- From: Dael Jackson <daelcss@gmail.com>
- Date: Mon, 13 Feb 2017 21:32:23 -0500
- To: www-style@w3.org
=========================================
These are the official CSSWG minutes.
Unless you're correcting the minutes,
Please respond by starting a new thread
with an appropriate subject line.
=========================================
Logical Properties
------------------
- RESOLVED: gCS() on logical props returns exactly the same value
as its equivalent physical property.
<url> in content
----------------
- RESOLVED: url() in 'content' only applies to images and nothing
else in visual canvas, and to audio files in aural
canvas.
SVG Integration
---------------
- RESOLVED: Define everything in CSS.
CSS 2.1
-------
- RESOLVED: Publish css21-testing NOTE.
Snapshot
--------
- RESOLVED: Make the additions to the snapshot listed here:
https://github.com/w3c/csswg-drafts/issues/826#issuecomment-272592572
- RESOLVED: Short name for the snapshot publication is "css-2017"
- Florian will make these edits after TabAtkins fixes an issue
with indexes and bikeshed.
===== FULL MINUTES BELOW ======
Agenda: https://wiki.csswg.org/planning/seattle-2017
Scribe: TabAtkins
Logical Properties
------------------
<Rossen> https://github.com/w3c/csswg-drafts/issues/384
dbaron: We should decide what happens when you ask for computed
value for logical properties.
dbaron: Three answers in circulation. I have a preference for one.
dbaron: Brief refresher:
dbaron: margin-block-start. If you're writing in horizontal text,
that's margin-top, etc.
dbaron: It affects the computed style of margin-top or
margin-right.
dbaron: So what happens if you call gCS().marginBlockStart?
dbaron: (1) empty string. Logical properties don't have computed
styles, computed styles only live on the physical
properties.
dbaron: That's not very satisfying, no good reason for it. But
Gecko does it now, because we can easily change from it.
dbaron: (2) (my pref) if you ask for margin-block-start, and
block-start == top, you get margin-top's computed value.
dbaron: (3) various quirky things for various physical properties,
and we'd prefer to fix them and not mirror them to the
logical properties. Such as "this is a used value" vs
"this is a computed value".
dbaron: I'd rather not do that.
dbaron: I'd prefer to just resolve on (2).
<fantasai> +1
dbaron: In (2), in vertical Japanese, gCS().marginBlockStart ==
gCS().marginRight.
dbaron: There are two props that alias each other; which pairs
alias which depend on direction and writing-mode.
fantasai: Strongly in favor of (2).
dauwhe: (2) as well.
<iank> for houdini would like logical props to have a computed
style in maps. (2/3) would be nice.
RESOLVED: gCS() on logical props returns exactly the same value as
its equivalent physical property.
<url> in content
----------------
Scribe: fantasai
<Rossen> https://github.com/w3c/csswg-drafts/issues/821
dbaron: First problem is that, if you look at definition of
content property, it says computed value as specified
below.
dbaron: And then none of the text below actually has the words
computed value.
Rossen: So it's underspecified.
dauwhe: Lots of it is copied from 2003.
plinss: Not clear how far below.
Florian: You mean, somewhere in Australian office?
dbaron: Don't want to go too far into this.
dbaron: Answering first part is editorial.
dbaron: Other issue is:
dbaron: Content accepts a url.
dbaron: In theory, could say url(foo.html) and it will load an
iframe
dbaron: or load url(...) and it will load a plugin.
dbaron: I would like to say that doesn't happen.
dbaron: I think we should just say it's an image.
fremy: Video?
tantek: svg profile for this, more like an image less like an app.
Florian: That is declared dead and out of scope of new SVG charter.
Rossen: SVG integration is not out of scope.
tantek: I don't think bg-image allows for mp4.
TabAtkins: Nor JS to run.
dbaron: I think we might want to discuss videos as images.
dbaron: Don't think we should do it for content and not bg-image,
or vice versa.
tantek: Are you asking for same constraints in content as for
bg-image?
Scribe: tantek
fantasai: I think this makes sense in the general case, which is
visual browsers.
fantasai: If we're rendering to an aural canvas, you should be
allowed to have sound
fantasai: e.g. mp3 file.
fantasai: Would be reasonable thing to do if you're rendering to
an aural canvas.
fantasai: We could say for visual browsers you only render things
that can be rendered visually, like images.
fantasai: But in parallel to that, if you're rendering for a
aural, canvas, you render a sound.
Rossen: Can we reference speech module?
fantasai: I think speech module already references content
property.
Rossen: Reasonable clarification.
Florian: The way you specified it .. you might get one or the
other.
fantasai: We have media queries.
Florian: The cascading will be done before it is an image or sound
Florian: so use media queries.
fantasai: But for loading, best practice is to use media queries.
fantasai: You ignore the thing you can't render to your canvas.
fantasai: Visual canvas defined in css2.1
fantasai: and aural canvas is defined in css3 speech.
Rossen: Current proposed resolution, url() in 'content' only
applies to images and nothing else in visual canvas, and
to audio files in aural canvas.
RESOLVED: url() in 'content' only applies to images and nothing
else in visual canvas, and to audio files in aural
canvas.
SVG Integration
---------------
Florian: SVG Integration is dropped as a deliverable.
dbaron: This is becoming a compat issue because blink implements
it.
dbaron: Just a note that there will probably be more issues on
this spec, since it's relatively vague, and Blink
implements it, and we're seeing compat issues from not
implementing it.
Rossen: Improvements could be made.
<Rossen> https://github.com/w3c/csswg-drafts/issues/308
(discussion about editing these specs)
Rossen: unslotted
Florian: We just said we need SVG integration spec.
Florian: Hopefully the SVG WG will come back to life, if not we
should take it over.
TabAtkins: We should take it over.
Rossen: We should talk to the chair of the SVG WG.
Rossen: There are some WG restructuring that is happening with SVG
at the moment
Rossen: with a new charter since current charter proposal had 5
FOs.
Florian: I'm suggesting we send a message to the AC.
RESOLVED: Define everything in CSS.
Florian: If we have a functional SVG WG it should do it
Florian: if not the document should not die.
Rossen: any other topics
* tantek brings up first letter interactions
CSS 2.1
-------
Bert: CR for CSS 2.1+?
fantasai: We resolved to redirect CSS22 to the new note.
<tantek> Bert, uh no, the note and the CR are different
fantasai: We will publish the note.
fantasai: In its previous versions it will link to ...
fantasai: css22 shortname will link to note.
Bert: Do we also start the process updating css21?
fantasai: If any of the changes in the css22 have tests and 2+
interop passing implementations, then we should
incorporate those into a css21 update.
Snapshot
--------
Rossen: What happened to snapshot?
fantasai: fantasai didn't do anything for the snapshot.
Rossen: Is this something we can carve.
fantasai: Have higher priorities, like Grid & Alignment.
Rossen: Can someone else take over?
Florian: Snapshot has 3 editors.
Florian: Do we have a resolution of what we want in the snapshot?
Rossen: We had an email .... on the phone ...
Florian: That went into github
Florian: issue #?
<dbaron> https://github.com/w3c/csswg-drafts/issues/826
dbaron Issue is 826 for snapshot issues.
CSS 2.1
-------
dbaron: For 2.1 issues, what is the preferred feedback mechanism
for minor comments on the edits.
dbaron: github issues?
dbaron: I reviewed the changes section.
dbaron: I have some minor comments and one substantive comment
which was the encoding detection.
Bert: That sounds like I shouldn't publish it as a CR yet.
fantasai: We are not publishing as a CR we are turning it into a
note.
RESOLVED: publish css21-testing NOTE
Rossen: Anything else on 21 publishing?
Rossen: Back to snapshot.
Snapshot
--------
<Rossen> https://github.com/w3c/csswg-drafts/blob/master/snapshot-2017-modules.md
dbaron: I feel weird commenting on github when the document is not
there.
gsnedders: There are three, no four copies of css2.
Florian: dbaron made a short list of things to be added.
Rossen: so I'll read them (from
https://github.com/w3c/csswg-drafts/blob/master/snapshot-2017-modules.md
presumably)
<Rossen> css-align
<Rossen> css-grid
<Rossen> css-ruby
<Rossen> css-text
<Rossen> css-writing-modes
<Rossen> filters-1
fantasai: Ruby needs another update. It is super far behind on
incorporating Xidorn's feedback.
fantasai: Writing Modes is pretty stable and should be inlined
into the main section.
fantasai: Text needs to be republished but is fairly stable
fantasai: and goes with the transforms group.
dbaron: My comment was proposed edits to Chris's list
dbaron: ChrisL's only addition to the snapshot was Variables?
Florian: So you're (fantasai) proposing ... ?
Florian: How do people feel about CSS Break?
fantasai: We don't have much implementation.
Florian: But that's ok.
Rossen: We've been nibbling on this one.
fantasai: I'm ok either way.
Rossen: We need tests for this one.
Florian: In the beware section.
Rossen: so... CSS break.
Rossen: What else?
Florian: ... the last comment from Chris on the issue.
fantasai: I think TabAtkins was going to.
TabAtkins: What am I doing?
fantasai: Indexes.
TabAtkins: I can do it.
Florian: Do we have resolutions?
fantasai: Goal for snapshot is to only index properties in the
snapshot.
fantasai: Probably the snapshot should only include ...
Florian: Can we do that?
TabAtkins: You can filter it by spec.
Florian: For the record we should not block publication for this,
but the recent discussions for incubation may mean we
want to consider revising the best practice section. but
we should not block on this.
fantasai: I think it is fine, and people should be better about
talking about their plans and pressuring us to publish
things that ought to be published.
Florian: There is a discussion to be had there, let's have it some
other time.
Rossen: Are we ready to record resolution to add those at the link
and publish?
dbaron: So we're adding Writing Modes to ... ?
fantasai: Main one.
dbaron: Top section
dbaron: and Variables plus my list minus Ruby plus Break to which?
fantasai: Should Variables be in top level section?
dbaron: The top section the snapshot is defined by these modules.
dbaron: 2 rough interop, 3 stable but beware.
fantasai: Top level is for specs in CR and testing & impl
experience.
dauwhe: Can we have subheads?
fantasai: Yeah. Didn't before cuz the list was so short...
gregwhitworth: Is custom props in CR?
TabAtkins: Yes.
gregwhitworth: We need to go to the second one.
fantasai: There are two CR phases.
fantasai: To get into the top level you have to be in the second
CR phase
fantasai: where all we are expecting is minor bug fixes, mostly
on the implementation side, not significant changes
to the spec.
Florian: The rest of the ...
dbaron: Which things are we adding where?
Florian: Variables goes where?
gregwhitworth: 2 or 3
Rossen: 2
Florian: Variables 2 ok.
dbaron: I'm trying to make a note of this in the github issue.
dbaron: [reads a list]
dbaron: What I have: primary: Writing Modes, Rough interop:
Variables, completed design: Align, Grid, Text, Filters,
and Break
dbaron: Did I miss anything?
Florian: Maybe some section 3 can go in 2? not sure
Florian: Text?
fantasai: Main reason we had 2 vs 3 was stuff not in CR but still
had spec work to do but was deployed.
fantasai: 3 was spec is done, but not enough feedback to know how
stable the spec is.
fantasai: Possible spec is fine, or spec needs a lot of work, we
don't know.
fantasai: Text is ... 2 is fine
fantasai: Hopefully in a couple of months we can get into 3?
dbaron: [confused]
fantasai: 1 is second phase of CR.
dbaron: Do you want me to move Text to be with Transformations?
fantasai: Yes.
dbaron: Align Grid ... are in the lowest one
<dbaron> https://github.com/w3c/csswg-drafts/issues/826#issuecomment-272592572
RESOLVED: Make the additions to the snapshot listed here:
https://github.com/w3c/csswg-drafts/issues/826#issuecomment-272592572
Florian: The action I don't want is the indexes, the rest can be
on me.
Rossen: TabAtkins is doing the indexes.
ACTION: Florian commit the edits and publish
<trackbot> Created ACTION-826
Rossen: I think that is everything.
Florian: Is that 2.16 or 2.17?
Rossen: 2.16.1.
Rossen: We are done.
Florian: What is the shortname?
RESOLVED: Short name for the snapshot publication is "css-2017"
Rossen: We are adjourned. See you in Tokyo!<div
id="DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2"><br />
<table style="border-top: 1px solid #D3D4DE;">
<tr>
<td style="width: 55px; padding-top: 13px;"><a
href="https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon"
target="_blank"><img
src="https://ipmcdn.avast.com/images/icons/icon-envelope-tick-round-orange-animated-no-repeat-v1.gif"
width="46" height="29" style="width: 46px; height: 29px;" /></a></td>
<td style="width: 470px; padding-top: 12px; color: #41424e;
font-size: 13px; font-family: Arial, Helvetica, sans-serif;
line-height: 18px;">Virus-free. <a
href="https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link"
target="_blank" style="color: #4453ea;">www.avast.com</a>
</td>
</tr>
</table><a href="#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2" width="1"
height="1"></a></div>
Received on Tuesday, 14 February 2017 02:33:28 UTC