Re: [css-compositing] why general blending is confusing

On 12 Oct 2013, at 7:00 pm, Robert O'Callahan <robert@ocallahan.org> wrote:

> On Fri, Oct 11, 2013 at 9:00 PM, Dean Jackson <dino@apple.com> wrote:
> Notice that in 4, the child does NOT blend outside its parents bounds. That’s
> because it doesn’t have any background pixels to blend with at that point. Sure, the
> translateZ(0) forced a stacking context (compositing layer) on the parent, but
> this could have happened for a number of reasons - e.g. a sibling of the parent
> having a hover effect that triggered compositing.
> 
> I’m not sure there are any good ways around this without accumulating buffers
> (potentially prohibitively expensive), and even then things like animations
> and media will make it harder.
> 
> I'm not sure what problem you're pointing to. Are you saying the spec is unclear

No.

> , or that it's too hard to implement in some situations (which situations?)

It will depend on what the specification requires, but it will be hard to implement
something that produces the same output for examples 3 and 4 in my test.

> , or that it's too complicated for authors to understand

Mostly this, yes. I don’t think an author will understand when layers are created,
and why behaviour changes dramatically when they are.

However, if the WG are happy with this, and we can describe all of the
complicated ways a page may be composited (in a cross browser way),
and it is something we can all implement easily, then I’ll be happy. I really
like this feature and would love to see it in engines.

Dean