TAG telcon

24 jan 2005


See also: IRC log


Roy, Stuart, Norm, Chris, TimBL, DanC, Noah
Paul, Ian



<tim-phone> if you can from the unforgiving minute get 60 seconds worth of distance run ....

<scribe> Scribe: Chris

Agenda review

<DanC> (side note on review of agenda: this agenda is not exhaustive w.r.t. action items in the group; sigh.)

DC: IETF call

Next meeting

SKW: Regrets
... transition telcons before new TAG participants terms

TBL: No objection

SKW: VQ agreed to work o agenda for first f2f

NW: Volunteer to chair the telcon next week

RF: Volunteer to scribe next week

approve agenda

SKW: Did not note that we accepted minutes of previous meeting

<DanC> yes they do, stuart: "Minutes of 20 Dec 2004 accepted." -- http://www.w3.org/2001/tag/2005/01/10-tag-summary.html

NW: No objection

TBL: Seconded

RESOLVED; accept minutes of last meeting

<Stuart> http://lists.w3.org/Archives/Member/tag/2005Jan/0025.html

public discussion of extensibility and versioning


Noahs email

SKW: which list - schema-dev, www-tag, etc

CL: Asking ppl to subscribe to www-tag gets them a high volume list; better to go on schema-dev

DC: As long as its public, fine with me. if its more general than just schema, should be on www-tag

SKW: So, schema-specific stuff on schema-dev

<scribe> ACTION: Stuart respond to Noah citing xml-schema-dev as forum for schema specific versioning discussion

<DanC> (if anybody is seeking a shared forum where both the schema WG and the TAG are obliged to pay attention, we don't yet have one)

SKW: Joint meeting with schema 14 Feb at regular TAG telcon slot

Tech Plenary

SKW: Net outcome: A single proposed Panel session on theme of Extensibility and Versioning. Paul Downey (BT) is owning the session for TPPC.

Anticipating participation from TAG (volunteers?)and other WG's inc. XML Schema and QA-WG.

SKW: Steve Bratt said just one session
... Perhaps DO, HT, NM on panel?

<Stuart> http://www.w3.org/2005/03/02-TechPlenAgenda.html

Plenary agenda:


CL: I'm interested in Cross-Specifications Test Suites

NW: Interested in XML futures

<DanC> (I feel similarly to CL re test foo)

TAG f2f

SKW: VQ is assembling an agenda
... TAG liaisons tracking table started
... little other interest in extensibility outside of XML and schema


DC: Is this up to date and maintained?

SKW: Yes, feel free to update

<DanC> (actually, what I asked was: does the page currently know everything stuart knows, and he said yes.)

RF: when are we meeting:

SKW: Mon 9-12

NW: plan to be there, may be slightly delayed'

QA Review

<Stuart> http://lists.w3.org/Archives/Member/tag/2005Jan/0009.html

CL: my draft http://lists.w3.org/Archives/Member/tag/2005Jan/0009.html

spec is http://www.w3.org/TR/2004/WD-qaframe-spec-20041122/

<DanC> (yes, it has a pleasant style to it. plenty of whitespace, not horribly long)

<DanC> (ah... now I see why I didn't read Chris's msg; went to tag, not to www-tag; and yet it's in the technical part of our agenda. disconnect, for me.)


its not clear whether the review is public yet, since we have not agreed to it

<Zakim> DanC, you wanted to request an agendum on uri scheme registry reivew, W3C/IETF telcon 27 Jan and to

DC: seems like a fine review, wish oit was sent to them directly
... Not read carefully. Critical to fix the optional conformance bit

(discussion - who owns and umbrella spec, what if its another WG). Cross-spec conformance

<DanC> (table http://lists.w3.org/Archives/Member/tag/2005Jan/att-0009/qaframework-recursiveconformance.html)

SKW: Needs to clearly indicate which section is being discussed
... Overal l positive tone not conveyed by tesxt, add a prefix on that

<tim-phone> timbl notes character set problems with that table.

SKW: Discussion at TP on these comments? CL available

CL: Sure

SKW: Who owns this after Chris turns into a pumpkin?

TBL: Can an external person contribute, or is this a tunnelling out of alumni until their actions are all done or transferred

CL: Does not seem like too much work

TBL: precedent, we invited DO to do similar

CL: OK agreed

SKW: Splendid
... Is this suitable to send as TAG feedback?

RF: No objection

(no objections)

TBL: Abstain, did not get chance to read the comments. Support the TAG sending it

NM: Abstain too, have not reviewed

<DanC> (I think "abstain" puts a motion at risk of failing due to lack of support, while "concur" does not)

SKW: Support CL

Please send Last Call review comments on this document before that date to www-qa@w3.org, the publicly archived list

<DanC> I gather we are so RESOLVED.

<scribe> ACTION: Chris Clean up and submit

RESOLVED: These , cleaned up are TAG comments

IETF URI Registry

<DanC> http://lists.w3.org/Archives/Public/uri/

<DanC> Duplication of provisional URI namespace tokens in 2717/8-bis http://lists.w3.org/Archives/Public/uri/2005Jan/0020.html

DC: http://lists.w3.org/Archives/Public/uri/2005Jan/0051.html

<DanC> http://ietfreport.isoc.org/idref/draft-hansen-2717bis-2718bis-uri-guidelines/

DC: new process drafted, a provisionl and a final registry
... good to cite WebArch
... IRI everywhere is related to this
... if you care about this, time is running out to fix/change tings

RF: they are ready to produce another draft
... probably best to wait for the new draft

SKW; could have multiple provisional registrations for the same URI scheme?

DC: yes, but not the permanent one

TBL: (scribe missed)

SKW: Larry asked us to review new schemes.

DC: expert review of new schemes as they move to permanent registry

TBL: Who assigns it?

DC: IESG last call, then its allocated

<Stuart> SKW: Larry asked us to review and comment on revision of the URI scheme registration process.

RF: If anyone raises a non-uniqueness then it would halt the IESG review
... Next draft wil make it more clear tat the permanent registry is unique. provisional registrsations that clas with permanent als not allowed

TBL: No warning on provisional clashes?

DC: Any sane (machine readable) registry can produce uniqueness

NM: Early/late registration - late can have an inadvertent clash

DC: 27 Jan IETF/W3C telcon
... Next IETF is when??

<DanC> "6-11 Mar 2005 Minneapolis, MN?

<DanC> 62nd IETF"

DC: 6-11 March

RF: Its not a WG so no meeting then

XML Chunk Equality

SKW: Suggested posting as a note, or a finding
... TBL asked for reasons for different types of equality, when to use each one

<DanC> "ACTION: NDW to make editorial improvements, point to other different schemes, why use them, things to avoid in XML Chunk Equality."

NW: Took some actions to improove the doc in this way. no due date. Not completed yet

<DanC> -- http://www.w3.org/2001/tag/2005/01/10-tag-summary.html#item08

SKW: So, discuss more once this revision is done

NW: Due date depends on XSL/XQ specification schedule... tell you next week

<DanC> "pc: good to see when F&O deep= works and when it does not"

TBL: Equality characterized by a number of parameters?

NW: Yes, deep= has options that can be set. Namespace-related options

<Stuart> Use cases from the Issur raising:

<Stuart> Cases I am aware of:

<Stuart> - XML itself uses it for an external entity

<Stuart> - XML schema has the "Deep equality" issue as to when any two chunks

<Stuart> are "equal".

<Stuart> - RDF has a "XML Literal" data type which it handles transparently. It

<Stuart> needs a notion of when two chunks are the same.

<Stuart> - XML-DSig signs, and therefore ensures the integrity of, a chunk of XML

<DanC> (timbl, why are you surprised that RSS feeds don't have namespaces? consumers don't require them. people naturally do the minimum work that achieves their goal.)

TBL: Amazed at how much RSS has no namespace

NW: question is of unused but declared namespaces?

DC: case of two non namespaced docs, equal or not???

F(equal) -> Yes | No | dunno

<DanC> i.e. did <p> in doc1 mean what <p> in doc2 meant?

NM: (starts to say something interesting, but phone fades)

Mark Baker issue on WS-Addressing

<DanC> (the best way to provoke a response is to threaten harm, somehow; i.e. start talking about the next topic, threatining somebody's ability to comment on the previous topic)


DC: Read him to say he was happy

WS-Addressing SOAP binding & app protocols

DC: (reads from email)

<timbl> wsa:to

DC: its not a new issue

NM: SOAP will wind up putting the URI where HTTP wants it, but will also be in the SAP header too
... is it a flaw to carry the info in an additional place?

<DanC> (doesn't seem like a new issue, to me; seems like issue http://www.w3.org/2001/tag/issues.html?type=1#whenToUseGet-7 )

TBL: Arch of the WS-* specs is not yet written.
... identify an endpoint in ws, but actually send it to a different URI of the service, which has some connection, but the sever has a URI
... so its a service end point, and the service can talk about multiple objects

objects and services are distinct

scribe: another achitecture, get on the URI of a book, but behind the scenes its broken down into multiple services, checking financials and stock etc so it looks atomic but i ssplit up behind the scenes
... not clear wheter to support marks issue because its not clear what architecture it is fitting into
... good to involve DO here, finsd how WS folks tend to do this
... may be some defacto or emergent architecture
... can't say its broken unless we can point to the part that breaks

DC: Prefer to discuss whether to add this as an issue, not the summary of the eventual finding

TBL: Happy to add it to the list

NM: or work it outafter some fact finding first


RF: seems the direction of all ws specs is to be binding neutral, but no statement that a given binding is required

<DanC> endPointRefs-NN?

RF: so entirely separate architectures all described as web services
... support adding it as an issue

SKW: TP liaison with WS Addressing

<DanC> ACTION: DanC edit http://www.w3.org/2001/tag/2005/01/TechnicalPlenaryLiaisons.html to reflect avaialability and interest

NM: Suggest asking Mark Nottingham

SKW: Calls question to add as an issue

DC: endpointRefs-NN
... Aye

CL: Concurr

RF: Yes

<timbl> Aye

NW: Yes

SKW Concurr

<Stuart> concur

NM: Yes

RESOLVED: New issue endpointRefs-NN

salt NN to taste

<DanC> (tradition is to announce new issues. I'm not in a position do that)

<DanC> (easily)

<scribe> ACTION: Stuart Tell Mark Nottingham we added the isse and would like to discuss it

tag-announce and www-tag?

SKW: End of agenda

DC: Seconded :)


Summary of Action Items

[NEW] ACTION: Chris to Clean up and submit [recorded in
... http://www.w3.org/2005/01/24-tagmem-irc#T20-47-36]
[NEW] ACTION: DanC to edit
... http://www.w3.org/2001/tag/2005/01/TechnicalPlenaryLiaisons.html
... to reflect avaialability and interest [recorded in
... http://www.w3.org/2005/01/24-tagmem-irc#T21-22-04]
[NEW] ACTION: Stuart to respond to Noah citing xml-schema-dev as
... forum for schema specific versioning discussion [recorded in
... http://www.w3.org/2005/01/24-tagmem-irc#T20-19-41]
[NEW] ACTION: Stuart to Tell mark Nottingham we added the isse and
... would like to discuss it [recorded in
... http://www.w3.org/2005/01/24-tagmem-irc#T21-24-28]

Minutes formatted by David Booth's scribe.perl 1.94 (CVS log)
$Date: 2004/10/22 18:32:00 $