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

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

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Fri, 25 Jun 2010 10:53:03 -0700
Message-ID: <AANLkTinqa7Ko9To1DqgcBaYhXFjG1b9Vpjo-Cr_Og7cH@mail.gmail.com>
To: Simon Fraser <smfr@me.com>
Cc: Brad Kemper <brad.kemper@gmail.com>, fantasai <fantasai.lists@inkedblade.net>, Brian Manthos <brianman@microsoft.com>, Brendan Kenny <bckenny@gmail.com>, Sylvain Galineau <sylvaing@microsoft.com>, "robert@ocallahan.org" <robert@ocallahan.org>, "www-style@w3.org" <www-style@w3.org>
On Wed, Jun 23, 2010 at 6:20 PM, Simon Fraser <smfr@me.com> wrote:
> On Jun 23, 2010, at 6:14 PM, Tab Atkins Jr. wrote:
>
>> On Wed, Jun 23, 2010 at 6:13 PM, Brad Kemper <brad.kemper@gmail.com> wrote:
>>> I'd still be interested in knowing if people think that a "10px transition
>>> from 100% opaque to 0% opaque" should be inclusive of a 100% pixel and/or a
>>> 0% pixel within the 10px. I think here on the list we've be viewing it as
>>> exclusive of those endpoints.
>>
>> I consider it exclusive.  I expect a black shadow on white background
>> to not hit #fff or #000 until just *outside* of the blur area.
>
> The shadow is a Gaussian blur. Gaussian functions never hit zero, so in theory the shadow has infinite extent; in practice, implementations do cutoffs for efficiency, and each may choose a different cutoff. I don't think arguing over the specifics of which pixel is the last affected pixel is productive. (Note that if we defined shadows in terms of SVG filters, we wouldn't have to have this discussion.)
>
> We do need to agree on whether a non-offset, non-spread shadow extends by the blur amount or half the blur amount, but I don't think we have to pixel-precise on what "extends" means.

We do need to be precise if we want to be able to test the feature,
right?  I don't like features that are so implementation-defined as to
be untestable.

~TJ
Received on Friday, 25 June 2010 17:53:56 GMT

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