[CSSWG] Minutes and Resolutions 2010-08-18

Summary:

   - Reviewed F2F agenda
   - CSS2.1 Beta 3 published
       http://lists.w3.org/Archives/Public/public-css-testsuite/2010Aug/0033.html
   - RESOLVED: fantasai's new proposal accepted for CSS2.1 Issue 56
   - RESOLVED: proposal accepted for CSS2.1 Issue 137
   - Discussed visibility and text-decoration; discussion moved to F2F
   - Please review new proposal for CSS2.1 Issue 159 for next week
       http://wiki.csswg.org/spec/css2.1#issue-159
       http://lists.w3.org/Archives/Public/www-style/2010Aug/0342.html
   - RESOLVED: min-height/max-height undefined on table rows, cells, etc. in CSS2.1
   - PROPOSED: Make Appendix G informative. Decision to be made next week.
   - RESOLVED: Accepted proposal for CSS2.1 Issue 187 to make atomic inlines
               other than replaced elements with display 'inline' neutral wrt bidi
   - RESOLVED: Proposal accepted for CSS2.1 Issue 190
   - In the interest of completing REC, will be setting a stop date for accepting
     CSS2.1 issues soon.
   - CSSWG will consider moving CSS2.1 from LC to PR if possible, instead of going
     back through CR again.

====== Full minutes below ======

Present:
   David Baron
   Bert Bos
   Beth Dakin
   Arron Eicholz
   Elika Etemad
   Simon Fraser
   Daniel Glazman
   Brad Kemper
   Hĺkon Wium Lie
   Peter Linss
   Steve Zilles

<RRSAgent> logging to http://www.w3.org/2010/08/18-CSS-irc
Scribe: fantasai

F2F planning
------------

   Daniel: First topic, F2F
   Daniel: We have a bunch of items on the telecon F2F. I suggest we
           move items > #2 to the F2F
   nobody says anything
   Daniel: ok, so let's do that

   Daniel: Do we know who is going to be in Oslo on Sunday night?
   howcome: I asked for who is coming to the picnic
   dbaron: there's also an itinerary page for the group
   <Bert> -> http://www.w3.org/Style/Group/2010/Oslo meeting page
   howcome: I would suggest bringing a wind jacket and a bathing suit

   Bert: I added Template Layout to the agenda, since César is coming
   Daniel: Peter and I will work on a schedule

   dbaron: howcome, where are we going to meet on Sunday?
   howcome: I'll put an email out about that tomorrow when I know more about the weather

CSS2.1 Test Suite
-----------------

   fantasai: Beta 3 published
     http://lists.w3.org/Archives/Public/public-css-testsuite/2010Aug/0033.html
   fantasai: Changes listed in the email
   Arron: There are still more updates to go in. I have another list
          from Gérard as well as a couple other emails.
   fantasai: I can do another publish as soon as updates are in

CSS2.1 Issues
-------------

   http://wiki.csswg.org/spec/css2.1#issue-56
   http://lists.w3.org/Archives/Public/www-style/2010Aug/0334.html
   Daniel: what if an image is replaced by alternate text?
   fantasai: Then it will be underlined, because it is text
   dbaron: This sounds good to me
   Bert: sounds fine
   RESOLVED: fantasai's new proposal accepted for CSS2.1 Issue 56

   Steve: So the EPUB group is proposing to use...
   STeve: There's a problem in Chinese and Japanese with place names
   Steve: Sometimes these are written differently than the standard
          Unicode characters.
   Steve: That means they would appear to be a graphic, but would
          actually be text.
   Steve: So I would like the minutes to note that there are times
          when people would use a graphic as a character.
   Steve notes that he is not expressing disagreement with the
     current proposal.
   fantasai: I think we discussed this wrt css3-text
   Steve: I think we were discussing the possibility of adding an
          option to underline images
   <fantasai> http://dev.w3.org/csswg/css3-text/#text-decoration-skip
   fantasai: Yep, we did that.

   http://wiki.csswg.org/spec/css2.1#issue-137
   http://lists.w3.org/Archives/Public/www-style/2010Jul/0226.html
   fantasai: I talked to bzbarsky and he thought this was fine
   dbaron: The float positioning rules do depend on anonymous blocks
           being containing blocks, so we can't make them not
           containing blocks.
   dbaron: This is fine with me though.
   <szilles> +1 for fantasai's soln
   RESOLVED: proposal accepted for CSS2.1 Issue 137

   dbaron: I believe that the visibility of text decorations should
           match the visibility of the text, since we've moved to a
           model where the decorations are drawn by the text.
   Daniel: It's predictable for authors
   Steve: And it fits the new description.
   <smfr> it's fine with me
   Arron: Implementations are all consistent and draw the underline
          for invisible text.
   Steve: The other thing that got discussed last week was do those
          implementations need to change to not draw under other
          things they're not supposed to draw under
   dbaron: For Gecko, you might want to look at our Quirks mode now,
           since that's closer to what the standard now says than
           our Standards mode
   Daniel: I think drawing underlines under text that's not there
           and not selectable is confusing
   Steve: We're now just repeating the discussion from last week
   Steve: I suggest moving this to the F2F
   <dbaron> For the record, I could live with the model where the
            visibility of the decoration goes with the visibility
            of the element the decoration is on.
   <dbaron> But I'd like to avoid anything other than those two choices...
   <dbaron> (well, maybe I'd be ok with the idea that it's visibile
            only if *both* are visible)
   Daniel: Moved to F2F
   * fantasai notes we might want to run other text-decoration tests
     on implementations to see what else need changing
   <arronei> i already have all the results for all the text-decoration
             tests. I will bring that info to the F2F

   http://wiki.csswg.org/spec/css2.1#issue-159
   http://lists.w3.org/Archives/Public/www-style/2010Aug/0342.html
   fantasai: If people want more time to review that, that might make
             sense; I only posted it last night

   http://wiki.csswg.org/spec/css2.1#issue-170
   Brad: Tab and I were saying that it should be undefined now, and
         that it should mean something in CSS3
   dbaron agrees with it being undefined
   Brad: I would love max-height to be a trigger for scrollable
         tbodys or trows
   Daniel: No objection to making it undefined for 2.1?
   RESOLED: min-height / max-height on table rows, cells, etc.
            undefined for CSS2.1
   <bradk> Will there be a note about min-height, etc. being
           unresolved for now?
   <bradk> undefined, I mean
   <glazou> bradk: yes
   <fantasai> Yeah, but I think Bert can write the wording for that

   http://wiki.csswg.org/spec/css2.1#issue-183
   howcome to forward issue 183 to Anne

   http://wiki.csswg.org/spec/css2.1#issue-186
   Daniel: I don't think we can make Appendix G non-normative
   dbaron: Can we say what it means then?
   Bert: I don't think there's anything in that grammar anymore that
         isn't in the prose. So maybe it can be made informative.
   Bert: This is not very useful for anything except the validator,
         and maybe people wanting to generate CSS, though not many
         of those
   Daniel: It could be useful for editors doing syntax highlighting
           and things like that
   fantasai: We're not proposing to remove the grammar, only to
             make it informative.
   fantasai: It can still be used for all those uses, it just
             wouldn't have any weight against the normative prose.
   Daniel: I suggest we write the proposal to make Appendix G
           informative, discuss and resolve during F2F meeting so
           people have time to review the issue

   http://wiki.csswg.org/spec/css2.1#issue-187
   Steve: Why, if I put a bidi property on a table or inline, why
          wouldn't I want it honored?
   fantasai: You do. But you generally want it to be affecting the
             contents of the inline-block, not the contents around it
   Steve: ...
   Steve: so the i18n discussion concluded that they should be neutral?
   fantasai: pretty much. Although they did also expect images to
             always be neutral, and they're not.
   Steve: I think I understand why you're proposing what you do
   fantasai: We're going for neutral because it's safer in general
   Steve: I can live with what you're proposing. I don't have enough
          use cases to know.
   Daniel: Steve, do you need more time?
   Steve: The only thing I would do is go ask my i18n guys, but I'm
          not sure that's going to change much
   RESOLVED: Accepted proposal for CSS2.1 Issue 187 to make atomic
            inlines other than 'inline' replaced as neutral wrt bidi

   <smfr> http://wiki.csswg.org/spec/css2.1#issue-190
   http://test.csswg.org/suites/css2.1/20100815/xhtml1/clip-inherit-001.xht
   Daniel: It seems there is interop against the current spec
   fantasai: I think when we introduced used values, this was one
             of the areas we missed the computation
   Daniel asks around
   <dbaron> I don't understand what the spec means without the change :-)
   Apple and Opera support the proposal
   Arron: I ran the tests in all the browsers and we all do match
   RESOLVED: Proposal accepted for CSS2.1 Issue 190

   Daniel: I will note that the issues list keeps growing. We need to
           close down the issues list and push the rest to errata at
           some point.
   <szilles> +1 for setting a stop date
   Daniel: Otherwise we will never release CSS2.1
   <Bert> +1
   <Bert> (It's not forbidden to publish a REC with open issues, they
          just have to be noted in the announcement.)
   Daniel: We should set a firm date at the F2F
   fantasai: We should also going from LC to PR, if we can
   fantasai: There's an AC meeting at TPAC, ideally we have PR ready
             for it so that we can enter REC by the end of the year
Meeting closed.
   <dbaron> fantasai, voting on PRs has zero to do with AC meetings
   <dbaron> fantasai, it's all by WBS surveys
   * fantasai didn't know that
   <Bert> Formally, yes. But AC reps do get more active around AC
          meetings (just like we get more active around ftfs :-) )
   <Bert> And if we need a Director's decision, either to enter or
          exit PR, at the AC meeting we have the required people all
          in one place, no need for a hard-to-schedule telcon.

Received on Wednesday, 18 August 2010 18:03:44 UTC