W3C home > Mailing lists > Public > www-style@w3.org > June 2010

RE: [css3-background] vastly different takes on "blur"

From: Brian Manthos <brianman@microsoft.com>
Date: Wed, 23 Jun 2010 20:41:53 +0000
To: fantasai <fantasai.lists@inkedblade.net>, Brad Kemper <brad.kemper@gmail.com>
CC: Brendan Kenny <bckenny@gmail.com>, Sylvain Galineau <sylvaing@microsoft.com>, Simon Fraser <smfr@me.com>, Tab Atkins Jr. <jackalmage@gmail.com>, "robert@ocallahan.org" <robert@ocallahan.org>, "www-style@w3.org" <www-style@w3.org>
Message-ID: <FA122FEC823D524CB516E4E0374D9DCF0158D5DB@TK5EX14MBXC132.redmond.corp.microsoft.com>
> > Agreed. I was thinking to include a note about that, and then ask a
> > question such as "which of the following renderings best describe a
> > blur of 10px (and don't cheat to see what you browser does)", and
> > include some variations that are all based on a black shadow (for
> > widest range of perceptiple changes to transparent). The variations
> > would include not only one version being twice the width of the other,
> > but also the possibility of clipping to within some range that (1% -
> > 99%?) that a human can discriminate as different from totally opaque
> > and totally transparent (and filling the 10px with that), and also
> > testing whether or not people consider a pixel of 100% and/or 0% to be
> > part of the countable part or not.
> 
> I feel that these other questions are much more easily resolved here, and I'd
> rather keep the question as simple as possible for the one issue where we
> really would benefit from feedback. Getting coherent answers from the web
> design community is difficult enough as it is.
> 
> ~fantasai

But that completely misses a critical point.

With all renderings I've seen so far, the mathematical impact of the blur and the human perceptible impact of the blur are significantly different.

If these two rules
	div {
		box-shadow: 200px 200px 10000000000000000px;
	}
	div {
		box-shadow: 200px 200px 200px;
	}
appear the same to human perception in a UA and that UA is considered compliant then I'd argue the feature is effectively untestable.

Taken to the extreme, a UA that renders
	div {
		box-shadow: 10px 10px 1px;
	}
	div {
		box-shadow: 10px 10px 10000px;
	}
identically is compliant.

Thus it becomes "UA-specific blur parameter" which does a disservice to interoperability.

-Brian
Received on Wednesday, 23 June 2010 20:42:39 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:28 GMT