Re: [css-containment] ED of Containment ready for review (was overflow:clip)

On Fri, Nov 8, 2013 at 4:03 PM, Simon Fraser <smfr@me.com> wrote:
> On Nov 8, 2013, at 2:33 pm, Tab Atkins Jr. <jackalmage@gmail.com> wrote:
>> The ED for the 'contain' property is now ready for review and possible
>> WD publication: <http://dev.w3.org/csswg/css-containment/>.
>>
>> This is the property that originated as my suggestion for "overflow:
>> clip", for stronger layout isolation.  As discussed on this list, it
>> grew a bit to encompass more than just overflowing, so that using the
>> property actually isolates the element's subtree *completely*
>>
>> Please review and let me know if I missed anything, or if there's
>> anything that's not sane there.
>>
>> If you have good suggestions for how to split the containment up into
>> a few pieces, and good justification for why you'd want this
>> (preferably pointing to something real-world that could benefit from
>> being isolated but needs to violate some of the assumptions of strict
>> containment), we can make the property alternately accept a few
>> feature keywords that turn on isolation per-feature.
>>
>> I'm aiming for a WD request in the next week or so, so review soon is
>> appreciated!
>
> It’s unclear to me if “contain: strict” is a hint from the web author to the UA
> to say that the set of conditions given in the spec are true, or whether
> it actually causes all those conditions to be enforced.
>
> I suspect that it’s the former, but does this allow the UA to just render
> incorrectly if the author says “contain: strict” but the conditions are not met?
> That’s pretty weaksauce.

There's a whole bunch of "must"s in there, saying exactly what the UA
has to do, and at least a few of the restrictions don't make any sense
as self-imposed restrictions.  I'm not sure how to make it clearer
that this property causes behavior changes.

~TJ

Received on Sunday, 10 November 2013 17:06:44 UTC