W3C home > Mailing lists > Public > public-i18n-its@w3.org > July to September 2006

[I18N ITS] Teleconference Minutes 2006-07-17

From: Felix Sasaki <fsasaki@w3.org>
Date: Tue, 18 Jul 2006 00:52:48 +0900
Message-ID: <44BBB250.7020007@w3.org>
To: public-i18n-its@w3.org
... are at http://www.w3.org/2006/07/17-i18nits-minutes.html

and below as text.

- Felix


      [1] http://www.w3.org/

                        i18n ITS working group

17 Jul 2006




          Felix, Richard, Yves





     * [3]Topics
         1. [4]four mails from i18n core
         2. [5]issue 5
         3. [6]issue 6 at
         4. [7]issue 7 at
         5. [8]issue 8 at
         6. [9]xml:lang issue (again)
     * [10]Summary of Action Items

four mails from i18n core

   see topic items 1-4 at


   Yves: everbody at the last call agreed in leaving them to the
   ... so we skip over them now, to gain some time

   Richard: sounds good to me

   Felix: me as well

issue 5 [12]http://www.w3.org/Bugs/Public/show_bug.cgi?id=3469

     [12] http://www.w3.org/Bugs/Public/show_bug.cgi?id=3469

   Yves: the wording regarding xml:lang

   yves comment is at

     [13] http://www.w3.org/Bugs/Public/show_bug.cgi?id=3469#c1

   Richard: we don't say in the spec that langRule points to s.t. which
   is to be treated like xml:lang

   Yves: you made that issues in other comments

   Richard: So we should have a different wording for xml:lang ?
   ... we come back to this issue later, after addressing the other
   xml:lang issue (see next topic)

issue 6 at [14]http://www.w3.org/Bugs/Public/show_bug.cgi?id=3468

     [14] http://www.w3.org/Bugs/Public/show_bug.cgi?id=3468

   Yves: you said that xml:lang should be in the ITS markup
   ... Sebastian had some discussion with you on that, Richard

   Richard: Sebastian said "xml:lang comes with xml anyway"
   ... I said you have to declare it in your format
   ... depending on the schema language

   Yves: I would add xml:lang here and to the rules element

   Felix: might create an issue with being schema language independent

   Richard: how about having a paragraph on that in the spec?

   Yves: this needs more discussion probably, with Sebastian

issue 7 at [15]http://www.w3.org/Bugs/Public/show_bug.cgi?id=3467

     [15] http://www.w3.org/Bugs/Public/show_bug.cgi?id=3467

   Yves: why did we do that?

   Felix: this is about the global case, right?

   Yves: yes, like locInfo

   Felix: I agree, having it the same like locInfo

   Yves: Richard, would you see this as an substantive change?

   Richard: not sure
   ... I described some comments as substantive, but it depends on you
   how you classify the changes
   ... this could go both ways
   ... a question on the content model of the ruby design: is it plain

   Felix: it is inline, like in the ruby TR

   Yves: so resolution: we change to have an element instead of an

issue 8 at [16]http://www.w3.org/Bugs/Public/show_bug.cgi?id=3466

     [16] http://www.w3.org/Bugs/Public/show_bug.cgi?id=3466

   Yves: this is also about ruby

   Richard: you cannot point to markup which does not match the RUBY TR
   ... that should be made explicit
   ... existing ruby markup is only the one which is conformant to the

   Felix: good comment, fine with that

   Richard: another issue: does XHTML 1.1 need global rules?
   ... i.e., does one need to point to the ruby text?

   Yves: from the ITS viewpoint it is not clear that it is ruby

   Felix: it depends on what you want to process
   ... it is not against the ITS conformance criteria *not* to
   associate xhtml 1.1. with ITS

   Richard: it is maybe a best practice
   ... whether people should use these mechanisms
   ... the same is true for xml:lang

   Yves: for me, it has to do with each vocabulary
   ... xml:lang is a different level
   ... you cannot say "every xml processor should know about ruby"

   Felix: I am not sure about the consequences of a must. It might be
   hard to test

   Richard: what happens if we have a "should"?

   Yves: we need more discussion
   ... as for ruby: we need to reword the paragraph which is after the
   example, right?
   ... what about the xhtml 1.1. issue?
   ... you should map ruby to ITS, if it is possible, that is it,
   ... we have an example of that with opendocument
   ... we could say "it is a best practice to associate as much as
   ... but we cannot force you to associate everything"

xml:lang issue (again)

   Yves: Felix, you agree that having the attribute would be a good
   ... only the implemenation with RELAX NG is an issue

   Felix: Is it a problem that other specs like XQuery use xml:lang
   without ITS or schemas?

   Yves: we could say "production rules cover only ITS attributes", for
   others you may need other stuff like xml:lang

   Felix: worried about the output of processing if we use xml:lang

   Yves: it does not to be logical that an ITS processor has to be
   aware of xml:lang

   Felix: we might double functionality with existing specs like XQuery
   ... how about saying "please use xml:lang locally", and have
   langRule if they don't use it

   Yves: having xml:lang in the schemas is useful from the practical
   ... and an ITS processor should be aware of xml:lang
   ... how about removing the data category?

   Felix: we have to continue we discussion

Summary of Action Items

   [End of minutes]

    Minutes formatted by David Booth's [17]scribe.perl version 1.127
    ([18]CVS log)
    $Date: 2006/07/17 15:32:22 $

     [17] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
     [18] http://dev.w3.org/cvsweb/2002/scribe/

Received on Monday, 17 July 2006 15:53:19 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:43:07 UTC