See also: IRC log
<scribe> Scribe: Roy Fielding
<scribe> ScribeNick: Roy
Regrets for Sep 13: Noah, Henry, TimBL
Plan for next week is discussion of GRID. Experts have been invited but we have not heard back yet.
VQ: should we still schedule discussion for next week if we only have one person from GRID?
NM: propose we take one round of discussion to get oriented and then perhaps take it further during the f2f
NW: will find out about other experts and send info to Vincent
<DanC> 30 Aug minutes
RESOLUTION: Minutes of 30 Aug meeting approved
VQ: I have made a pass through the issues list to prepare for a f2f agenda
<DanC> ah... good... http://www.w3.org/2001/tag/issues.html?type=1#httpRange-14 is now marked "agreed"
VQ: only a rough draft has been prepared so
... I would like to make progress on several of the open issues, try to close them, and make progress where possible
... I would also like to see some setting of priorities for the TAG
NW: that would be good
VQ: Henry, are there constraints on the schedule at F2F?
HT: No, we can go til midnight is desired.
... dinner on Tuesday will probably be at 7:30 pm, time not yet confirmed
<DanC> (let's see... europe is that funny place where I wake up at 3 or 4am, right? so I'm happy to start earlier than 9am... much earlier... and by 5pm, you'll be lucky to communicate with my conscious self.)
HT: do people have a preference for organizing a group meal on Wed or Thurs?
all: Wed is better
VQ: anything else about the logistics?
<dorchard> this is all online right Henry?
HT: location is three-story row house with lots of doors. I will stand out on the sidewalk from 8:30 to 9 or so to let people in. If you are late, enter door 2, go upstairs one level, and enter through main entrance [Henry will supply details]
<ht> ... ring bell marked "HCRC", secretary will let you in and take you to the meeting
<ht> or call my cell phone [number elided]
<DanC> it seems to be covered at http://www.w3.org/XML/Group/2005/09/edinburgh-meetings-info.html#location
NW: first entry is from Chris, suggests that we
add an editorial correction and that looks fine to me
... anyone disagree (silence)
<DanC> "Interpretation of prose in 3.3"
<DanC> ^that's a pretty good name, for these minutes
RESOLUTION: TAG agrees to CL errata 1, Interpretation of prose in 3.3
Second errata is from Roy to update references to URI and IRI
RESOLUTION: TAG agrees to RF errata for URI and IRI references
<scribe> ACTION: NW to include items 1 and 2 in webarch errata [recorded in http://www.w3.org/2005/09/06-tagmem-irc]
HT: document is not ready yet and unlikely to be able to spend more time on it before f2f
VQ: should we spend time on it during f2f?
HT: probably not
DC: this topic doesn't seem to be urgent at this point
<ht> Noah is talking about namespaceDocument-8
NM: there is concern about use of namespace names and http, may be worth f2f time
VQ: that is on the f2f agenda under namespaceDocument-8
DC: IETF stated practice currently is to use
URNs for namespace names, for example in geopriv (geo privacy) specs.
... IETF using XML in various ways and recommending use of URNs, so there is some need to recommend dereferenceable identifiers instead
VQ: we have not done work on this since it was announced. does it make sense to discuss it at f2f?
DC: people working on protocols/formats tend to start with short names and then later encounter name collision
Roy: I agree with Dan's summary of the issue.
... I suggest that DC take an action to write something up ;-)
DC: I could do that
NM: that would be valuable to me to get a handle on the issue
<DanC> ACTION: DanC to write an update on issue 51 and microformats [recorded in http://www.w3.org/2005/09/06-tagmem-irc]
... any progress?
NW: no progress to report, we need to figure out who has the ball
<Zakim> DanC, you wanted to ask who has the ball on c14n
<DanC> HT: XML Core
NW: I will try to write something up for f2f
<scribe> ACTION: NW to provide something written for namespaceState for f2f [recorded in http://www.w3.org/2005/09/06-tagmem-irc]
<noah> David Orchard to contextualize his scenarios, such as more on what is happening with SOAP and WSDL. [recorded in Minutes of the W3C XML Schema Working Group 4th (37th) F2F meeting, action TP5-8]
that document is not public
DO: I have made significant progress on that action, should be resolved within a week or so
VQ: let us know when we should review something
in the TAG
... any more items to discuss?
<DanC> I just sent mail about namespaceDocument-8, if we want to add that today's agenda
<DanC> ns8 stuff
DC: last week we talked about ns8...
afterwords, I remembered that if you take the whole URI of a name within a
schema, remove the fragment, and dereference (like a machine), what comes
back as RDDL does not have an id attribute that can be used to ground
fragment ids as schema anchors
... straightforward solution is to place anchor ids in the namespace document such that they are identifiable in the dereferenced resource
NM: I had assumed that people would want to use RDDL as a form of indirection. Perhaps if RDDL had its own media type then it could define its own mechanism for indirect definition of anchors
HT: with a namespace described by multiple schema documents, one must acknowledge that there will be different names defined by different schema documents, and therefore the namespace document needs to act as a union of names in a complex namespace [scribe interpretation]
<ht> The relevant Bugzilla entry for the Schema WG is 1974, so URI is http://www.w3.org/Bugs/Public/show_bug.cgi?id=1974
<DanC> ACTION: HT to track progress of #int in the XML Schema namespace document in the XML Schema WG [recorded in http://www.w3.org/2005/09/06-tagmem-irc]
HT: in settling httpRange-14, we talked a great deal about regular URIs but not as much about URIs with fragment
NM: when we introduced "information resource", our definition distinguished between obvious things but left unclear whether a namespace falls under the category of "information resource" or not
DC: if it returns 200, then you have answered
... whether a namespace resource is an information resource or not is up to the namespace creator.
NM: so, summary would be that it is okay for a namespace name to be an information resource provided that the URI is used consistently