12 April 2005 Tag Teleconference

Present

DanC, Ht, Norm, EdR, VincentQ, DOrchard, TimBL

Regrets

Noah

Chair

VincentQ

Scribe

EdR

<ht> scribe: EdR

<DanC> DO volunteers to scribe 3 weeks hence

<scribe> Chair: Vincent Quint

<scribe> Scribe: EdR

<scribe> Meeting: 12 April TAG teleconference

Agenda

Agenda accepted

David possible regrets for next week.

Henry offers regrets for next week.

<DanC> (I'm here on the 19th but at risk due to travel for the 26th)

Roy to scribe next week.

Tim offers regrets for next week.

Minutes from last weeks meeting.

RESOLUTION: Minutes are approved as they stand

<DanC> fixed title. http://www.w3.org/2005/04/05-tagmem-minutes.html $Revision: 1.4 $

Action items;

Henry and Ed to continue action item for posting of meeting minutes.

<DanC> (found it.

<Norm> You broke up for a moment, DanC, what did you have to implement?

<Norm> ah, resolve-uri

DanC - action closed on document review, limited response in email so issue

DanC Document XSLT/XQuery/Xpath documents review complete

Timbl it would be cleaner to have seperate modes, but not the level the TAG

<timbl> One converts a URI (loosely construed) into a URI (strict charset) with false

<DanC> mkPathSegment(string) -> uri-path-segment

<Norm> The QT WGs have struggled more than once with the fact that anyURI isn't a

<DanC> anyURItoURI(anyURI) -> URI

<timbl> With true, it converts any string into a string constrained to the path

<DanC> (yes, not making anyURI a subtype of string is a mistake I convinced the XML

<Norm> I'm not the source of the echo

<DanC> anyURItoURIref(anyURI) -> URIref

<timbl> (Yes, that makes sense)

<timbl> URIencodeString

<timbl> or encodeStringForURI

<DanC> ACTION: DanC to draft comment about splitting fn:escape-uri into separate [recorded in http://www.w3.org/2005/04/12-tagmem-minutes.html#action01]

<ht> quoteURI (== arg2==true), fullURI (== arg1==false)

<timbl> cleanURI

<DanC> timbl: "canonicalize" has the right connotations w.r.t. being idempotent, but

XRI Document review

<timbl> It was very difficult to exp[lain to people in the old days why one might vary

<ht> HST notes that "path element" doesn't immediately help me, and google didn't

Henry has prepaired a first review

Henry: Henry took David Orchard's document and folded in with Henry's document

<Zakim> DanC, you wanted to say it seems headed in a useful direction

Ed Raises question of procedure on Oasis document

feedback due to Oasis by end of the month.

<DanC> (Ed, pls copy [email protected])

Ed to send Henry feedback to see if it makes sense to fold into the document

Dan: In general re-use URI schemes where appropriate. If your doing an

<ht> DanC: Admin Hierarchy (via DNS) plus structured thereafter (path) is done,

<Zakim> dorchard, you wanted to channel Dan on domain/path strings.

We will iterate our position on [email protected], Vincent will send more formal

Binary XML documents

<Zakim> DanC, you wanted to ask for a summary of the argument in the characterization

<timbl> http://www.w3.org/TR/xbc-properties/

<timbl> 4.3.2 Description

<Norm> Audio gone?

<Norm> #$@$##@

<timbl> http://www.w3.org/TR/xbc-characterization/

<DanC> Subject: Dave Orchard individual review of XBC for the TAG

<DanC> Date Mon, 11 Apr 2005 12:35:40 -0700 (14:35 CDT)

<timbl> http://www.w3.org/mid/[email protected]

<DanC> indeed, the { xml generality no} cell is hard to swallow

<timbl> "There is a consensus in the Working Group, based on significant collective

<DanC> hmm... "Evaluation of existing approaches has shown that there is at least one

<timbl> 3 or 5

<timbl> have checks in the boxes

<DanC> (swapping in http://www.w3.org/2001/tag/issues.html?type=1#binaryXML-30 )

<Zakim> ht, you wanted to ask what the _Web Architecture_ issue(s) are here

<timbl> qq= to ask whether the bullets after the tabl elabel the columns

<DanC> (swapping in 4.1. Binary and Textual Data Formats

<DanC> (wherein I find "quantitative studies are essential to a correct understanding

<Zakim> timbl, you wanted to ask whether the bullets after the tabl elabel the columns

<DanC> note # why obscure Feasibility section? Dan Connolly (Thursday, 7 April)

<Zakim> DanC, you wanted to (a) propose that

<timbl> "A compact encoding can be achieved in different ways: a number of different

<timbl> A compact encoding can be achieved in different ways: a number of different

<timbl> 5.2.2

<dorchard> I don't mind spending many telcons/time on this. I think this is very

I think there IS a case where binary xml format could be justified (or binary

Vincent: Close discussion for today... running out of time.

David: This issue needs more time, there are a wide varieties of opinions and

Dave put several options on the table which were not discussed yet today.

<scribe> ACTION: Ed to take Dave Orchards list and add his information as a working [recorded in http://www.w3.org/2005/04/12-tagmem-minutes.html#action02]

Dan's new issue - Squatting on link relationship names, x-tokens,

deferred to next week

<DanC> (so scribe fore next week is Roy, provided he confirms, else Norm, right?)

<Vincent> DanC: right

I'll edit the minutes and get them out today.

<DanC> great.