See also: IRC log
<trackbot> Date: 02 June 2009
<mgylling> [hmm, got disconnected]
In any case he is not yet online
though I spoke to him an hour or so ago
<mgylling> ... and heres another +1 for relaxng
Well, it's good to have some Ralxing expertise around
Masayasu used to be our relax expert
<mgylling> I know relaxng quite well too (consider it my second tongue after swedish)
<MoZ> what happened ?
We're dropping the call until Shane arrives
staying in the IRC channel
<MoZ> ok please ping me when you're back so
<Tina> ShaneM: Steven is on a school run, I believe; I'll call in when lunch is done.
<alessio> hi all, wasn't it at 13.00 UTC?
<ShaneM> Steven is tied up for a little while. roland, what's your plan
<alessio> ok... thx shane
<Tina> ... my phone is broken, again. I'm IRC only, folks.
<ShaneM> hey steven. some of us are here - just waiting on you and roland.
Given the number of issues we need to review in order to resubmit the PERs, I request that we schedule another 4 hour vF2F meeting. I suggest we try to do it Tuesday, 2 June at noon UTC: http://www.timeanddate.com/worldclock/meetingdetails.html?year=2009&month=6&day=2&hour=12&min=0&sec=0&p1=136&p2=179&p3=215
We are starting, MoZ, Alessio, Tina, mgylling
<scribe> Scribe: Steven
Shane: I reduced the list to 11
... We received a formal objection from Ian Hickson
Steven: About text/html?
Shane: Yes, he doesn't like it
Steven: I believe that doesn't have to be a show-stopper
Shane: And there is Bjoern's mail, that wasn't sent to us
<Tina> ShaneM: do we have a link to the formal objection?
Steven: I haven't checked it
rigorously, but I believe that points to issues he had already
... but we should make sure
... that all his issues made it into the issue tracker
... So we have 11 issues that we should look at
Shane: I replied to this ages
... anyway appx C is gone
Steven: Who, quite honestly is going to misunderstand this?
Shane: We replied, he didn't answer, what do we do?
Steven: we have followed process. Issue closed
Shane: Marked as closed,
... we advise against the XML declaration
... and say use UTF 8 or a higher-level protocol
... he asks a question, Steven replied, he didn't respond. Therefore we have met process.
Steven: Issue closed
Shane: This is an error in the
... The DTD has been fixed
... is it OK for a PER?
Steven: Yes, it doesn't break any browsers, it was a clear error
Shane: there is an updated editors' draft
<ShaneM> Updated references section is at http://www.w3.org/MarkUp/2009/ED-xhtml1-20090528/#refs
Shane: I have split the
references into informative and normative
... need an OK from you
... POSIX is for MAY, SHOULD etc
... there was a formal objection because the objector wanted us to point to 5th edition of XML
... but we decided against that because of the change to name token
Steven: I remember having that
... isn't 5th edition just v1.1 back ported to 1.0?
Steven: We deliberately made this
decision, so I think we should defend it at the transition
... I mean, the real thing we are trying to achieve here is adding schemas
Shane: Anyone object to pointing to 4th edition?
Shane: we refer to 2e of
namespaces, an obvious improvement
... Are there any of the informative refs things that have to be normative? (I will update the ref to M12N)
... some of them aren't even referenced
Steven: Chop 'em out then.
... they may be leftovers from appx C
Shane: Fixing them now
... three left, MIME, MOD, and MathML
Steven: Looks good
... closes this issue
Shane: I will reply
Shane: Tobias asked us to fix it
... but Jim Ley brings it to XHTML 1
<ShaneM> I said "objects are independent entities within the page. Navigation is local to the
<ShaneM> object and its own processing. This is defined in HTML 4."
Shane: but I claim it is an HTML
... where the question is already answered
Steven: Good answer
<MoZ> Moz : I'm with 4th edition. Since Namespace with XML is still not in sync with 5th edition
Steven: Thanks MoZ, that is good
Steven: lang and xml:lang are
normatively defined elsewhere; I think it would be wrong to
normatively define anything here
... if you use both, with different values, well, that's your own silly fault
Markus: Are the datatypes the same?
Shane: I think one is a subset of
... so it is possible to have a value of xml:lang that is not a legal value of @lang
... (in theory)
Steven: Didn't we want both lang and xml:lang for the same reason as name/id
Shane: There is a difference,
largely because of accessibility software
... we were trying to be good XML citizens by including xml:lang
<MoZ> You have no choice to use xml:lang. It comes with XML 1.0 (as well as xml:space)
Steven: If we change the wording on this now, we would be leaving PER territory
<ShaneM> Good point. In XHTML 1.1 we say "This specification also adds the lang attribute to the I18N attribute collection as defined in [XHTMLMOD]. The lang attribute is defined in [HTML4]. When this attribute and the xml:lang are specified on the same element, the xml:lang takes precedence. When both lang and xml:lang are specified on the same element, they SHOULD have the same value."
Steven: it could invalidate
... In XHTML 1.1, we are expanding the number of allowable documents, and not invalidating existing UAs
Shane: Section c.7 has been removed, so the comment doesn't apply
Steven: It does, because he
wanted something added to the normative requirements
... which is out of scope, because these attributes are defined elsewhere
... and I will argue that at the transition call
5 mins break for coffee
Shane: WHat do we do; I have kept the anchors with a pointer to the place where they now are
Roland: That's the best we can do
Shane: He's written a validator against appx C, and that broke the error messages in his validator
<Tina> Who wrote a 'validator' ?
Steven: This isn't the first time this has happened with a W3C spec (anchors changing)
Shane: Well, they don't get a 404
<ShaneM> sorry - just a minute
Steven: We defer to XML on the
definition of whitespace
... and even point that out in 4.7
Steven: I still think that XML
definition of whitespace on input, and CSS definition on output
is the good answer
... and not add anything else. The spec at 4.7 says exactly what it should do
Shane: Issue closed
... we are rejecting his request, no edits are associated with this change
Shane: Nothing to do with this
... Do we want to add a schema?
Steven: Let's not give ourselves any more work
Shane: OK, this comment is not associated with XHTL 1.0 anymore
Shane: I think he was misinterpreting this section
<ShaneM> http://www.w3.org/MarkUp/2009/ED-xhtml1-20090528/#normative item 3
Shane: It means "@name does not create a fragment identifier"
Steven: Looking at the wording of his comment, he is asking for changes wrt to appx C, which is gone. So I think the issue is closed
Shane: I have made the
... they make no changes to an UA
... and anyway, all the datatypes map to CDATA
Steven: So this is just
... names of datatypes that are otherwise equivalent
Steven: So we just take out the second part of the sentence?
Shane: Yes, get rid of duplication
Steven: I can live with that
Shane: I will make that
... It is in an informative section anyway
<Tina> Which list is 6332 on?
<ShaneM> 6232 is in the voyager issues thing - Steven is getting a link now
<ShaneM> Ian's objection is here: http://lists.w3.org/Archives/Public/www-html-editor/2009AprJun/0020.html
Steven: What can I say except
that I disagree?
... ALl we do is give guidelines on getting documents to render in HTML UAs
<ShaneM> 5.1 in third edition current reads: XHTML Documents which follow the guidelines set forth in [XHTMLMIME] may be labeled with the Internet Media Type "text/html" [RFC2854], as they are compatible with most HTML browsers. Those documents, and any other document conforming to this specification, may also be labeled with the Internet Media Type "application/xhtml+xml" as defined in [RFC3236]. For further information on using media types with XHTML, see the informativ
<ShaneM> For information on delivering XHTML 1.0 Documents to user agents that do not natively handle this media type, see [XHTMLMIME] .
<ShaneM> So the final text for 5.1 would read: XHTML 1.0 Documents should be be labeled with the Internet Media Type "application/xhtml+xml" as defined in [RFC3236]. For information on delivering XHTML 1.0 Documents to user agents that do not natively handle this media type, see [XHTMLMIME] .
<ShaneM> any objections to this text?
Steven: This is an informative section, right?
... I'll suggest the change to the commentator
... So we are done, modulo checking for missed comments, and checking new comments on the PERs
<Tina> I have none
<ShaneM> ... Roland searches email archives ...
Shane: Yes, this is new
Roland: We'll talk about it when
we get to 1.1
... it is in the wrong place, but let's add it to the issues system
Shane: There are 21 issues
... most trivial
... I haven't done the final pass
... first issue is @id on script
... fixed and closed
Steven: HTML4 disallows these
... and for XHTML we used English text
<Tina> What's the reasoning behind that change?
TIna - because XML doesn't have exclusions
Steven: Oh but wait, Modularization uses exclusions!
Shane: We were so clever!
... we win!
... "Content exclusions should be defined by the rules in M12N, which can't be expressed in all schema grammars"
<MoZ> That's where I think Relax NG + NVDL could help
<MoZ> (It's more Relax NG + Schematron to be fair)
Shane: Ruby does it right too!
Steven: So the reply is that the spec is clear, and does the right thing
<ShaneM> XHTML 1.1 references the XHTML M12N and Ruby Recommendations. Those Recommendations define content models using an abstract content model grammar, and these are specifically excluded already. While the underlying implementation(s) may not be able to enforce these exclusions because of limitations in their grammars, the restrictions are nontheless normative.
Steven: using the definitions as required by M12N
Shane: We don't make changes to
work round bugs in software
... wrong approach
Roland: Keep it as is
Steven: We could consider it for
... it has a reasonable use-case
... this would take us out of PER territory
Shane: We have an objection because this is not fixed
Roland: You think we should be pragmatic
Shane: Yes, we should just do what UAs do now
Steven: The tests make it look like it should be a URI ref
<ShaneM> <img usemap="lala">
Shane: external references have
never been implemented
... so this wouldn't affect current implementations
... You have to change this in m12n
Steven: then the issue is on the wrong spec
Shane: Then I'll move it there
Shane: Editorial; fixed
Shane: Handle this tomorrow
This is scribe.perl Revision: 1.135 of Date: 2009/03/02 03:52:20 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/SH/Sh/ Succeeded: s/tand/stand/ Succeeded: s/TH// Succeeded: s/is is an/This is an/ Succeeded: s/draft/ draft/ Succeeded: s/edition/edition of XML/ Succeeded: s/Stevrn|/Steven/ Succeeded: s/�m/'em/ Succeeded: s/MOz/MoZ/ Succeeded: s/edf/def/ Succeeded: s/SH/Sh/ Succeeded: s/anyway// Succeeded: s/tence/tence?/ Succeeded: s/63/62/ Succeeded: s/SH/Sh/ Found Scribe: Steven Inferring ScribeNick: Steven WARNING: Replacing list of attendees. Old list: Markus Roland Steven New list: Steven Roland ShaneM Alessio Default Present: Steven, Roland, ShaneM, Alessio Present: Steven Roland ShaneM Alessio Tina MoZ Markus Regrets: Gregory Agenda: http://www.w3.org/MarkUp/xhtml2/wiki/2009-06-02-FtF-Agenda Found Date: 02 Jun 2009 Guessing minutes URL: http://www.w3.org/2009/06/02-xhtml-minutes.html People with action items:[End of scribe.perl diagnostic output]