W3C home > Mailing lists > Public > www-style@w3.org > February 2009

Re: [CSSWG] Minutes and Resolutions 2009-02-04: box-shadow and border-image

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Thu, 5 Feb 2009 10:55:48 -0600
Message-ID: <dd0fbad0902050855q285c59ecw449cb5cf325bd7c4@mail.gmail.com>
To: Brad Kemper <brad.kemper@gmail.com>
Cc: fantasai <fantasai.lists@inkedblade.net>, www-style@w3.org

On Thu, Feb 5, 2009 at 10:31 AM, Brad Kemper <brad.kemper@gmail.com> wrote:
>
>
> On Feb 4, 2009, at 4:47 PM, fantasai wrote:
>
>>  RESOLVED: box-shadow is not suppressed by border-image
>>  RATIONALE: It's useful in many cases, and the author can suppress
>>            it himself as needed.
>
> Can this be reopened? I strongly disagree with the reasoning above, and the
> minutes do not mention the arguments I gave in other e-mails[1], and ask for
> a test case but don't mention the one I provided[2]. My main points:
>
> 1. Having box-shadow and border-image visible at the same time is NOT
> useful. Those who are creating the images for border-image can very easily
> create them with shadows already included in the images[3]. Created that
> way, they can follow the complex contours of the raster border in whatever
> fashion the artist chooses. There is no reason to have the UA do it instead
> at that point. In most all cases, except only where the images all have
> straight edges and square or round corners that all exactly align with the
> border-box, the UA generated shadow will not be in the right place.
>
> 2. The PRIMARY reason an author would specify both is for fallback reasons.
> Some UAs may support box-shadow without supporting border-image, or the user
> may have turned off images in their browser (in which case border-image
> should really be ignored too) or may have set a user style sheet to
> 'border-image: none !important' (because they really hate loading images
> that are only used in borders). In those cases, box-shadow provides a good
> fallback. BUT... only if the UA does not render both at the same time.
>
> So, having both box-shadow and border-image _visible_ at the same time is
> NOT useful in many cases, and the author CANNOT suppress it himself in the
> case in which having these two properties in the same rule would be most
> useful.

I hadn't gotten around to responding to the original email yet, but I
was planning on saying the same things as Brad here.

Box-shadow is only necessary when border-image is not being used.  It
can easily be *harmful* when border-image *is* being used (and at best
is superfluous).  The note that authors "can suppress it [themself] as
needed" is incorrect, because there is a legitimate reason to
*specify* both (fallback behavior, to provide a shadow when the
border-image isn't being used, for whatever reason), but almost never
a reason to *display* both.

~TJ
Received on Thursday, 5 February 2009 16:56:25 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:16 GMT