RE: Proposal: will-change property (formerly will-animate)

No, we’re reading the same things.  I’m just suggesting that we need another sentence in the ‘will-change’ spec along similar lines:
If any non-initial value of a property would make the element a containing block for fixed-positioned descendants, then specifying that property in “will-change” must make the element a containing block for fixed-positioned descendants.

I don’t see how any language currently in the ‘will-change’ spec implies this statement.

Sorry for being insufficiently explicit before,
               Pavel

From: James Robinson [mailto:jamesr@google.com]
Sent: Tuesday, January 21, 2014 2:15 PM
To: Pavel Curtis
Cc: Tab Atkins Jr.; Robert O'Callahan; Benoit Girard; www-style
Subject: Re: Proposal: will-change property (formerly will-animate)

From Benoit's email:

"Stacking context behavior:
will-change will induce a stacking context if the non-initial value of any of the specified <custom-ident> match a CSS property induces a stacking context."

A non-initial value of the ident "transform" induces a stacking context.  The same text exists in Tab's draft.  Maybe we're reading different text?

- James

On Tue, Jan 21, 2014 at 11:01 AM, Pavel Curtis <pavelc@microsoft.com<mailto:pavelc@microsoft.com>> wrote:
Yes, but the ‘will-change’ spec doesn’t call out that it also has this effect when its value contains ‘transform’.

From: James Robinson [mailto:jamesr@google.com<mailto:jamesr@google.com>]
Sent: Tuesday, January 21, 2014 10:58 AM
To: Pavel Curtis
Cc: Tab Atkins Jr.; Robert O'Callahan; Benoit Girard; www-style

Subject: Re: Proposal: will-change property (formerly will-animate)

http://dev.w3.org/csswg/css-transforms/#transform-rendering


"For elements whose layout is governed by the CSS box model, any value other than none for the transform results in the creation of both a stacking context and a containing block. The object acts as a containing block for fixed positioned descendants."

- James

On Mon, Jan 20, 2014 at 11:04 AM, Pavel Curtis <pavelc@microsoft.com<mailto:pavelc@microsoft.com>> wrote:
I don't see any language in the spec that implies this behavior...

        Pavel

-----Original Message-----
From: Tab Atkins Jr. [mailto:jackalmage@gmail.com<mailto:jackalmage@gmail.com>]
Sent: Friday, January 17, 2014 6:15 PM
To: Pavel Curtis
Cc: Robert O'Callahan; Benoit Girard; www-style
Subject: Re: Proposal: will-change property (formerly will-animate)
On Fri, Jan 17, 2014 at 5:37 PM, Pavel Curtis <pavelc@microsoft.com<mailto:pavelc@microsoft.com>> wrote:
> Does the declaration
>         will-change: transform
> also cause the element to become a containing block for fixed-positioned descendants?  I would hope so.

Yes, since 'transform' generates a containing block for some of its values, it generates a containing block when specified in 'will-change'.

~TJ