i18n core working group

ation items

<scribe> ACTION: all to look at charmod resid test suite and gather feedback (ONGOING) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action01]

francois: no new feedback
... this will be ongoing

<scribe> ACTION: all to give feedback on the LTLI draft (ONGOING) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action02]

<scribe> ACTION: Francois, ping Martin about when we can make the tests public (DONE) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action03]

francois: did hear from Martin
... all pages are no accessible
... as for hosting: that will be o.k. as well

Richard: problem is: we need somebody to edit it or change s.t.
... in the W3C side, everybody can do that

Francois: there are system dependencies which are part of the tests

Richard: we could probably do that in W3C as well
... another problem: How to accept tests from his Organization
... since they are not a member of W3C?
... again a patent policy problem

Francois: do we say *we* maintain a test suite?

"as soon as the following conditions are met:": no need for *us* to procude the test suite

francois: so no requirement to have a W3C test suite
... for XML we have a test suite. It was hosted by MIT
... NIST is a member, and there were some processes to accept that as a contribution.

Richard: one question: is a non w3c test suite sufficient to get through CR?

<r12a> three main possible scenarios:

<r12a> [1] stays on martin\s servers and just point to those tests over there to fulfill CR criteria

<r12a> [2] stays on martins servers, but we call them W3C tests

<r12a> [3] move the tests to W3C servers so taht we can handle maintenance more easily

<r12a> each scenario 1 -3 probably involves increasing administrative complexity

<r12a> ACTION: RI, FS, determine what is the minimal course necessary to pass CR [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action04]

felix: I will send a mail to plh

<fyergeau> Correction: XML test suite is now hosted at W3C

<scribe> ACTION: Felix to find out where the tests from Martin are (DROPPED) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action05]

francois: I mentioned to Martin a start / presentation page, he said it would be a good idea

<scribe> ACTION: Mary will make the updates discussed in Mandelieu on ws-i18n [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action06]

<scribe> ACTION: Mary to review http://www.w3.org/TR/its/ after 18 May before 30 June (ONGOING) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action07]

<scribe> ACTION: Francois to build a current issues list on charmod norm (PENDING) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action08]

<scribe> ACTION: Felix to implement discussion results from the i18n-comments list (DONE) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action09]

<scribe> ACTION: Felix to create examples for defined terms (DONE) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action10]

<scribe> ACTION: Felix to ask for a review extension for WCAG (DONE) [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action11]

i18n activity restructuring

Richard: we thought of i18n activity restructuring
... which AC reps at the AC meeting will see
... it is a discussion, not decided yet

<r12a> http://www.w3.org/International/Activity.html

richard: take a look at the bottom of the document
... benefit is both for core and the spec work
... key benefit: we have been slowing with charmod because we have other stuff on the radar
... on the other hand, spec work sometimes takes time from the review work
... separating out the core work from the specification work will help a lot
... web services work will probably be finished very quickly

Francois: What will happen for specs which are a REC?

richard: we have to think about that as well

Francois: I like the proposal, only life after REC needs to be made clearer
... if there is a version 2.0 of a spec and there is a working group, that's o.k.

richard: we will also stagger meetings / telefone conferences
... e.g. a charmod call every two weeks
... key benefit would be that work on specific topics is more focused, not distracted by other topics

francois: so the new core working group will be not under patent policy?

richard: yes
... it would be beneficial to many people who want to join core

francois: one problem: this goes for the AC, and there is a call for participation
... there is a risc that there is not enough participation to start a working group

richard: for core that's no risk, we have the GEO folks
... for the ws working group, we need to have enough support to generate a working group
... I think we don't need to have a huge working group
... as for charmod, it would be maybe only three people
... everybody fine with the proposal?
... (people on this call)

no objection

<karuneshvijay> yes

topics: LTLI

Francois: Felix implemented many changes

Felix: Addison send new comments, please have a look

Francois: so we send a publication request on Thursday
... anything else on LTLI?
... nice to see the progress, if you have comments, please send them in

<scribe> ACTION: Felix to send a re-publication request for the LTLI draft on Thursday [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action12]

first letter discussion

Richard: it is a pseudo element

francois: there was a call for details, and a lot of mail about the Indic languages
... discussion on the www-international list
... should we write a Note about this?
... the CSS people seem to be very reluctant to make changes

Richard: Erika and I talked about it
... she asked me about (indic) examples, but not major changes for the spec
... maybe we should push for a stronger wording

Francois: the examples show that "first letter, just one unicode character" is not sufficient

Richard: the default grapheme cluster rule would not cope with Indian scripts, would it?

Francois: I'm not sure
... this should be checked

Richard: currently we say:

Richard quotes the current texts on "first letter"

<r12a> original text:

<r12a> Some languages may have specific rules about how to treat certain letter

<r12a> # combinations. In Dutch, for example, if the letter combination "ij" appears

<r12a> # at the beginning of a word, both letters should be considered within the

<r12a> # :first-letter pseudo-element.

<r12a> first proposal:

<r12a> Note: In some cases the :first-letter pseudo-element should include more

<r12a> than just the first non-punctuation code point on a line. For example,

<r12a> combining characters must be kept with their base character. Additionally,

<r12a> some languages may have specific rules about how to treat certain letter

<r12a> combinations. In Dutch, for example, if the letter combination "ij"

<r12a> appears at the beginning of an element, both letters should be considered

<r12a> within the ::first-letter pseudo-element. The UA should include at least

<r12a> the entire default grapheme cluster as defined by [UAX 29 Text Boundaries]

<r12a> and may include more than that as appropriate.

Francois: we where proposing "default grapheme cluster" as the base line
... they (CSS group) did not want to have that as a normative part, which is regretable
... because it does not cover many common things
... maybe we should push back now

Richard: karunesh, Vijay, is where a language specific varitation of what a first letter is?
... in latin scripts, there is language specific variations on how first letters shoud be treated, e.g. "ij" in Dutch
... the question is: are there language specific variations of first letter styling for various indic scripts?

Karunesh: If you use Dev., the rules would be the same
... the situation might be different for other (indic)scripts
... the grapheme cluster seems to be an approriate unit

karunesh gives some examples

Richard: gathering this information sounds interesting

Karunesh: we are investigating from many specialists

Richard: Should we set up a wiki for this?

Francois: good idea

Karunesh: wiki sounds also good to me

Richard: would be good to involve the Unicode list
... many people on that list could make some input on this
... I'll put out another blog entry on this, which will point to the new wiki, and send a mail to the Unicode list about this

<scribe> ACTION: Richard to create a wiki on "first letter", post a blog and write a mail to the Unicode list about it [recorded in http://www.w3.org/2006/05/16-i18ncore-minutes.html#action13]

other business

francois: next week no meeting