thing. Looking at June before work can begin on this

<ChrisW> Agenda Item 4: Units & Measures

<ChrisW> Elisa: still making slow progress on evaluation criteria for all the ontologies out there

<ChrisW> ...ODM still a big obstacle, finished in two weeks and then this will start moving

<ChrisW> ...being able to publish this in W3C space a big motivation

- DRAFT -

SWBPD Vocab Managment TF

12 May 2005

Attendees

Contents

Summary of Action Items

Scribe.perl diagnostic output