W3C | TAG | Previous: 20 Jan teleconf | Next: 3 Feb 2003 teleconf

Minutes of 27 Jan 2003 TAG teleconference

Nearby: IRC log | Teleconference details issues list www-tag archive

1. Administrative (20min)

  1. Roll call: SW (Chair), DC, PC, RF, CL, TB, DO, IJ. Regrets: TBL, NW
  2. Accepted 20 Jan minutes
  3. Accepted this agenda
  4. Next meeting: 3 Feb (short)

1.1 Meeting planning

1.2 FTF meeting agenda

1.3 TAG at tech plenary


DO: Tech plenary planning committee has met.
"short" and "5 segments" don't both fit in my brain at the same time
CL: Good idea to pick top three issues and get input. XML id, XML subsetting good.
DO: We'll use BOF to talk about linking.
[Discussion of fact that xml id and xml subsetting orthogonal.]
PC: Looking to have 90 minutes with non-TAG moderator. The message is that people want more technical content.
DO: I was hoping to have no process discussion; we'll have a short discussion.
I'd be happy letting any process discussion come up as Q&A.
I think 10 mins is an acceptable compromise but lets hold it to that and not let it spread into the technical part
DanCon, you wanted to ask about discussion mode/logistics (panel? presentations? other?)
PC: I think Steve Zilles should moderate.
DO, CL: Good idea.
Confirmed to attend and participate: CL, DC, PC, DO
LIkely regrets: TB, RF, SW
PC: NW should be there since he'll be there that week for XSL meetings.

[Defer discussion to ftf meeting]

2. Technical

2.1 httpRange-14

  1. httpRange-14:
    1. Request to establish the relationship between URIs and Resources is many to many; from Bill de Hora.


DC: I feel no obligation to discuss this in previous meetings.
what Dan said
CL: BDH's question has not been ignored on the list.
TB: In light of further traffic on www-tag (and xml-dev in parallel), my opinion has gotten stronger that our chances for consensus on this point are small. I agree that TBL's world view is consistent; I just don't believe it.
IJ: I think this issue may be blocking us in practice, even if we have said in the past it's not critical path.
DO: I don't know what new information there has been to cause us to reopen this.
I agree on the 'what text needs to change' criterion
RF: I've spent a lot of time drafting messages; they had nothing to do with httpRange-14 except in the periphery. The purpose of those messages was to establish what we should put in the arch doc. If TBL wants to describe the Web as everything the W3C is doing, we have to write another document. We need to write a model that describes it all. The REST model does not describe the SemWeb architecture.: The reason we have to make a choice: I can describe the system that exists without talking about anything in the future (by looking at implementations). Whether we want to talk about the future will affect our description.
DO: There is similar discussion going on in Web Services area.
but if there are no implementations in that area, isn't it premature to try and abstract out their properties?
[Agreement to talk about "What should go in the arch doc" at the ftf meeting.]
PC: I think it's important that at our ftf, we talk about www-tag and perhaps a need to organize threads.
SW: We shan't discuss httpRange-14 today; will not be on agenda any time soon.

2.2 Architecture document

  1. 6 Dec 2002 Editor's Draft of Arch Doc:
    1. Next TR page draft? IJ proposes after ftf meeting.
    2. Action CL 2002/09/25: Redraft section 3 based on resolutions of 18 Nov 2002 ftf meeting.
    3. Complete review of TBs proposed principles CP9, CP10 and CP11
CL: I've started putting together a chapter three. I will have something shortly before the ftf meeting. I'll try to have enough in advance to read on plane.

The TAG reviewed proposed principles CP9, CP10, and CP11 raised at November 2002 ftf meeting

CP9. Designers of protocols should invest time in understanding the REST paradigm and consider the role to which its principles should guide their design:
- statelessness
- clear assignment of roles to parties
- flat addrss space
- limited uniform set of verbs
yes, agreed
DC: Are these the top four things for REST?
RF: Say "Uniform address space" instead of "flat address space."
DC: I have no objection to this, but would rather explain how to do a Web page / how not to. I'd be willing to write something up.
ACTION DC write two pages on correct and incorrect application of REST to an actual web page design
(I was talking about a finding... webarch editor to steal as much/little text as he likes)
or a web service
agree with Roy s/flat/uniform/
note to self: OpenGIS mapping service is a *great* example of a straightforward "uniform address space" service
DO: I'm writing something in Web Services area that's related.
DO: I'd have to publish this info....
DO: My goal is different - someone who believes in REST can say "this is how I would build a REST-based Web service" and someone more interested in RPC could do things according to that style.
ACTION DO: Please send writings to tag@w3.org. DO grants DC license to cut and paste and put into DC writing.
TB: Per our charter, please send info to www-tag.
RF: One reason is that we are talking about WS Arch....
TB: Ok, but I'll keep pushing for www-tag.
Resolved: Accept CP9 with s/flat/uniform.
CP10. Agents which receive a resource representation accompanied by an Internet Media Type MUST interpret the representation according to the
semantics of that Media Type and other header information. Servers which generate representations MUST not generate Media Types and other header information (for example charsets) unless there is certainty that the headers are correct.
CL: +1 to add this as is
IJ: See related language in Internet Media Type registration, consistency of use:
"The architecture of the Web depends on applications making dispatching and security decisions for resources based on their Internet Media Types and other MIME headers. It is a serious error for the response body to be inconsistent with the assertions made about it by the MIME headers. Web software SHOULD NOT attempt to recover from such errors by guessing, but SHOULD report the error to the user to allow intelligent corrective action."
CL: I like the more general second sentence of CP10 (compared to findnig). I can elaborate this; and give examples.
I can elaborate this no problem
DC: I prefer what we have in the finding. wget and link checkers don't pay attention to the mime type, and they're still correct.
they are not, arguably correct
DanCon, you wanted to wonder about clients like WGET that totally ignore MIME types, in a way that's good.
DanC: ... and so the way it's phrased here in CP10 http://lists.w3.org/Archives/Public/www-tag/2002Nov/0107.html doesn't appeal to me.
IJ: I think there is an ok behavior where the tool says "I'm doing something different." This is different than a user agent doing something and hiding it's incorrect behavior.
TB: I support moving finding info into arch doc. I think we need to say more about "Servers which generate representations MUST not generate Media Types and other header information (for example charsets) unless there is certainty that the headers are correct." in arch doc.
RF: One slight problem - because of some charset interpreting errors in some browsers, often the charset is explicitly given in the doc itself to prevent a cross-site scripting security problem.
TB: If you are sending well-formed XML, it knows what charset it's in. It's almost always wrong to say what the charset is (and you could get it wrong)
I understand this and can write it in the arch doc
TB: Don't serve as text/xml; transcoders can possibly get it wrong.
ACTION CL: Draft language for arch doc that takes language from internet media type registration, propose for arch doc, include sentiment of TB's second sentence from CP10.
CP11. Designers of protocols should give serious consideration to avoiding such design activities in favor of existing well-established
protocols such as HTTP that fit well into REST.
TB: Another way to say this is : Avoid designing new protocols if you can accomplish what you want with HTTP
DC: I think that TBL has an old action to build a table saying: what patterns would imply appropriate protocols.
TB, IJ: We withdrew that action.
TB: This is a statement worth making; people should think about using HTTP before running off to design a new protocol.
DO: What are the criteria for using/not using?
TB: That's a larger question that may be worth working on.
DC: "In the Web context" is vague to me. The IETF best practice on record is "Don't use HTTP except for hypertext.": Handwaving won't help us here.
TB: I have not understood why WebDev was needed as another protocol.
RF: It's written as an extension to HTTP.
DC: It runs on port 80.
DC: My examples are POP and SMTP.
Action TB: Develop CP11 more.
DC: I suggest describing GET/PUT/POST in a para each, then say "if your app looks like that, use HTTP".
TB: If your protocol needs a notion of a temporally extended session, then HTTP won't help you.
DO rhetorically: Why would you need one of those? You'll need to include some examples.

2.3 IRIEverywhere-27

  1. IRIEverywhere-27
    1. Action MD and CL 2002/11/18: Write up text about IRIEverywhere-27 for spec writers to include in their spec.
    2. Action CL 2002/11/18: Write up finding for IRIEverywhere-27 (from TB and TBL, a/b/c), to include MD's text. Process with IJ; awaiting comments from MD


CL: IJ and I discussed this last week. We drew up some text.: Text based on input from Martin Duerst.
DC: Value of treating e and E as equivalent is a huge cost. What actual value do you get?
CL: Actual value is that both are equivalent to the same character.
TB: Lots of software is already treating %7e and %7E as the same.
that they are equivalent to the *actual character* represented
uri spec is way fuzzy on this
actual practice is that they are the same
TB: Per 2396, I think Web robots are in their rights; lots of Web robots do this.
RF: Yes, that's my understanding.
PC: Chris, could you explain impact?
I think it's straightforward to read the URI spec as saying that http://a/%7E and http://a/%7e are distinct URIs, and may or may not refer to the same resource.
roy, you think otherwise?
I think otherwise
I read it the same as Dan :-(
CL: There is a bigger effect on IRI spec and suggestions for RFC2396.
sigh; each of those URIs is a sequence of 12 characters. they differ in their 12th character. hence they're different URIs. RFC2396 says otherwise?
this has more effect on IRI comparison (which is done by transformation to URI and then comparing)
Action CL: Please propose text IJ and CL worked on to www-tag (flipping the ACL).
%7e is one character -- three octets
it means that the *actual kanji* and the sequence of hexifyied octets compare to the same
which helps in roundtripping a very great deal
oops one octet -- char[3]
"%7e" is *one* character???
"character" is defined in spec
was ignoring IRC... yes, lots of software will decide those two URIs are the same in their cache
no, %7e is one octet
hence *sequence of *
DanCon, you wanted to suggest the value of having %7E specified to be equivalent to %7e is purely aesthetic, and not *nearly* worth the cost.

2.4 binaryXML-30

  1. binaryXML-30
    1. Action CL 2002/12/02: Write up problem statement about binary XML; send to www-tag. See CL email to tag
TB: I agree that a general public service announcement along the lines of what CL has done would be beneficial.
yes, chris's msg is the first 80% of a finding, IMO.
TB: I propose that we close the issue; no substantive proposal at this time. TB summarizing CL mail: The only out there that seems to be making a serious reach at being a broad-spectrum solution is the BIM thing.
CL: I'd like to post this so that people can say "You aren't aware of this other thing...."
DC: I'd like a week to think about the proposal.
so, getting public review would flush out some alternative solutions we have missed
DanCon, you wanted to ask for a week or so to consider Bray's proposal
Action CL: Send email 0067 to www-tag for feedback.
TB: I will be astounded if a proposal comes forward that really does meet a broad spectrum of needs. Some requirements are fairly contradictory with each other.
SW: We'll likely discuss this at the ftf meeting.

2.5 xmlProfiles-29

  1. xmlProfiles-29
    1. See email from Chris on options for ID
    2. See email from NW (TAG-only) on ID attributes.
    3. See comments from Paul Grosso to treat xml:id as separate spec.
    4. See NW summary.
TB: Was there much disagreement about this?
DO: My pushback was attribution; I don't think there's consensus in the TAG yet. NW wrote his note as NW's position, not TAG position.
DC: I expect that NW will integrate feedback on his position. At the last meeting, we agreed to comment on NW's text.
TB: I don't think we have a TAG position without further discussion of NW's draft.
Orchard hasn't responded in substance, has he? he only objected on process grounds. i.e. he hasn't sent his technical position/argument, did he?
DO: I think that there are at least 3 or 4 people who could live with xml:id.
TB: I thought we were trying to write a note for the AC on a way to proceed. My sense is that we pretty much agree with NW's draft except for the xml:id part.
PC: I thought I saw public pushback on having profiles at all.
See also: http://lists.w3.org/Archives/Public/www-tag/2003Jan/0217.html
TB: Mostly I see questions about SOAP and PIs. We could change "I feel strongly that it would be a mistake to introduce a single new feature, or a single change of any sort that would not be completely compatible with XML 1.1, in the work that subsets XML."
"The question remains open..."
DC: Seconded.

TB summarizes NW's conclusions.

CL: That would mean that you cannot have xml:id; that's not in xml 1.0.
TB: 1.1 processors would not know what it means, but wouldn't have any problems with it.
SW: I'd like NW to concur with this.
DC: I'm happy for him to do that after the fact.
TB: My proposal is to ack that NW feels this way.
DO: One possibility is to ask the AC what they think; another is to hammer this out further.
TB: I think it's cost effective for us to tell the world that we think that there should be an xml 1.1.
DC: WRiting to the AC is logistically awkward. I recommend we write to XML Activity lead and recommend that it go to the Director.
DanCon, you wanted to ask that the chair put the question, including who we're writing to. Writing to the XML Activity lead, suggesting he take it to the AC, would be my preference
Action IJ: Change one sentence, sent to XML Activity Lead, cc www-tag.
DO: I have some concerns about this; seems like we're pulling a fast one. I think we could do with more discussion about this. Perhaps we could do a straw poll on xml:id.
TB: Why don't we accept a new issue on xml:id and get the ball rolling.
PC/TB/DO agree.
2nded, to accept an issue on xml:id (in case chair is counting toward majority in favor)
DO: If we treat xml:id as a new issue, then ok to send out.
Action DO: Raise new issue about xml:id (separate from xmlProfiles-29).
DO: I will raise issue by tomorrow.
TB PROPOSED: Close this issue with the sending of this letter.
Resolved: Yes.

Postponed: Findings in progress

See also: findings.

  1. Findings in progress:
    1. deepLinking-25
      1. Action TB 2002/09/09: Revise "Deep Linking" in light of 9 Sep minutes.
    2. URIEquivalence-15
      1. TB's "How to Compare Uniform Resource Identifiers" draft finding.
      2. Action TBL 2003/01/20: Send email to uri@w3.org requesting terminology change (regarding definition of "URI").
      3. Action TB: Before Feb 2003, produce a new draft of How to Compare Uniform Resource Identifiers that incorporates comments from DC and TAG.

2.2 Priority issues

  1. namespaceDocument-8
    1. Action PC, TB 2003/01/13: Write up a Working Draft that recommends a data format for namespace docs (not compulsory) and that such a document should follow the Rec track process. The initial content of the document should be taken from the RDDL challenge proposals; they are isomorphic in tecnical content. Please include drawbacks in the draft.
    2. Please read NW summary of the following proposals:
      1. RDDL Proposal from Tim Bray.
      2. RDDL Proposal from Chris Wilper
      3. RDDL Proposal from TBL
      4. RDDL Proposal from Jonathan Borden
      5. RDDL Proposal from Micah Dubinko
      6. RDDL Proposal from Sandro Hawke
      7. See also proposal from Garrett Wilson
  2. fragmentInXML-28 : Use of fragment identifiers in XML.
    1. Connection to content negotiation?
    2. Connection to opacity of URIs?

Ian Jacobs for Stuart Williams and TimBL
Last modified: $Date: 2006/12/08 10:16:47 $