[CSSWG] Minutes and Resolutions 2010-05-04

   - Discussed charter
   - Plan to trim down css3-images, will discuss on www-style
   - RESOLVED: Publish CSS3 Lists WD
   - RESOLVED: jdaggett, szilles as editors of css3-linebox
   - RESOLVED: Publish CSS Snapshots once unprefixed legacy properties issue
               is noted
   - RESOLVED: Remove phonemes property; include note to explain removal
   - Discussed CSS logo

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

   David Baron
   Bert Bos
   John Daggett
   Arron Eicholz
   Elika Etemad
   Simon Fraser
   Sylvain Galineau
   Daniel Glazman
   Koji Ishii
   John Jansen
   Brad Kemper
   Håkon Wium Lie
   Peter Linss
   Edward O'Connor
   David Singer
   Alan Stearns
   Daniel Weck
   Steve Zilles

<RRSAgent> logging to http://www.w3.org/2011/05/04-css-irc
Scribe: Sylvain Galineau


   glazou: extra agenda items ?
   jdaggett: I'd like to take over editing of the line layout spec
   jdaggett: and an update on the charter if possible

CSSWG Charter

   jdaggett: looking at the draft charter, the line grid spec ended on the
             charter as a medium priority spec even though we haven't
             discussed this
   jdaggett: i'm not sure it should be a separate spec vs. being part of
             line layout. there seems to be a lot of overlap. why is it in
             the charter and how did it end up on the status page ?
   szilles: I believe it was discussed in the context of line alignment
            across columns. and layout for east asian languages
   jdaggett: sure but when did we agree to make it part of the charter ?
   jdaggett: I don't think it warrants a separate spec.
   szilles: for the sake of making progress it may be easier to list them
            as they are now
   glazou: if it's on the charter we can work on it in the course of the
   jdaggett: it is medium priority and the decision to make it so is unclear
   <dsinger> This is a draft charter for discussion, right? And now we are
             discussing it.
   szilles: the criteria for medium priority is a champion to push it
            forward. And Koji is the champion
   <dsinger> Anyone can propose into the DRAFT
   (agreement to wait for Chris Lilley)
   glazou: what else about the charter ?
   * Bert sorry for being late. Had some business to finish.
   howcome: I'm championing GCPM so should it be medium ?
   szilles: I think it also needs to get to LC or CR in the period
   fantasai: we also used to require implementors working on it
   szilles: that is a reasonable criteria for High Priority
   szilles: should we poll vendors on implementation interest ?
   plinss: we did a private poll last time. it could be helpful to update
           the data we collected last time
   glazou: we should stabilize the list and then run the poll
   howcome: if those specs expected to reach PR are high priority then
            css3-multicol should be High.
   <glazou> http://www.w3.org/2010/09/CSSWG/charter.html
   howcome: everything being in scope I'm not sure we can read much in the
            hi/med/low ranking
   <dsinger> We ought to prioritize staff, wg, f2f time etc. On the priorities
   glazou: taking the hi pri documents to CR is not the hard part. moving
           test suites along for them all is the larger work item
   glazou: let's continue this discussion when Chris is present.
   szilles: with the exception of jdaggett's and howcome's comment, does
            anybody on the call think something is missing or shouldn't
            be there ?
   szilles: I thought ChrisL believed the draft was ready to go in front of
            the AC membership in two weeks so we should at least identify
            obvious bugs without waiting for ChrisL

   glazou: my only concern is the number of documents in high-priority.
           are we creating an expectation of achieving PR for everything
           in high priority ?
   szilles: there is no harm in having fewer things in high priority and
            exceeding expectations
   glazou: yes
   szilles: any good candidates to move from high to medium ?
   glazou: this is something where poll data would be useful input since
           browser vendors commitment to implement and submit testcases is key
   <dbaron> I think it may be a good idea... but I'd sort of like to think
            about it... and perhaps try it to see how it goes.
   <sgalineau> dbaron, agree. something we should pilot
   <dbaron> (it == requiring test suite submission for dropping prefixes)

Publishing updates to css3-lists and css3-images

   fantasai: these are request from Tab. I'm generally in support of publishing.
   fantasai: for lists, we should resolve the issue at the end of the agenda
             and publish

   glazou: css3-images. any objection to publishing a new draft ?
   <smfr> http://dev.w3.org/csswg/css3-images/
   dbaron: this draft seems to get more features
   dbaron: was originally something we were going to get out quickly, but
           seems to be growing more features, which makes that harder
   sylvaing: I'd like to move gradients forward in level 3; the rest in level 4
   fantasai: one of the items was the image annotation which we moved out
             from backgrounds & borders.
   fantasai: we have implementation of object-fit et al. but under other names.
             these are implemented but not in browsers
   dbaron: now I'm a little less scared after reading through the TOC more closely
   glazou: should we extract the parts that are ready for LC/CR ?
   fantasai: I would prefer to cut features than extract them into another module
   (some agreement to thin Level 3 and move the rest to Level 4 WD)
   <dbaron> I think cross-fade() is also wrong -- using the wrong definition
            in terms of porter-duff.
   glazou: it'd be better to do these edits before republishing
   fantasai: cutting features is easy as long as we have WG agreement on what
             to keep/what to drop
   fantasai: I can do the edits while Tab is away
   glazou: I suggest we move this discussion to the mailing list so we conclude
           on the cuts at the next telcon
   <glazou> http://lists.w3.org/Archives/Public/www-style/2011Apr/0764.html
   <smfr> http://dev.w3.org/csswg/css3-lists/
   ACTION: fantasai start mailing list discussion on trimming css3-images
   <trackbot> Created ACTION-320

   jdaggett: I think it would be helpful if Tab were here to decide on what
             list items are to be edited out
   fantasai: for complex counter style, we need to define cjk-ideographic
             because it's been defined in CSS2.0 and implemented in Gecko and
             WebKit already. Also EPUB needs it
   glazou: are we OK with publishing now ?
   fantasai: there was discussion about the longhand style for ideographic
             counters up to large numbers. this has never been published
             and I'd like to at least have it as an appendix so we can refer
             to it later
   fantasai: it'd be an informative appendix today, we don't lose this info
             and we still have the option of making it normative in the future
   fantasai: ...so the spec can limit itself to a 10k range but we don't lose
             the correct information for ranges beyond that
   <dsinger> and indeed, if someone wants to implement something that is
             robust into the trillions, they should be able to
   howcome: I think there is merit in documenting this. I'm not sure it needs
            to be in the WD.
   jdaggett: numbering systems are complex. I think the features need to
             target use-cases.
   fantasai: my concern is that we have published definitions that were wrong
   <dsinger> "Note: this algorithm truncates at 10,000; previous CSS
              publications went to a higher limit but had errors in the
   glazou: I propose to publish this WD as-is
   jdaggett: sounds fine
   No objections
   RESOLVED: publish css3-lists WD

   szilles: when we decide to publish a WD, we should also think of people who
            do not follow www-style on a daily basis

Line Layout module editor

   jdaggett: I would like to take that up as css3-fonts is completed
   jdaggett: dbaron is the current editor and I'd work with him to get his
             pending edits in and take it from there
   szilles: I would like to be considered as a co-editor.
   jdaggett: I'm fine with that
   RESOLVED: jdaggett,szilles editors of Line Layout

Kyoto F2F - agenda items

   szilles: since we have the workshop the day before, i'd like items
            discussed in the workshop to being our f2f agenda
   szilles: so that some attendees of the workshop can easily attend both
   szilles: I thought the discussions of css3-text and css3-writing-modes
            would be relevant to the workshop. any other ?
   fantasai: these are the main ones
   szilles: maybe Line Grid as well.
   szilles: I would like to have a discussion of Regions and Grid Layout
   glazou: everyone, please gather agenda items for Kyoto

CSS Snapshot publication

   fantasai: also our handling of prefixing which is defined in the snapshot
   <fantasai> http://dev.w3.org/csswg/css-2010/#experimental
   fantasai: our previous discussion was to clarify that section.
   sylvaing: I thought dbaron wanted to think about it
   dbaron: I think I'm fine with publishing
   smfr: I think this is fine
   howcome: no objection
   szilles: i'm ok
   dbaron: I think we need some sort of exception for features that other
           browsers have already shipped unprefixed (e.g., 'text-overflow')
   fantasai: add, e.g. "the CSSWG may allow certain features to be unprefixed
             for legacy reasons"?
   szilles: I think it would be better for the WG to grant the exception
   dbaron: maybe we can note this as an issue for this publication
   no objections
   RESOLVED: release the document once this issue has been noted


   <glazou> http://daniel.weck.free.fr/CSS-Logo/
   howcome: i like it
   glazou: i like it too
   howcome and sylvain appreciate the historic connection to the old logo
   smfr: it looks 3d so i like it :)
   <jdaggett> not a huge fan of gill sans though....
   <bradk> I'm not terribly excited about it. Why 3 boxes? Something about
           the box model?
   glazou: thanks a lot to daniel for contributing this logo
   plinss: i like it to but we were asked for a design brief
   glazou: a proposal can't hurt
   plinss: I'd like to capture the intent of what we're going for
   plinss: i.e. 'this is what we want out of a logo and this is a proposal'
   <bradk> I'm also tired of Gill Sans
   Bert: so who will produce a design brief ?
   danielweck: I wrote my own. maybe we can just improve on that
   fantasai volunteers bradk...
   <bradk> gee thanks.
   <bradk> As far as the logo goes, I don't have anything better in mind,
           and I thank Daniel for his work. I'd like to see us agree on a
           design brief and see something grow from that, starting with
           multiple sketches in order to define the direction for further
   <bradk> Usually when designing a logo, there is more than one round of
           sketches and more than one week after the first round in order
           to arrive at a final logo.

CSS3 Speech

   <danielweck> request to drop phonemes from css3-speech
   <danielweck> http://lists.w3.org/Archives/Public/www-style/2011Apr/0746.html
   <danielweck> http://www.w3.org/TR/css3-speech/#phonemes
   danielweck: the first link explains the current situation. there is a
               consensus that the phonemes property in the current css3 draft
               breaks the separation of content and presentation
   danielweck: i also argue that EPUB will drive this issue forward using an
               SSML-based solution.
   glazou: so you argue this is not part of css3-speech because it'll be
           defined elsewhere
   danielweck: right. and also that this is more likely to be part of HTML
               given an EPUB solution
   <Bert> (I once proposed: 'phoneme-map: "ape" = \676\324\486, "bar" =
           \9084\7854\78=4' which picks whatever string actually occurs in
           the text.)
   glazou: i suggest a note be included to explain this removal
   danielweck: that's fine with me
   <Bert> Peter's is indeed one possible implementation of my principle.
   RESOLVED: remove phonemes property; include note to explain removal

Meeting closed.

<fantasai> bradk: I think the proposal was for you to draw up the design brief
<fantasai> bradk: not to design the logo
<bradk> Ah. I could probably do that if I can find time. Maybe a draft
         to see if we all agree. I pretty much hate writing up design
         briefs, but I could probably knuckle down and come up with something.
* fantasai has no clue how to come up with a design brief
<fantasai> Working backwards from the existing logos, I'd say
<fantasai> - Connection to the previous logo seems important to CSSWG
<fantasai> - Sense of "cascading"
<fantasai> - Sense of building something
<plinss> - modular components building on one another
<plinss> - 3d logo vs 2d logo - more modern feel, impression of evolution
            from prior logo
<plinss> - brings a feeling of "style"

Received on Wednesday, 4 May 2011 23:48:00 UTC