CSS WG weekly meeting minutes 2008-06-04

Meeting: Cascading Style Sheets (CSS) Working Group Weekly Teleconference

Date: June 4, 2008
Time: 09:06am to 10:00am PDT

Attendees:
David Baron
Arron Eicholz
Elika Etemad
Ming Gao
Daniel Glazman
Molly Holzschlag
Peter Linss
Jason Cranford Teague

Official regrets:
Bert Bos
Anne van Kesteren
Håkon Wium Lie
David Singer
Mohamed ZERGAOUI
Steve Zilles

Scriber: Ming Gao


<RRSAgent> logging to http://www.w3.org/2008/05/21-css-irc

daniel: quick agenda review; any new item?
...... No addition


Topic:  Charter, by Peter
-------------------------
peter: module list updated last week; will send it out this week
jason: have you received my note on hypertext link style?
fantasai: which note?
jason: regarding pseudo class
fantasai: then should go into selector module
jason: make sense; though does affect link style
... ok; I agree to cancel my note
... more about link pseudo class; not a huge thing, from a designer point of view

daniel: peter, please send the complete list of modules out today.
peter: yes

fantasai: we have not talked about whether to do selector 4 module. does not have  feedbacks, since f-2-f meeting (march);
... should this be listed in the charter? I think it should be in scope
daniel: selector was the first thing that had been worked in CSS WG
... making it an extension would be a better course.
fantasai: ok; I can't disagree with this; I haven't heard the implementor's view.
daniel: would like implementor focus on other modules.


Topic: background and borders issues
------------------------------------
fantasai: bert and I am working on these issues
... would seek implementors' view

daniel: is there a third way than accept or reject (the current draft), e.g. working on  this later
fantasai: don't want to keep working on the draft, would spend energy somewhere else.
peter/molly: would like to have a place to keep the work done so far, for next round.
... don't want losing the work
fantasai: has listed some of these work on the wiki pages

issue #1
--------
fantasai: is anyone (implementors) wanting the multiple borders in module 3?
david b.: yes
daniel: anyone else, Microsoft, Opera?
arron: can't say for sure; likely won't need it for a while

david b.: not sure what is to be implemented
fantasai: not sure either
<fantasai> dbaron: Mozilla has a multiple borders feature, but I don't think it's what  we want here
daniel: so better to drop it for this round; need more work
daniel: all agree; done
<fantasai> RESOLVED: no multiple borders in level 3

issue #2
--------
fantasai: percentage border width, issue 26
... any pressing reason to add this?

daniel: is there a use case for this?
fantasai: don't know one
daniel: don't see border width specified this way, rather in pixel; jason?
jason: agree. can't think of a use case where the percentage border width is to be  used.
<dbaron> I would note that one reason you don't see non-pixel borders is that
... some implementations (e.g., older version of Gecko) will often make them uneven,
... which isn't what authors want.

molly: the only place (I can think of) is scalable design; not someone is practicing  these days.
david b: [repeating his comments above]

<fantasai> jason: I don't think it will make or break any designs
daniel: so we can resolve it by not doing this for now
<fantasai> RESOLVED: no percentage borders

issue #3
--------
fantasai: next issue: shadows; looking at the pictures
<fantasai> http://bradclicks.com/cssplay/Shadows.html
<fantasai> jason, molly: designers would be very interested.
<fantasai> dbaron: would this feature give you 5% of what designers want from
... this, or 80% of what designers want?
daniel/molly/jason: a substantial feature (i.e. the 80%)
jason: more designers want is to do this like they can with photoshop

jason: is this to provide some webfloor effect?
daniel: no, just the shadow effect

david b: want to make a distinction between box shadow effect vs text shadow effect
fantasai: yes.
molly: need to do both

david: a general question for the WG: do we want to pull one feature at a time based on  what SVG can do, for
... next 20 years?

daniel: do we want to do features depending on external engine, or focus on CSS style
... suggest this question for SVG/CSS WG joint session in the fall
molly/david: agree
daniel: any protocol on coordination with SVG features?

molly: if do shadow effect, can't do a half job, as this is a common feature in web  design
<fantasai> molly: if we're doing shadows, we should do inner shadows
fantasai: doing it as an inner shadows than external shadows
<fantasai> fantasai: I would probably just add an 'inset' keyword to the shadow

daniel: continue working on this box shadow, don't drop it
<fantasai> RESOLVED: continue work on inner shadows

issue #4
--------
daniel: next, positioning from corners
fantasai: the current approach might be a bit awkward.
... an alternative approach would require a different syntax.

<fantasai> background-postion: 10px 20px;
fantasai: designer tend to prefer to do this positioning from bottom right.
<fantasai> background-position: bottom 10px right 20px;
fantasai: my proposal is to use keyword and distance
... from the edge of the element

daniel/fantasai: would like to hear what authors want to say
jason: never really have a case to do it from left or right;
... could be because I can't do it in the past

<fantasai> fantasai: question is whether to create a syntax for this, or to wait
for calc()
fantasai: people really want to do from bottom right
jason: agree; what is the syntax to write it from right to left?

fantasai: I am posting some notes to show the syntax
<fantasai> background-position: bottom 10px right 20px;
<fantasai> background-position: start 10px center;

jason: just looking at these syntax, it is hard to understand (what they do, from  authors' point of view);
... maybe easy for computer to understand

peter: Q: do you mean 10 px from righ edge of the box ?

fantasaI: could be right edge from the right edge(?)

david b: calc() expression has the percentage in it.
david b: two passes of calculation, one for image and one for the other

jason: is this to create the padding of what the background is?
... sometime, need to create both background content and background box

daniel: do we have consensus to work on this?

<fantasai> http://lists.w3.org/Archives/Public/www-style/2008Jan/0365.html
fantasai: inclined to add this; maybe put in the next draft and ask for feedbacks
daniel: agree? ok, and resolved.

fantasai: post my proposal and david b. has some comment. are we ok?
<fantasai> http://lists.w3.org/Archives/Public/www-style/2008May/0148.html

daniel: resolved.


Topic: template layout
----------------------
<glazou> http://www.w3.org/Style/Group/css3-src/css3-layout/Overview.html

<fantasai> Any issues that received no comment in that message will be marked as
resolved?

daniel: can current draft be considered as a working draft?
peter: don't see why not
... the current draft is a year old
daniel: this is the third working draft

fantasai: think this module need a lot of work, before LC
... though no problem publishing it as working draft for comments

daniel: resolved


Topic: moving css3-color to LC
------------------------------
david: several issues
dbaron: one about z-index required adding a new paragraph
dbaron: another I proposed no change
dbaron: another I resolved by pulling in diffs from css2.1

daniel: I personally have no objection to release it to LC; others?
fantasai: no comments
arron: no objection
molly/jason: no objection

daniel: resolved: release to last call

david b: when the proposal was in CR, we had received some comments
david: when should we respond to them? will people get confused?
<dbaron> the question was really about when we publish the disposition of
comments

daniel: color is something a lot of groups depending on, so good to respond now
<glazou> dbaron: we can have an online document ready but formally not called "DoC"



Topic: test suite and test review process
-----------------------------------------
<fantasai> Ming: In April I proposed a review process that me, elika and arron
put together
<fantasai> Ming: to use wiki pages for review comments
<fantasai> Ming: One of the key steps in the process is the peer or approver --
... the final review

<fantasai> Ming: Recently Arron and I discussed how a person qualifies to become a peer
<fantasai> Ming: The reviewing is the bulk of the work right now.. it's the bottleneck
<fantasai> Ming: If we're looking at peers we have today, we don't have a lot of active  peers
<fantasai> Ming: Maybe David will have time to help more now ...
<fantasai> Ming: but we lack peer resources
<fantasai> Ming: So I proposed a process for someone with enough creditials to apply  for peer status
<fantasai> Ming: My proposal is based on Elika's policy
<fantasai> http://csswg.inkedblade.net/test/css2.1/review
<fantasai> Ming: I took those one step further, and said that person has to
review 50-100 testcases across modules and with sufficient complexity
<fantasai> Ming: to demonstrate competency
<fantasai> Ming: When a person applies for peer position and get approved, then person  has to continue contributing
<fantasai> Ming: e.g. review (approving/rejecting) 30-50 testcases a month

<fantasai> Daniel: Have you discussed this process with W3C management?
<fantasai> Daniel: Like it; This could be a general process for test case review in  W3C.

<fantasai> dbaron: In open source projects, the rules are less formal.
<fantasai> dbaron: and becoming a peer is more a recognition of the person's status
... rather than something you apply for

<fantasai> Ming: Then maybe we need another status, not peer, but something like  "Approver"
<fantasai> Ming: Someone who demonstrates ability to make high quality comments on  testcases

<fantasai> ... some discussion ...

<fantasai> Peter summarizes: We like the idea of having a list of active reviewers.
... We want a call to the public for more people to join this list and help review  tests
... We want members of the public to help review tests.
... We want to have a list of "final reviewers", and publish process for becoming a  final reviewer

<fantasai> Molly: we can publicize through www-style, our blog, Eric Meyer's group,  personal blogs
<fantasai> Elika: I've been waiting for licensing issues to issue a call for help with  the test suite

<glazou> fantasai: I can extend the wiki to host comments on the test suite
<fantasai> Elika: There are several ways to help: reviewing testcases,
... reporting failures, fixing testcases in response to review comments, writing  testcases

<fantasai> Peter: We shouldn't hold back on that for issuing a call for reviewers
<fantasai> Elika: Ming and I discussed using the wiki for reviews, should set that up  first
fantasai: I would like to have that set up first before our communication; should be  done in two weeks;
peter: agree; but let's not wait for all pieces in place.

daniel: meeting adjourned

original minutes URL: http://www.w3.org/2008/06/04-css-minutes.html

Received on Thursday, 5 June 2008 21:14:07 UTC