See also: IRC log
<trackbot> Date: 02 September 2010
<DKA> Scribe: Dan
<DKA> ScribeNick: DKA
Resolved: minutes of 19th approved
Noah: Call for next week is at risk depending on getting agenda / chairing set up...
hrm
Noah: on Webapps - I want to make some progress in September. Any thoughts?
<noah> Let me state a bit more forcefully on WebApps: I don't think our level of investment and rate of progress has been consistent with our agreement that significant writing on WebApps would be one of our major goals for the year.
<noah> I intend to work with TAG members in Sept to see whether serious writing can be done in time for discussion at the F2F.
<Zakim> ht, you wanted to ask for a session on 3987
Henry: I support face-2-faces...
Larry: In the last week, the IETF area directors have got together with the wg chairs to push the work forward.
Henry: I feel we can't usefully respond to Roy without knowing if your idea for re-architecting the situation has support.
Larry: The problem is: there's
some work that needs to get done to resolve the differences
between what the specs currently say and what really happens
and what should happens...
... also there is a venue discussion (w3c-whatwg-ietf).
... or the unicode consortium...
<ht> LM, last spring during an MIT TAG meeting we walked together to the pub, and you described your ideas for reworking the whole idea of URI grammar
Larry: the only rational way of making progress is to start doing some of the work...
<noah> ACTION-409?
<trackbot> ACTION-409 -- Henry S. Thompson to run Larry's plan for closing IRIEverywhere by the XML Core WG -- due 2010-06-22 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/409
<noah> ACTION-410?
<trackbot> ACTION-410 -- Larry Masinter to let the TAG know whether and when the IRIEverywhere plan in HTML WG went as planned -- due 2010-11-01 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/410
<noah> ACTION-448?
<trackbot> ACTION-448 -- Noah Mendelsohn to schedule discussion of http://lists.w3.org/Archives/Public/public-html/2010Jun/0394.html on 26 August (followup to 24 June and 12 August discussion) -- due 2010-09-28 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/448
Larry: the driver for making this
happen is IDN - because if you translate from an IRI with an
international name not in ascii to a URI then you get something
hex-encoded which you have to unencode in order to IDN it...
which presents some problems.
... Roy's point was slightly different.
<noah> http://lists.w3.org/Archives/Public/public-html/2010Jun/0394.html is about URL processing in HTML
Noah: Happy to see people pushing on substantive work leading up to the f2f. I have put action 409, 410 and 448 into the IRC - Henry can you comment?
Henry: I think 409 is done.
Noah: any objections?
[none heard]
<noah> close ACTION-409
<trackbot> ACTION-409 Run Larry's plan for closing IRIEverywhere by the XML Core WG closed
<masinter> The IETF IRI working group chairs have indicated that they're going to start going through issues .... I hope that will result in making some progress
<noah> HT: They have replied to us
action-410?
<trackbot> ACTION-410 -- Larry Masinter to let the TAG know whether and when the IRIEverywhere plan in HTML WG went as planned -- due 2010-11-01 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/410
Noah: Do we need any new actions?
Henry: No. I want a session at
the f2f to hear from Larry about this.
... [of IRI bis status and the status of larry's
proposal]
<noah> ACTION: Noah to schedule F2F discussion of IRIbis status and Larry's proposal due: 2010-10-05 [recorded in http://www.w3.org/2010/09/02-tagmem-minutes.html#action01]
<trackbot> Created ACTION-459 - Schedule F2F discussion of IRIbis status and Larry's proposal due: 2010-10-05 [on Noah Mendelsohn - due 2010-09-09].
Noah: anything else on oct f2f?
Privacy workshop report: http://www.w3.org/2010/api-privacy-ws/report
DKA: See workshop report link http://www.w3.org/2010/api-privacy-ws/report
DKA: Workshop headlines: 1) very well attended, some reported it as most comprehensive in 5 years.
DKA: Participation from academic groups, good representation from IETF
DKA: We discussed need for better coordination between the IAB and the TAG...look for better progress now that summer is over
DKA: There was lots of focus on device APIs. What's been learned from geolocation api deployment. Should privacy information be carried along with device data (e.g. location) in the context of an API call. Also a UI dimension.
DKA: Privacy questions may have to be asked at time that user is asked for permission to collect data.
DKA: We also discussed "privacy rulesets", presented by ????. Creative commons-like model that allows users to pick from standardized options for privacy settings.
DKA: Can link a license for that piece of data, the link being carried along in user agent and onward into the network. Can indicate preference for allowing 3rd party access, etc.
DKA: In summary, it was a very good opportunity for discussion. We probably achieved somewhat less consensus than I hoped, but the topics discussed were very pertinent for the DAP F2F that followed immediately after the workshop. Chairs reported it was valuable.
DKA: Next steps are that we need to figure out coordination between TAG and Internet Architecture Board. I could take an action.
<noah> ACTION: Appelquist to coordinate with IAB regarding next steps on privacy policy [recorded in http://www.w3.org/2010/09/02-tagmem-minutes.html#action02]
<trackbot> Created ACTION-460 - Coordinate with IAB regarding next steps on privacy policy [on Daniel Appelquist - due 2010-09-09].
<noah> ACTION-460 due 2010-09-14
<trackbot> ACTION-460 Coordinate with IAB regarding next steps on privacy policy due date now 2010-09-14
DKA_> http://lists.w3.org/Archives/Public/public-privacy/
<DKA_> ScribeNick: DKA_
Noah: Are you [dan] or is someone else willing to take on a writing assignment about "what the tag wants to tell the world about API design issues for webapps or something smaller like policy..."
<noah> . ACTION: Appelquist to draft "finding" on Web Apps API design
<noah> ACTION: Appelquist to draft "finding" on Web Apps API design [recorded in http://www.w3.org/2010/09/02-tagmem-minutes.html#action03]
<trackbot> Created ACTION-461 - Draft "finding" on Web Apps API design [on Daniel Appelquist - due 2010-09-09].
<noah> ACTION-461 due: 2010-10-11
Tim: [+1 to Noah's comments on producing substantive findings]
Noah: Roy's finding on authoritative metadata is a good example of a good finding.
Noah: I thought the idea was to
have it at our f2f on the west coast...
... I've had some reservations...
DKA: I took action to look into logistics. Raman seemed a bit negative, so I went looking for alternate hosts who migh contribute space.
DKA: Goal is/was co-location with TAG meeting. Actually spoke with Carnegie Mellon West, but they couldn't manage it either.
DKA: At this point, I'm not sure how much energy I have to push this forward.
<noah> Chair is curious whether anyone else thinks this is high value?
<noah> We need to settle soon so travel can be arranged.
<noah> We also need to reach out to other attendees.
<noah> I remain somewhat skeptical, but maybe I'm being too conservative.
<johnk> I still think it's a good idea, but I don't have the time to do anything any more prior to October
Tim: We could pick well-known established architects and/or people who have been making decisions that we care about...
Noah: What kind of format and invite list would you have in mind?
Tim: For format: maybe get people we don't know to present what the most important properties that we haven't mentioned in the architecture document?
Noah: Invitation-only? Or open to
anyone who signs up?
... Day-long thing? or Smaller meetings?
... what kind of discussion are we trying to foster with
whom?
DKA: Could do day long, or just invite experts.
DKA: Logistics would be much easier.
Noah: As I think about pulling
this together - within about 2 weeks we need to know which days
we need to be on the west coast. Currently we have 19th-21st.
If anyone needs to come on the 18th [e.g.] then we need to know
in 2 weeks.
... option B (invited experts coming to talk to us) is less
difficult.
... Anyone want to push for a full-day developer-camp style
thing? If not, I propose we let it go...
... Anyone else interested in working with Dan on option (A) -
a bigger developer camp on a different day?
Tim: My concern about the "big" one is peoples' time may already be committed.
Noah: What I'm hearing is it doesn't work... If you have ideas for other things that might fit in the 3 days - but inclined to let go.
<noah> ACTION-454?
<trackbot> ACTION-454 -- Daniel Appelquist to take lead in organizing possible Web apps architecture camp / workshop / openday -- due 2010-07-22 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/454
<noah> . ACTION: DanA to Take lead in organizing outside contacts for TAG F2f
<noah> ACTION-455?
<trackbot> ACTION-455 -- Noah Mendelsohn to schedule discussion on privacy workshop outcomes. -- due 2010-09-07 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/455
<noah> close ACTION-455
<trackbot> ACTION-455 Schedule discussion on privacy workshop outcomes. closed
Noah: any objection to close 455?
[none heard]
Larry: I have no opinion on this. I see nothing that I object to. I'd be happy for it to go either way.
<noah> http://www.w3.org/2001/tag/2010/08/12-minutes#item06
Tim: I was very concerned about this. I've got code in tabulator that throws up an error message when it hits this.
<noah> ACTION-456?
<trackbot> ACTION-456 -- Yves Lafon to locate past HTTP WG discussion on Location: A#B change, and make the TAG aware of it -- due 2010-08-17 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/456
<noah> Tim: I was hoping TAG would promote best practices
Tim: I'd been hoping that the TAG would say "there are best practices for redirecting from the object to the document about it"
Tim: we see PERL site redirects with 302 from A#B to C#D, and then discover that there's no info in that result about C#D, only about A#B
Tim: my code is unhappy with this
<Yves> the redirection is from A to C#D
<Yves> handling #B is done client-side
Noah: Can the TAG write on this?
<timbl> Ok, my current problem is with IIRC dcterms:title
<noah> LM: I thought HTTP redirect is being addressed in http committee. Thus no need for TAG finding.
Larry: I thought that the HTTP redirect was being addressed in http bis committee and that the TAG didn't need to write a finding.
<noah> TBL: The httpbis committee?
<noah> LM: Yves?
<timbl> _________________________
Yves: My impression is that there are more people interested in solving that issue here than in http bis...
<timbl> $ curl -I http://purl.org/dc/terms/title
<timbl> HTTP/1.1 302 Moved Temporarily
<timbl> Date: Thu, 02 Sep 2010 17:59:31 GMT
<timbl> Server: 1060 NetKernel v3.3 - Powered by Jetty
<timbl> Location: http://dublincore.org/2008/01/14/dcterms.rdf#title
<timbl> Content-Type: text/html; charset=iso-8859-1
<timbl> X-Purl: 2.0; http://localhost:8080
<noah> YL: I think there's more interest here on the TAG than there, because the focus is on the interpretation.
<timbl> Expires: Thu, 01 Jan 1970 00:00:00 GMT
<timbl> Content-Length: 283
Yves: if not enough people are interested in working on that here then we can say "allow http bis to do what they want."
<timbl> __________________
<timbl> $ curl -I http://dublincore.org/2008/01/14/dcterms.rdf
<timbl> HTTP/1.1 200 OK
<timbl> Date: Thu, 02 Sep 2010 17:53:42 GMT
<timbl> Server: Apache/2.0.59 (Unix) DAV/2 mod_ssl/2.0.59 OpenSSL/0.9.8g SVN/1.4.3
<timbl> Last-Modified: Mon, 30 Jun 2008 03:54:54 GMT
<timbl> ETag: "8cd2f-133a9-38dddf80"
<timbl> Accept-Ranges: bytes
<timbl> Content-Length: 78761
<timbl> Content-Type: application/rdf+xml
<timbl> ___________________________
<noah> TBL: I'm looking up to find dcterms:title, that's what I first typed in.
Tim: The first thing - I'm looking up to see what /dc/terms/title
<noah> Tim: It's telling me that I need to look at http://dublincore.org/2008/01/14/dcterms.rdf#title
<noah> Tim: from that you get:
Tim: And it's telling me that I need to go look at that local id (#title)... and then if you look at what you get back from http://dublincore.org/2008/01/14/dcterms.rdf#title
<timbl> So then you look in what you get from that you find:
<noah> HT: You get 80K of data, only a bit of which is of interest
<timbl> <!ENTITY dctermsns 'http://purl.org/dc/terms/'>
Tim: you get - it declares a namespace -
<Yves> #title seems to be an absolute reference in the redirected content, even if that content got a # in it
<timbl> ... xmlns:dcterms="http://purl.org/dc/terms/"
<timbl> <rdf:Description rdf:about="http://purl.org/dc/terms/title">
<noah> Tim: the key bit is <rdf:Description rdf:about="http://purl.org/dc/terms/title">
Tim: the information in that document says : <rdf:Description rdf:about="http://purl.org/dc/terms/title">
<timbl> RDF isn't about parts of documents, but things in this case the concept of title
<timbl> so C has no infor about C#D
<timbl> C#D basically does not exist
Tim: RDF doesn't use
anchors.
... If it was a hypertext document you'd be looking for an
anchor...
<timbl> Theer is information about A
Tim: it didn't define a local name - it used the fully qualified URL which has purl.org in it - about "a"
<timbl> Ther is no #B in this example
Tim: [in the current state of
affairs you would be forced to write code that ignores these
fragment identifiers - which is not good]
... I've concluded from HTTP that I can use that URI to talk
about this document...
<Zakim> ht, you wanted to ask how the situation would change if the reply had been a 303
Henry: What you're saying is that there are two problems here - one is that there a 302 and the other is that there is a hash in the response. What if the first were fixed - a 303 response but still a hash.
Tim: Then I would have required C#D to identify a document - again, I am expecting a document from the 303.
<timbl> From the 302 and the 200 my code concludes that C is a document and A is a document
Tim: Yes, they could have done a 303 but then give me a document about that other document - for example a document that provides me a SPARLQL query...
Henry: But I though the point of
the http range 14 finding was that you get a document that
doesn't pretend to be what you requested...
... it is 303 that we recommend in http range 14 - yes?
<Yves> I would note that one of the option was to delegate the "fragment combination" to the mime type definition. RDF can tell its story there, like ignoring the #D part (but you know that only when you dereference the URI)
Tim: Yes where the original is a predicate.
<timbl> for teh case where the original is a thing like dublin or the concpet of a title.
Henry: We've got a URI for Dublin
- of course we don't get Dublin back, bur we get a 303 to a
document that says it's about Dublin: RDF serialized as XML. In
that document it says it's about Dublin.
... Next question - I'm not convinced that the range 14 finding
envisaged RDF that was not ONLY about Dublin.
Tim: No; there are lots of cases where people wrote an ontology in one file and they've used a slash in their URIs but all the URIs redirect to the same file...
Henry: Someone might think - "actually this document contains info on every city in Ireland then I should put a hash on it to direct me to the part of that document about Dublin"
Tim: But RDF documents don't have parts.
Henry: But RDF tells me what the semantics of # are ?
<Yves> semantic of # should be described in the application/rdf+xml type, no ?
Tim: the RDF spec says: when you
get one of these things you parse it - it tells you how to
parse it when you get ... (?)
... the tutorials show you the fragment identifier is a local
identifier in a local name space...
... there could be no other semantics to fragment IDs [than
what the RDF spec states].
... The C#D issue is up the stack a bit.
Henry: I was trying to see if based on a reasonable reading, someone in the position of the Purl people might think that putting the # on was doing the right thing. I think they did.
Tim: It may well be that if we
got back to [Purl] then they could tweak their system
accordingly.
... Anyone know the webmaster at dublin core?
... Anyone know anyone else who does this? Redirecting to a
#?
Henry: [points the finger at Yves]
<noah> ACTION-456?
<trackbot> ACTION-456 -- Yves Lafon to locate past HTTP WG discussion on Location: A#B change, and make the TAG aware of it -- due 2010-08-17 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/456
Yves: I want to continue working on this, with Tim, to understand if the issue is only with RDF documents or if it's a more general one. E.g. part of a video, part of a document, etc...
<noah> . proposed ACTION: Yves to work with Tim to propose next steps regarding redirection for secondary resources
<noah> Would close 456
Tim: We have to know what the semantics are and we have to specify it differently for hypertext and RDF.
<noah> . proposed ACTION: Yves to write draft of best practices on redirection for secondary resources (with help from Tim)
<noah> close ACTION-456
<trackbot> ACTION-456 Locate past HTTP WG discussion on Location: A#B change, and make the TAG aware of it closed
<noah> ACTION: Yves to write draft of best practices on redirection for secondary resources (with help from Tim) [recorded in http://www.w3.org/2010/09/02-tagmem-minutes.html#action04]
<trackbot> Created ACTION-462 - Write draft of best practices on redirection for secondary resources (with help from Tim) [on Yves Lafon - due 2010-09-09].
<noah> ACTION-462 due: 2010-10-05
<ht> http://www.ietf.org/rfc/rfc3870.txt is the media type registration for rdf+xml
<noah> ACTION-462: due 2010-10-05
<trackbot> ACTION-462 Write draft of best practices on redirection for secondary resources (with help from Tim) notes added
<ht> And as TimBL said, it doesn't really answer the question, but points elsewhere: "More details on RDF's treatment of fragment identifiers can be found
<ht> in the section "Fragment Identifiers" of the RDF Concepts document
<ht> [
<ht> http://www.w3.org/TR/2004/REC-rdf-concepts-20040210
<ht> ]
<timbl> tracker, help action
<noah> ACTION-458?
<trackbot> ACTION-458 -- Noah Mendelsohn to schedule discussion of followup actions for TAG to coordinate with IETF on MIME-type related activities -- due 2010-09-07 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/458
<noah> ACTION-447?
<trackbot> ACTION-447 -- Yves Lafon to coordinate TAG positions on media type related work with IETF, and to represent TAG at IETF meetings in Mastricht Due: 2010-07-20 -- due 2010-07-20 -- CLOSED
<trackbot> http://www.w3.org/2001/tag/group/track/actions/447
Noah: Yves said it didn't line up as well as we hoped - we said we would pick it up when Larry is back.
Larry: I wrote this blog post on it - if you think it's good we could pick it up as a TAG note...
Noah: if you did want to go
forward with it - anything else need to be done?
... do you view this as "step on".
... I think October is a good target for this.
... Can we take a week or 2 to re-read it.
Dan: should we bring it into W3C space?
Larry: I can supply it.
Noah: I can help to put it
up.
... I think TAG members don't want to see this work lost or
just left in Larry's blog.
Larry: Anything that came up in the June meeting?
Noah: I don't think we got to the point where we know what success is. In what further ways does the TAG want to engage?
Larry: What I would like - some of this belongs in changing the ways in which MIME types are registered. So for this to have an effect on the Web it would need to be an IEFT document.
Noah: Do you have time to help formulate [e.g.] a proposal to the IETF?
Larry: [Yes I can do that.]
+1
Noah: We will re-schedule in a week or 2.
Larry: I'm willing to turn it into an Internet Draft - I can put it in that format.
Tim: Regrets for next week.
Noah: Adjourned.
thanks!
<noah> New note on ACTION-458:
<noah> On the 2 Sept 2010 call, Larry decided that what he'd like is TAG endorsement of this as a submission to the IETF. He is willing to put it in IETF draft form. We agreed to schedule (soon) discussion of TAG endorsement for an IETF submission.