W3C

GRDDL Working Group Meeting Minutes

11 Jul 2007

See also: IRC log, agenda

Attendees

Present
DanC, john-l, Chimezie_Ogbuji, HarryH, rreck, FabienG, jjc, danja
Chair
Harry
Scribe
john-l

Contents


Convene GRDDL WG meeting of 2007-07-11T11:00-0400

<HarryH> PROPOSED: to approve GRDDL WG -- 27th Jun 2007 as a tru e record

<HarryH> http://www.w3.org/2007/06/27-grddl-wg-minutes.html

<HarryH> RESOLVED: approved GRDDL WG -- 27th Jun 2007 http://www.w3.org/2007/06/27-grddl-wg-minutes.html as a tru e record

<HarryH> Scribe?

<HarryH> DanC is scribe next time.

GRDDL Outreach

HarryH: What do we want to do with our time while GRDDL is in PR?
... Outreach!

<HarryH> http://lists.w3.org/Archives/Public/public-grddl-wg/2007Jun/0385.html

HarryH: Dublin core metadata profile need to be GRDDLified.

<HarryH> Andy Powell?

<HarryH> Eduserv?

<DanC> danja, do you know what's up with dublin core profile stuff?

<DanC> i.e. does any of that leak from IanD to you?

<danja> no sorry, I completely forgot

<HarryH> Try to figure this out.

<danja> I can ask him now (in #talis)

<DanC> pls do

<Zakim> DanC, you wanted to comment on "Kingsley and try to get GRDDL EARL results "

DanC: Kingsley sent EARL results privately.

Chimezie: Received copy, haven't gr0ked, will ping about making results public.

<rreck> as for outreach, i was going to propose the use of GRDDL for a presentation at http://ncor.us/oic2007

<danja> DanC, re. DC -

DanC: I get to promote GRDDL at XML Summer School.

<danja> iand: no progress - i will email andy powell this afternoon

<HarryH> RSS?

<HarryH> Atom?

HarryH: Easiest deployment would be to seed namespace documents for main formats.

<rreck> if someone could help me see how to add a namespace correctly to Zthes i could do many ISO vocabs

<rreck> zthes doesnt use a namespace

<danja> HarryH, RSS 2.0 doesn't have an XML namespace :-(

<DanC> brainstorming on outreach possibilities: Atom, microformats profiles, dublin core, OAI, Creative Commons [timbl/tabulator]

<HarryH> Atom GRDDL?

<danja> not sure of current status of Atom XSLT - Henry Story has one for XSLT 2.0

<danja> also David ??? (will look up) has another, fairly verbose one

DanC: Once upon a time RSS had an RDF version, there had been hope of an RDF version of Atom, then there were efforts towards Atom OWL
... now there are bits of code for producing Turtle from Atom

<DanC> http://www.w3.org/2005/Atom is owned by the Atom WG

DanC: the Atom NS is owned by the Atom folks, whom we need to ask if we want to get a namespace doc at the NS URI.

HarryH: Need to convince the Atom WG that a namespace doc with a GRDDL ref is value added.

DanC: Need to get that by Tim Bray, who is... dubious about the value.
... Needs to be a well-engineered solution.

HarryH: Follow up with Henry Story about his transformation.
... Any takers?

<danja> I can

<danja> (I also did an XSLT myself, but am pretty sure it's far from complete)

<HarryH> Danja does this sound right - ACTION: danja to find out the status of Atom GRDDL and e-mail listt to find other people with transformations or schemas.

<DanC> which list? IETF Atompub WG?

<danja> fine

<HarryH> Danja does this sound right - ACTION: danja to find out the status of Atom GRDDL and IETF Atompub WG list to find other people with transformations or schemas.

<HarryH> ACTION: danja to find out the status of Atom GRDDL and IETF Atompub WG list to find other people with transformations or schemas. [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action01]

<danja> finer

HarryH: Anyone want to help rreck out with his namespace issues?

rreck: Representation of thesauri in XML

<danja> re. Outreach, is there anything that might be useful to pass through SWEO?

DanC: Looked at SKOS?

rreck: Yup!
... All the pieces are in place (e.g. translation to SKOS exists)

<rreck> http://zthes.z3950.org/

DanC and rreck discuss the possibility of adding namespaces to Zthes.

<DanC> (I wonder if this is worth some SemWeb CG time)

<DanC> (or SemWeb Deployment WG time)

HarryH: Stepping back, we may just want to continue discussing GRDDL deployment on the ML.

DanC: I can see using GRDDL offline.
... Keep cached copies of namespaced documents.

jjc: I could even have a URN, and simply state a relationship between a URN and a namespace document.

HarryH: But how do we bootstrap to that point?

DanC: But this is reasonable in an enterprise setting.

<danja> (re. Quick Reference Card - I plan to make public to coincide with PR...Agenda #5)

Straw poll: what would you do to support GRDDL?

<rreck> im bringing it up at a conference on saturday and in a paper that is due today

<chimezie> I plan to push the issue of using GRDDL for extracting links to bioinformatic ontologies from 'legacy' healthcare communication (XML) dialects

<chimezie> this is a theme i keep pushing and will keep pushing

john-l: Interested in helping with some follow-up with Atom.

<danja> I plan to fill in some of the gaps in XSLTs, also try and get some online demo stuff (@talis)

FabienG: Included GRDDL materials in our graduate school curriculum, integrating GRDDL transforms and agents in various internal projects

<DanC> (fabien explains lots of cool stuff; I hope the scribe is getting that, but if not, maybe fabien can elaborate in mail)

jjc: Maintaining Jena GRDDL support

HarryH: Could add microformat transforms to the XHTML namespace document

<danja> (might a page on the ESW Wiki for GrddlOutreach be useful?)

<chimezie> +q about DanC's comment to the HTML WG about keeping @profile (not related to advocacy, but)..

<DanC> (could be)

<danja> maybe a cellphone nearby

danja: I'd like to get some live demos with some practical, interesting results.

<DanC> ACTION: Chime to ask Kingsley about distribution of GRDDL EARL results [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action02]

<DanC> ACTION: IanD and Danja to e-mail maintainer of Dublin Core Metadata Profile to upgrade to GRDDL. [CONTINUES] [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action03]

Standard Library Transform

<danja> iand: no progress - i will email andy powell this afternoon

<DanC> indeed, http://www.w3.org/2001/sw/grddl-wg/library has @@s

jjc: There is code in the library that still needs work and exercise.
... There needs to be an action in this space to ensure work proceeds.

<jjc> http://www.w3.org/2003/g/xml-attributes

<chimezie> GRDDL Standard Transform Library

<DanC> xml-attributes.xsl,v 1.24 2007/06/25 10:52:08 jcarroll

jjc: The xml attributes library module needs review.

DanC: There are tests that exercise the critical features...

<scribe> ACTION: john-l to read through documentation for xml-attributes and to be sure there aren't any gremlins. [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action04]

<DanC> "Also, it doesn't use

<DanC> the @@@link W3C XSLT URI library."

<DanC> http://www.w3.org/2001/sw/grddl-wg/td/grddl-tests#grddl-library

<DanC> hmm.. no, that's just a link to the library

DanC: Should review what tests have what status.
... We didn't approve the library tests.

<danja> (btw, http://www.w3.org/TR/xlink2rdf/ may be worth looking at re. Standard Transforms)

<chimezie> xml:lang from xml-attributes

<chimezie> "unit"/"regression" tests

jjc: I can partition out the tests relevant to the library to their own list.

<DanC> ACTION: jjc to organize library tests; link from http://www.w3.org/2001/sw/grddl-wg/library [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action05]

<DanC> ACTION: jjc to dispose of @@s in http://www.w3.org/2001/sw/grddl-wg/library (and in docs of 3 txforms) [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action06]

Relationship to XHTML 2 and HTML 5

<DanC> Harry, did you send something to the XHTML 2 WG?

<DanC> that's supposed to be in place, e.g. for the meeting later today

<HarryH> I did email Steve Pemberton directly.

<DanC> with public archive?

<HarryH> Yep.

<DanC> ok

<HarryH> http://lists.w3.org/Archives/Public/public-grddl-wg/2007Jun/0273.html

<HarryH> Note Steve is cc'ed.

<HarryH> No response.

<DanC> I wrote http://lists.w3.org/Archives/Public/public-html/2007Jul/0571.html

<HarryH> I will take an action to follow up with Steve if needed.

<DanC> ... in the name of this WG, without review.

chimezie: I like it.

<HarryH> So do I take an action for that?

DanC: At least there isn't a lot of negative reaction.
... It's possible that it's just being ignored.

<DanC> no, no more action needed, Harry

<HarryH> My only fear would be if XHTML also drops profile

<HarryH> in XHTML2

PR request

<HarryH> The call is today 3PM EST.

<HarryH> I have no major concerns.

DanC: There might be a loop to close with respect to XQuery and Schemas...

chimezie: How does that relate to the URN case?

DanC: In the URN case, you're using a custom resolution mechanism...
... In another case, there's an undefined binding between the document and the schema.
... That would be a "larger asterisk".

john-l: Implementation report was confusing?

<chimezie> If need be , I can suggest text for covering the XML CMS link to schema usecase..

<chimezie> Testing for Multiple Representations

Discussion of "Not Applicable" confusion in the test results document.

Looking at http://www.w3.org/2001/sw/grddl-wg/td/grddl-tests#langconneg1 and the two following tests...

Each of these three have the same "input" URI.

Thus, they are alternatives.

<chimezie> gTest:subsumes is used for the maximal conneg situation

<chimezie> otherwise gTest:alternative is asserted where tests share the same input

jjc: HP would have a preference to be finished before 2007-09-05

DanC: Looking like 2007-09-11 at the moment
... 2007-07-13 or 17 for PR, perhaps
... Then a number of members need to encourage it going to Rec
... Also important is press release testimonials

chimezie: I prefer 2007-07-13

<DanC> anything else re the PR request, HarryH ?

#issue-base-param

<DanC> . ACTION: DanC to remove base param from online GRDDL service and home2rss.xsl

DanC: This action should go under outreach.
... Dom is looking to migrate to another GRDDL implementation.

chimezie: What infrastructure is present?

DanC: Precedent exists for Jena

<DanC> ACTION: DanC to replace the W3C GRDDL service implementation with Jena or redland [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action07]

<DanC> and my other action is withdrawn

DanC: ETA on Jena release?

jjc: ~PR time, 13 July

SPARQL WG wants to know about our test harness

<danja> ew

Discussion ensues to some end talking about existing tools for manipulating EARL.

DanC: I've sent mail about my work; if they need me to explain it further, then it's prolly not mature enough. if they can go with what I've done and explained to others, maybe it is.

GRDDL press release noodling

DanC: Press release usually goes with Rec.
... Someone else is working on the first draft of the press release.
... Anyone else want to talk to reporters?
... We also need testimonials from member companies.

<DanC> (looking at http://www.w3.org/Press/ )

<DanC> http://www.w3.org/2004/01/sws-testimonial

jjc: Press release aimed at mid September?

DanC: Yes.

<danja> Talis may be willing to provide a testimonial, I'll ping iand

DanC: Get your name in lights, write a testimonial!

jjc: HP, yes.

chimezie: Yes.

<rreck> possibly

<danja> may be worth pinging sweo too

<DanC> yes, please, danja

<FabienG> I can help with french translation ;-)

<FabienG> amour, amour amour

<danja> er, HarryH "we need testimonials by July" - end of?

<HarryH> End of July at latest, better if earlier.

OpenID and FOAF and GRDDL

<HarryH> Also, let's ask Kingsley for a testimonial.

DanC: OpenID: use your homepage as your login key.

<HarryH> ACTION: HarryH to ask Kingsley for a testimonial [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action08]

DanC: also like callback auth using web pages instead of email.
... I think it would be cool if we could mix RDF metadata into such webpages using GRDDL.
... How do we trust this data?

<DanC> http://dig.csail.mit.edu/breadcrumbs/blog/2

DanC: Web log spam sucks, so now we require OpenID logins.
... But a spammer can make up an OpenID login.
... Now, Advogato IDs include FOAF with relationships of trust.
... This is resistant to spam through earned RDF karma.
... So you might be able to comment if you pass a certain karma policy.

<chimezie> sounds like a similar problem space as PAW

DanC: Advogato's FOAF files don't have all the pretty data you might want, though.

<rreck> sounds interesting

<DanC> ADJOURN.

Summary of Action Items

[NEW] ACTION: Chime to ask Kingsley about distribution of GRDDL EARL results [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action02]
[NEW] ACTION: DanC to replace the W3C GRDDL service implementation with Jena or redland [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action07]
[NEW] ACTION: danja to find out the status of Atom GRDDL and IETF Atompub WG list to find other people with transformations or schemas. [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action01]
[NEW] ACTION: HarryH to ask Kingsley for a testimonial [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action08]
[NEW] ACTION: jjc to dispose of @@s in http://www.w3.org/2001/sw/grddl-wg/library (and in docs of 3 txforms) [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action06]
[NEW] ACTION: jjc to organize library tests; link from http://www.w3.org/2001/sw/grddl-wg/library [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action05]
[NEW] ACTION: john-l to read through documentation for xml-attributes and to be sure there aren't any gremlins. [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action04]
 
[PENDING] ACTION: IanD and Danja to e-mail maintainer of Dublin Core Metadata Profile to upgrade to GRDDL. [recorded in http://www.w3.org/2007/07/11-grddl-wg-minutes.html#action03]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.128 (CVS log)
$Date: 2007/07/11 16:36:19 $