Intent to ship: CSS Grid

[Mats Palmgren]

As of Gecko 52, I intend to let CSS Grid ride the trains on all platforms.
https://drafts.csswg.org/css-grid/
This also includes all parts of the CSS Box Alignment spec that are
relevant to Grid layout:
https://drafts.csswg.org/css-align-3/

It has been developed behind the "layout.css.grid.enabled" preference,
and has been enabled by default in Nightly and Aurora for some time now.

Other UAs shipping this or intending to ship it are:
Chrome - intends to ship:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/hBx1ffTS9CQ
Safari - is implementing it (independently):
https://bugs.webkit.org/show_bug.cgi?id=60731
Edge - status unknown to me

Note about the 'subgrid' feature:
We will *not* support the subgrid feature of CSS Grid in this first
release. This feature is marked as "at-risk" in the spec and thus
not needed for spec compliance. (Chrome is likewise not going to
support subgrid in their first release.)

This feature was previously discussed in this "intent to implement" thread:
https://groups.google.com/forum/#!msg/mozilla.dev.platform/jSmfRivZOrU/ZMPcwySUEW4J

Bug to turn on by default:
https://bugzilla.mozilla.org/show_bug.cgi?id=1217086

CSS Grid meta bug:
https://bugzilla.mozilla.org/show_bug.cgi?id=css-grid

The Devtools team is implementing specific tools for Grid:
https://bugzilla.mozilla.org/show_bug.cgi?id=1181227
(some of which will ship in 52 I'm told)


/Mats

[Simon Sapin]

On 16/11/16 01:40, Mats Palmgren wrote:
> Note about the 'subgrid' feature:
> We will *not* support the subgrid feature of CSS Grid in this first
> release. This feature is marked as "at-risk" in the spec and thus
> not needed for spec compliance. (Chrome is likewise not going to
> support subgrid in their first release.)

I don’t have a strong opinion on subgrid specifically, and other
arguments can be made, but I’d like to note:

It is marked at-risk in the spec *because* several vendors had said they
might not support it at first. So using this to justify not supporting
it is a circular argument.

--
Simon Sapin

[Jet Villegas]

Unlike other parts of the Grid spec, we don't have sufficient implementor
feedback (from ourselves or others) to evaluate the subgrid feature's
suitability for implementation. That will take time to get, and we do plan
to get it, but we want Grid available to people while we do so. That's not
the same as "nobody's implementing, so we won't either."

--Jet

On Wed, Nov 16, 2016 at 9:22 AM, Simon Sapin <simon...@exyr.org> wrote:

> On 16/11/16 01:40, Mats Palmgren wrote:
>

>> Note about the 'subgrid' feature:
>> We will *not* support the subgrid feature of CSS Grid in this first
>> release. This feature is marked as "at-risk" in the spec and thus
>> not needed for spec compliance. (Chrome is likewise not going to
>> support subgrid in their first release.)
>>
>

> I don’t have a strong opinion on subgrid specifically, and other arguments
> can be made, but I’d like to note:
>
> It is marked at-risk in the spec *because* several vendors had said they
> might not support it at first. So using this to justify not supporting it
> is a circular argument.
>
> --
> Simon Sapin
>

> _______________________________________________
> dev-platform mailing list
> dev-pl...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/dev-platform
>

[Tantek Çelik]

Agreed. In the perspective of the increasing shift toward incubation
at W3C (before WG and certainly before WD or CR), subgrid has yet to
be sufficiently incubated to really implement in the context of a
spec, unlike the rest of grid which has been prototyped with multiple
iterations (by other implementers) to get us where we are (hence it's
a great thing that we're shipping grid itself, well done Mats).

The circularity that Simon refers to can be broken by prototyping
subgrid separately, i.e. behind a feature flag, and iterating with
developer feedback, and getting that iteration in the spec, and then
taking that feature to an actual CR with actual shipping
implementations.

Unless by some lucky guess subgrid has been perfectly specified
without incubation, we should be ok with it being dropped from the
spec, or rather, pushed out into a future version.

Tantek

On Thu, Nov 17, 2016 at 1:19 AM, Jet Villegas <jvil...@mozilla.com> wrote:
> Unlike other parts of the Grid spec, we don't have sufficient implementor
> feedback (from ourselves or others) to evaluate the subgrid feature's
> suitability for implementation. That will take time to get, and we do plan
> to get it, but we want Grid available to people while we do so. That's not
> the same as "nobody's implementing, so we won't either."
>
> --Jet
>
> On Wed, Nov 16, 2016 at 9:22 AM, Simon Sapin <simon...@exyr.org> wrote:
>
>> On 16/11/16 01:40, Mats Palmgren wrote:
>>

>>> Note about the 'subgrid' feature:
>>> We will *not* support the subgrid feature of CSS Grid in this first
>>> release. This feature is marked as "at-risk" in the spec and thus
>>> not needed for spec compliance. (Chrome is likewise not going to
>>> support subgrid in their first release.)
>>>
>>

[Boris Zbarsky]

On 11/15/16 2:40 PM, Mats Palmgren wrote:
> As of Gecko 52, I intend to let CSS Grid ride the trains on all platforms.

Mats,

What performance testing have we done on this implementation? Have we
done edge-case stress-testing? Stress-testing with large but
"real-life" layouts? Something else?

This is getting a huge amount of developer interest, so we can expect
quite a bit of usage once we ship it; we should make sure it performs well.

-Boris

[Benjamin Smedberg]

Have the various grid properties be added to our fuzzers and we've been
fuzzing the implementation fairly comprehensively?

--BDS

[Jet Villegas]

+cc: abillings re: fuzzing CSS Grid.

On Monday, December 5, 2016, Benjamin Smedberg <benj...@smedbergs.us>
wrote:

> Have the various grid properties be added to our fuzzers and we've been
> fuzzing the implementation fairly comprehensively?
>
> --BDS
>
> On Tue, Nov 15, 2016 at 7:40 PM, Mats Palmgren <ma...@mozilla.com

> > _______________________________________________
> > dev-platform mailing list
> > dev-pl...@lists.mozilla.org <javascript:;>
> > https://lists.mozilla.org/listinfo/dev-platform
> >
> _______________________________________________
> dev-platform mailing list
> dev-pl...@lists.mozilla.org <javascript:;>
> https://lists.mozilla.org/listinfo/dev-platform
>

[Mats Palmgren]

On 12/05/2016 08:00 PM, Boris Zbarsky wrote:
> What performance testing have we done on this implementation? Have we
> done edge-case stress-testing? Stress-testing with large but "real-life"
> layouts? Something else?

Nope, I've done no /dedicated/ performance testing so far.
Both I and dholbert (who did the reviews) have been mindful of
potential performance issues throughout development though, fwiw.

Now that the implementation of features for release is complete,
I intend to do the kind of dedicated testing you're asking about.
And I invite everyone else with an interest in Grid to do the same.

There are two O(n^2) bugs that I'm aware of, but those only occur
in edge cases (fragmentation) so we decided they weren't release
blockers, since they won't affect viewing typical Grid layouts.
They will certainly be addressed in v53 though.


/Mats

[Mats Palmgren]

On 12/05/2016 09:53 PM, Benjamin Smedberg wrote:
> Have the various grid properties be added to our fuzzers and we've been
> fuzzing the implementation fairly comprehensively?

Yes, I believe Grid is included in our normal fuzz testing.
Grid has been enabled by default in Nightly and Aurora (including our
testing profile) for about a year now. I've seen a few bug reports
from fuzz testing in that time. There's also a fairly extensive reftest
suite for Grid which I think is fed into the fuzz testing and with
the usual DOM/CSS fuzzing on top of that I think fuzzing Grid layout
is well covered.


/Mats

[Mats Palmgren]

On 12/05/2016 09:53 PM, Benjamin Smedberg wrote:

> Have the various grid properties be added to our fuzzers and we've been
> fuzzing the implementation fairly comprehensively?