- From: Dael Jackson <daelcss@gmail.com>
- Date: Wed, 22 Aug 2018 15:21:56 -0400
- 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.
=========================================
CSS Grid
--------
- RESOLVED: Align grid layout static position as content box rather
than padding (Issue #3020)
Logical Properties
------------------
- RESOLVED: Drop section 5, backgrounds and borders, from logical
properties. (Issue #3028)
- RESOLVED: Accept proposal [Have values that affect alignment
compute in containing block. Values that affect
alignment of content of a box of an element are computed
in writing mode of element itself]. (Issue #3013)
- RESOLVED: The flow relative keyword values compute to themselves,
not to physical equivalents. (Issue #2821)
- RESOLVED: Publish a new WD of logical properties.
CSS Containment
---------------
- RESOLVED: Layout containment censors baselines and size
containment does not (Issue #2995)
Selectors 3
-----------
- RESOLVED: Publish Selectors 3 as PR
Fragmentation
-------------
- RESOLVED: Accept the changeset:
https://github.com/w3c/csswg-drafts/commit/23d5a4c7b28f9767f7f7a621a7b895d4b858bbca
[Change from an allow list to a disallow list for
break-inside] (Issue #1904)
- RESOLVED: Specify 0 size elements are positioned as early as
possible in the fragmentation flow (Issue #1529)
CSS Text Decor
--------------
- The people on the call leaned toward browsers defaulting to on by
default for text-decoration-skip-ink (Issue #2818), however some
individuals that had previously been against on by default
weren't available for the call so discussion will continue on
Github and this can be discussed on the next APAC timed call (5
Sept.)
CSS Overflow
------------
- RESOLVED: Rename overflow to block-ellipsis. Rename auto to none.
Rename ellipsis behavior to auto. (Issue #2561)
CSS Fonts
---------
- RESOLVED: No change to spec, leave piece of text out (Issue #2430)
===== FULL MINUTES BELOW =======
Agenda: https://lists.w3.org/Archives/Public/www-style/2018Aug/0021.html
Present:
Rachel Andrew
Rossen Atanassov
Tab Atkins
Dave Baron
Garrett Berg
Tantek Çelik
Dave Cramer
Alex Critchfield
Benjamin De Cock
Elika Etemad
Tony Graham
Dael Jackson
Brad Kemper
Chris Lilley
Peter Linss
Myles Maxfield
Anton Prowse
Melanie Richards
Jen Simmons
Alan Stearns
Lea Verou
Eric Willigers
Jeff Xu
Regrets:
Chris Harrelson
Florian Rivoal
Sean Voisen
CSS Grid
========
Static position should use content-box, not padding-box
-------------------------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/3020
Rossen: TabAtkins and fantasai noticed this while reviewing Grid and
they asked if we should s tick to this or align to
everything else. Opinions?
Rossen: TabAtkins is calling in
fantasai: I imagine he's in favor since we submitted this together.
<TabAtkins> ^_^
<dbaron> makes sense to me
Rossen: We can resolve to align grid layout static position as
content box rather then padding. Objections?
RESOLVED: Align grid layout static position as content box rather
than padding
Logical Properties
==================
Drop / defer border/background logical transforms
-------------------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/3028
fantasai: This is to drop or defer border and background section.
There's a bunch of issues against it which we can handle
but these were put in and sketched out on how to handle
it. Parts critical for implementing writing modes are not
this.
fantasai: We propose to defer this section so we can stabilize the
rest
fantasai: And address these in L2 if we think approach is good
<fantasai> https://drafts.csswg.org/css-logical/#background-and-borders
fantasai: Dropping all of section 5^
Rossen: In favor, but want to hear if others have opinions.
<dbaron> sounds good to me
<bradk> +1
Rossen: Objections to dropping section 5 from logical properties?
RESOLVED: Drop section 5, backgrounds and borders, from logical
properties
Should the mapping for logical values depend on the element or
containing block?
--------------------------------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/3013
fantasai: Question was about should the mapping for logical values
depend on element or containing block. Not about logical
properties, but values like float: inline-start or
text-align: start
fantasai: First thing is currently the spec...it was stated values
compute to themselves. That's first question. Second is do
you map against containing block or element.
fantasai: Supposed to depend on kind of element. text:align-start
needs to map against element. For alignment we use
containing block so makes sense to do same for float
because you don't want element to move based on content,
but rather context
fantasai: Proposed is no change to spec but clarify the values map
to themselves.
fantasai: I copied that in. Does anyone else have comment?
Rossen: Question. There are...all properties in box model, are they
defined to resolve against own writing mode?
fantasai: Yes own. It was sometimes correct and because you have to
do computation of writing mode with cascade it's much
easier computation.
fantasai: It's unfortunate that's the case because there's a lot of
cases to use containing block writing mode for something
like margins but we did element's own writing mode for
simplicity.
fantasai: Don't have same problem for values as they resolve at
layout time
fantasai: We should have done issue #2821 first as it talks about
computed
Rossen: Other opinions?
Rossen: I asked about box model properties because they are used for
aligning box and sometimes used to align content inside.
Align-self and content align properties and values are very
similar in layout impact so one doing self and the other
containing block is weird.
fantasai: align-self:start you use containing block,
align-self:self-start you use your own writing mode
fantasai: text-align:start your children is what's being aligned.
Rossen: Get it. Not making case against it. In same virtue you can
say position left and right can be used to align-self and
padding start and end can align content box. From that PoV
they are similar.
Rossen: For box model all properties and values compute to writing
mode of element itself
Rossen: There is discontinuity here.
Rossen: Can live with it.
Rossen: Other opinions or resolve on proposal to have values that
affect alignment compute in containing block. Values that
affect alignment of content of a box of an element are
computed in writing mode of element itself
fantasai: That's what's spec in grid, flexbox, and alignment
<bradk> +1
RESOLVED: Accept proposal
flow-relative values should say what their computed values are
--------------------------------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/2821
fantasai: This is that the values we spoke about compute to self,
not left or right.
fantasai: This is necessary for text-align property so for
consistency we should do that for all properties and if we
don't do that CSSOM physical coordinates won't align
dbaron: Flip side is that it means anybody looking at computed
values to act on them has to do something more complicated.
If a web page looks at computed value they need to consider
2x possibilities. They might not and therefore have bugs
fantasai: Yeah but you have to do for inherited properties why treat
non-inherited differently?
fantasai: You have to do that on inherited.
dbaron: Not as strong a case for not inherited ones
dbaron: and I think there are more of them, or maybe not
fantasai: Consistency and so author can work in logical coordinates
if they want to. Make these computed values be what they
are and if browser needs to worry it should add
convenience to its code before reporting to the user
dbaron: What about the CSSOM
Rossen: I think giving them all the values and having them make the
choice what to use would be better then the result of calc
that will mask what value ended up computed and trying to
piece that back to the value's origin. Especially for
inherited.
Rossen: I agree with dbaron it will require a bit more handling on
user side but probably not that much.
Rossen: We can always simplify later.
dbaron: Okay
Rossen: Other opinions or try to resolve?
fantasai: Also necessary if keeping previous resolution
Rossen: Yes, but we could revert.
Rossen: Objections to CSSOM exposes both logical and physical values
and the resulting values are that of the cascade?
Rossen: Is that the summary?
fantasai: What?...no. Resolution is that the flow relative keyword
values compute to themselves, not to physical equivalents
Rossen: Objections to this?
RESOLVED: The flow relative keyword values compute to themselves,
not to physical equivalents
Publication
-----------
Rossen: This was last logical issue
fantasai: There's 2 more, but that's what we wanted to resolve
before publishing
fantasai: Other 2 significant issues are marked in draft. #3030 and
#3029. I think those need more time to discuss and
publishing a draft with them marked as such is good way to
go forward
Rossen: Agree. Objections to publishing a new CR for Logical
properties?
fantasai: We're on WD.
RESOLVED: Publish a new WD of logical properties
Rossen: We'll try for CR soon
CSS Containment
===============
containment vs baseline alignment
---------------------------------
github: https://github.com/w3c/csswg-drafts/issues/2995
Rossen: TabAtkins can you do without florian? Or do we punt?
TabAtkins: We concluded and wanted to see if there are objections.
TabAtkins: Spec wasn't clear what effect containments have on
baseline. florian convinced me. Layout containment should
censor the baseline because it acts like there's nothing
going on inside
TabAtkins: Don't have to do layout on contents to layout parent.
Baseline requires you to know what's inside
TabAtkins: Size containment does not censor. Baseline doesn't affect
that size in any way. Sizing works fine and you can
baseline align
TabAtkins: Only potential weird is an overflow:visible element it
can shrink to 0 but still have a baseline. That can
happen today, though.
TabAtkins: Conclusion: layout containment censors baselines and size
containment no change. Other containments don't matter
here.
TabAtkins: Sound good?
Rossen: So this proposal only changes layout containment?
TabAtkins: Yes.
TabAtkins: Turns out in Chrome we don't pay attention to size but do
for layout. We should be able to change. Per spec this
only changes layout containment.
Rossen: Okay. Rest is clarification
Rossen: Reasonable. Seems florian agrees. Other comments?
<fantasai> +1
<fantasai> I think this is the correct resolution
Rossen: Objections to layout containment censors baselines and size
containment does not
RESOLVED: Layout containment censors baselines and size containment
does not
Selectors 3
===========
Selectors PR
------------
chris: Selectors 3 was...we produced new CR and we have 1 test of
the 1 errata. CR was just mandatory for patent policy. That
time has passed and there's no open exclusion. So can resolve
for PR.
chris: I see gsnedders asked about test suite. He's got a pull
request about build system not working for WPT. I don't see
relevance. This CR is one errata and making it normative. I
don't see this holds up PR and then REC. Selectors 4 is
current spec.
Rossen: Is gsnedders on?
Rossen: I didn't see regrets from him.
[seems like no]
chris: Do we move forward?
Rossen: I'm in favor. Want to hear from group.
fantasai: Makes sense
<astearns> +1 to PR from me
<tantek> +1
Rossen: Objections to publish Selectors 3 as PR?
RESOLVED: Publish Selectors 3 as PR
Rossen: chris do you need help?
chris: Fine. Done transition request except for resolution.
<chris> transition request here https://github.com/w3c/transitions/issues/83
Fragmentation
=============
Define behavior for replaced elements
-------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/1904
Rossen: Way replaced elements affected by break-inside properties
Rossen: Change we made with fantasai to go from a white list to a
black list of elements
Rossen: Currently break-inside is spec that it applies to elements
in normal flow that establish FC or are [list from changeset]
<fantasai> Changeset is
https://github.com/w3c/csswg-drafts/commit/23d5a4c7b28f9767f7f7a621a7b895d4b858bbca
Rossen: Currently excludes flexbox, grid, etc. Changed that to be a
blacklist
Rossen: Proposal is break-inside applies to all elements except
[list from issue]
Rossen: While we did this we noticed break-before and break-after
weren't applied to grid and flex so we fixed that too
Rossen: That's the summary. Want to hear from WG if we missed
anything or if we can resolve.
plinss: Can we stop using term black list and white list?
<TabAtkins> +1, blocklist and allowlist
Rossen: disallow list and allow list
* dauwhe plinss: +1
Rossen: Objections to accept the change:
https://github.com/w3c/csswg-drafts/commit/23d5a4c7b28f9767f7f7a621a7b895d4b858bbca
<bradk> Abspos includes fixed?
Rossen: Yes bradk it includes fixed
<bradk> Ok
RESOLVED: Accept the changeset:
https://github.com/w3c/csswg-drafts/commit/23d5a4c7b28f9767f7f7a621a7b895d4b858bbca
Empty fragment at fragmentainer boundary
----------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/1529
fantasai: About an empty fragment at a fragmentainer boundary. If
you have a 0 height block or 0 width inline etc and the
previous item that if they filled the whole line does the
0 size element stay on the page?
fantasai: Specs don't say anything now about where required to break
so UA can be intelligent.
fantasai: For 0 size might make sense to spec explicitly
fantasai: Question is do we want to spec that for fragmentation in
general or just for flexbox?
fantasai: flexbox is only place we define that precisely right now.
<fantasai> because flexbox requires consistency more than
quality-of-implementation, so its breaking rules are
defined
Rossen: I believe we discussed in the past. Don't remember if
resolved.
Rossen: As far as I recall reason is the empty elements are
exhausted opportunistically as much as possible. If there
are empty boxes at the end of fragment we assume they fit.
Done so you can reduce subsequent fragments.
Rossen: That makes sense.
Rossen: Other side is such elements or boxes are used for
positioning or to create containing blocks or abspos items
that go in and out for UI
Rossen: For those cases harder to argue it's better that such items
are consumed asap or pushed. Again, counter is there are
avoid break-inside and break-after properties where you can
use such and control correctly
Rossen: In both cases makes sense to position and consume empty
boxes as early as possible on fragmentation where they are
encountered rather then pushing
Rossen: That's how I remember previous
Rossen: Curious if there are other arguments or if we can resolve on
that and spec it so we don't forget again.
Rossen: Objections to Spec 0 size elements are positioned as early
as possible in the fragmentation flow?
<bradk> +1
RESOLVED: Specify 0 size elements are positioned as early as
possible in the fragmentation flow.
Publication
-----------
fantasai: Need to do DoC first
fantasai: Look next week?
Rossen: Fine.
CSS Text Decor
==============
Consider adding a third value (skip?) for text-decoration-skip-ink
------------------------------------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/2818
Rossen: astearns you added this from the F2F. I wanted to hear from
Emilio or myles or anyone involved
myles: This is asking for a distinction between off/on/do what
platform does. For us on and do platform is the same.
Proposal asks for explicit on. I wanted to know what other
vendors thought about skipping on by default
fantasai: I don't think it's by default. Request is a value, doesn't
mean we change initial. Initial is auto.
Rossen: Do we need auto?
fantasai: Yes. Wanted to allow UA to do what it felt like.
myles: Cool if all browsers decided default was do skipping so we
could have 2.
Rossen: Our position is we just finished re-writing inline layout.
ink-skipping was on my shortlist, but that shortlist wasn't
short so we didn't get to it. As a targeted behavior we'd
want on by default once we have it. Just like we enable
kerning by default. Doable, not super concerning for
performance.
myles: Sounds like you're okay with 2 values
Rossen: Yes
myles: We are also happy with 2
Rossen: Having said that there's no way to force it. If platform
supports but if it decides on that device it disables you
can't force it.
myles: When you produce a product that can have that behavior we can
re-open?
Rossen: Fine.
Rossen: Is Emilio on?
dbaron: I don't think xidorn wanted it on for all
myles: Do you remember reasons?
dbaron: I think some was related to what he saw as default on
platform. Maybe windows primarily.
dbaron: Don't remember that well. Underlying was xidorn wasn't
comfortable with on by default for all.
myles: Maybe let this go into issue and ask him to comment in issue?
fantasai: Yeah. Can also take up on Sept 5 call.
myles: Sounds good
Rossen: Sounds reasonable.
Rossen: Let's stop here and move on.
CSS Overflow
============
'overflow-block' and 'block-overflow' are too similar
-----------------------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/2561
Rossen: florian is not here. Dunno if heycam or rachelandrew are here
fantasai: I'm in favor of accepting proposal in last comment fwiw
<fantasai> https://github.com/w3c/csswg-drafts/issues/2561#issuecomment-413144634
Rossen: Other opinions?
rachelandrew: I'd agree
Rossen: Rename overflow to block-ellipsis. Rename auto to none.
Rename ellipsis behavior to auto
<astearns> +1 to block-ellipsis
rachelandrew: Yes I agree [missed]
rachelandrew: I agree and as I mentioned in comments I wrote docs
for MDN and this seemed sensible way to explain it.
Rossen: We can try and resolve. Objections to rename overflow to
block-ellipsis. Rename auto to none. Rename ellipsis
behavior to auto
RESOLVED: Rename overflow to block-ellipsis. Rename auto to none.
Rename ellipsis behavior to auto.
CSS Fonts
=========
font-size: 'medium' value is the user's preferred font size
-----------------------------------------------------------
github: https://github.com/w3c/csswg-drafts/issues/2430
chris: It seems...I thought this affected implementations and
rereading it looks like this should be the user's default.
Apparently it already does impact medium. It's restoring
language that was in the spec and was removed. Thought it was
a change, but I no longer object. We can put the language
back. I'd like to hear if other vendors see a problem.
Otherwise trivial change.
myles: One thought. Proposal has changed through issue. In Apr 25
post from Amelia it says [reads
https://github.com/w3c/csswg-drafts/issues/2430#issuecomment-384406289
].
I think that's more specific than original text. Helps me
understand what text trying to say. If we make a change,
should be one Amelia proposed.
Rossen: You're okay with changes?
myles: Yeah.
dbaron: One thing we've found in past is if you change the meaning
of medium and default font size meaning many web pages
break. Turns out to not be a good thing rather then zooming.
Maybe that's changing as web changes
myles: One detail, we don't support changing meaning of medium, but
do support bumping every font size by a percentage. No way to
affect keywords without affecting font-sizes and that works
well
chris: I think that was the basis for my original reluctance. We
seemed to hope that would work in CSS 1 and 2, but now you
can zoom entire page and that seems more robust and what
people do. didn't want to restore on basis of it was in CSS2.
myles: Question for dbaron. You said if you change definition of
medium to other then 16 web pages break, but someone in the
issue said chrome and firefox allow changing definition of
medium
dbaron: We do have a setting and over time it has gotten more
hidden, but haven't removed. It's not a great idea to set.
<astearns> I have changed that setting and have had pages break
<chris> so we shouldn't really encourage changing it, then
Rossen: What does that leave for this issue?
myles: This is a natural pull between browsers custom a11y that they
do without spec changes vs something normative the spec can
say about how to improve font sizes. dbaron's comment on how
browsers evolved their solution and maybe normative spec text
isn't necessary makes sense. Could go either way
chris: Could go either as well. Don't want to encourage people to do
a bad practice
Rossen: Leaning resolve no change?
<tantek> I don't think we have enough data to justify a specific
change on this
<chris> fine with no change here if WG so resolves
Rossen: Going to take silence as agree
Rossen: Objections to no change to spec, leave piece of text out
<tantek> +1
<chris> +1
RESOLVED: No change to spec, leave piece of text out.
Rossen: We're at top of hour, one more font issue from fantasai
Rossen: Happy to push if can't resolve in a minute.
Rossen: Okay, we are done
Rossen: We'll chat next week. Bye!
Received on Wednesday, 22 August 2018 19:22:54 UTC