TAG telcon

12 Jul 2005


DanC, Vincent, Norm, EdR, DOrchard, Ht


<Norm> Finally

<DanC> DO notes record low sunlight for the month of June (or was it July?)

<DanC> for Vancouver

<scribe> Scribe: dorchard

Date: 12 July 2005

possible regrets, WS-Addressing f2f

norm: possible regrets

Chairs concern: Grid rep from HTP and possibly IBM will be present, so we should have a fairly full group of attendees

general discussion about postponing or not postponing grid discussion..

<DanC> "20-22 Sep 2005, in Edinburgh" -- http://www.w3.org/2001/tag/

Next telcon cancelled.

<DanC> next telcon: 23 Aug 2005

Next telcon: Aug 23

discussion about minutes from last week...

minutes approved with changes

<DanC> ed regrets in http://www.w3.org/2005/07/05-tagmem-minutes.html $Date: 2005/09/01 07:36:08 $


does namespace name identify information resource or not?

<DanC> (er... of course... who asked otherwise? oh.)

norm: should talk about what ns name does

dan: some ns names have # in them

<scribe> ACTION: norm to follow up on noah's message on ns name

<DanC> looks good... http://www.w3.org/2001/tag/2005/06/23-rddl/

<Norm> I think some folks would argue that namespace names aren't information resources and they must therefor use the 302 response code work-around that we discussed. In that case, it's a bad idea to put any document at the namespace name.

<DanC> examples e.g. http://www.w3.org/2001/tag/2005/06/23-rddl/rddl1.xml

<DanC> I think I'll review those.

web services architecture

<DanC> (ed, did you make much progress on reviewing the ws-addressing docs?)

<DanC> (hmm... maybe ws endpoints should use id@domain syntax rather than scheme:path URI syntax.)

<DanC> DO explains how message passing is orthogonal to having representations of resources available (e.g. by HTTP GET)

ed: worked on xri rather than ws-addressing
... why does ws need state

<DanC> (is the "my yahoo" pattern on our issues list yet? people keep asking me about it and I keep saying "yes, the TAG should say something about that")

state is on the web, not just web services. http cookies are almost always used for stateful identifiers.

<Norm> Hello ht

dave: the vast majority of web services are not on the web. Same with almost all uses of web sites with http cookies.

<DanC> (except for formal decisions and actions, anything the scribe can capture is gravy, imo)

Henry: somebody should write up discussion about "state".

<ht> So the questions of context, statefulness, etc. have been raised a number of times

<DanC> cookies are no different from other forms of authentication, architecturally.

<ht> Surely there's something in the issues list already about cookies vs. the 'back' button???

<DanC> actually, I just looked, and no, there isn't, ht. I agree there should be.

I volunteer to write up something about state in distributed application design

<DanC> ("state" is an overloaded word... GET is supposed to GET you the state of a resource, too, but that's an orthogonal meaning of "state")

<scribe> ACTION: David to write material on state in distributed application design

<DanC> 28 Jun minutes?

<DanC> looks ok to me. $Date: 01-09-2005 - 09:36$

<ht> I confess that I always have to stop and think before I can explain that interaction, so I sure hope a nice Dirk and Nadia story about that can be part of what Dave is writing . . .

<DanC> (re namespaceState , xml:id is a PR today)

<DanC> NDW to work with HT, DO on namespaceState continues

Norm: work on a write up on namespace-48

rdf in xhtml

<DanC> http://www.w3.org/2001/tag/issues.html?type=1#standardizedFieldValues-51

<DanC> ht's mention of xpointer registry reminds me of uriMediaType-9: Why does the Web use mime types and not URIs? http://www.w3.org/2001/tag/issues.html?type=1#uriMediaType-9

henry: Within a week, AC will be asked to approve opening up of xpointer scheme namespace space by an online registry.

<Norm> I have qualms about the registry too

<DanC> (persistent identifier workshop... any public records, ht?)

<ht> my talk: http://www.dcc.ac.uk/docs/persistence.html

<DanC> dan's action on 25 is withdrawn

Dan's action is withdrawn.

<DanC> action on RDFinXHTML-35 , rather

<DanC> my action is subsumed by what we did in Cambridge. http://www.w3.org/2001/tag/2005/06/14-16-minutes.html#item021

making progress on schema use in web services action item

Summary of Action Items

[NEW] ACTION: David to write material on state in distributed application design
[NEW] ACTION: norm to follow up on noah's message on ns name
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.115 (CVS log)
$Date: 2005/09/01 07:36:08 $