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

My use case, just for background: Containment could be a huge win for web
apps with left navs (Gmail <http://screencast.com/t/PPjvic0bpaV>,
Feedly<http://screencast.com/t/PPjvic0bpaV>,
Drive, etc) or content-heavy sidebars
(Facebook<http://screencast.com/t/NS0pr49pQR>).
Content is often dynamically updated/inserted into these spaces (above and
below their various folds) which an author could likely "guarantee" doesn't
have any effect on the main page.



On Tue, Dec 3, 2013 at 8:15 AM, Tab Atkins Jr. <jackalmage@gmail.com> wrote:

> On Mon, Dec 2, 2013 at 7:36 PM, Robert O'Callahan <robert@ocallahan.org>
> wrote:
> > On Tue, Dec 3, 2013 at 3:38 PM, Tab Atkins Jr. <jackalmage@gmail.com>
> wrote:
> >>
> >> Yup - if it's possible to scroll, then we have to (a) pay attention to
> >> the elements "below the fold", even if they'd otherwise be ignorable,
> >> and (b) paint at least some of the off-screen stuff, so that it'll be
> >> smooth if you start scrolling.
> >
> > That's only true for elements that are themselves visible, or nearly
> > visible. I think we could drop this requirement.
>
> Yes, but still.  The scrolling restrictions come from requests on our
> end.  ccing Ojan for an elaboration on the reasoning.
>
> > I agree with Simon that the text is ambiguous. Instead of "An element
> that
> > is strictly contained operates under the following restrictions:", I
> would
> > say
> >>
> >> An element that is strictly contained has the following restrictions
> >> applied to it by the user-agent:
> >> 1. The contents of the element are clipped to the element’s content box.
> >
> > etc
>
> Sure.
>
> > In part 1 you should be more clear about "contents". Presumably the
> contents
> > of an element don't include its border, for example, but this is unclear.
>
> I'm not sure how to more clearly talk about contents.
>
> > I believe restrictions 2 and 3 should be dropped.
>
> Let's let some of the engineers that asked me for this restriction to
> chime in first.
>
> ~TJ
>