- From: Brian Manthos <brianman@microsoft.com>
- Date: Sat, 17 Jul 2010 20:22:25 +0000
- To: Brad Kemper <brad.kemper@gmail.com>, Tab Atkins Jr. <jackalmage@gmail.com>
- CC: Aryeh Gregor <Simetrical+w3c@gmail.com>, "L. David Baron" <dbaron@dbaron.org>, Simon Fraser <smfr@me.com>, Brendan Kenny <bckenny@gmail.com>, www-style list <www-style@w3.org>
On Jul 17, 2010, at 7:54 AM, Brad Kemper wrote: >>On Jul 16, 2010, at 3:30 PM, Tab Atkins Jr. wrote: >> On Fri, Jul 16, 2010 at 3:14 PM, Aryeh Gregor <Simetrical+w3c@gmail.com> wrote: >>> On Thu, Jul 15, 2010 at 5:41 PM, Tab Atkins Jr. <jackalmage@gmail.com> wrote: >>>> ?_? That sounds plenty testable to me. Grab the pixels, verify they >>>> form a gradient, then check where the 99% point is hit. Sounds easy >>>> to me. >>> >>> What algorithm do you propose to "verify they form a gradient"? Other >>> than specifying a canonical type of gradient (say, Gaussian) and just >>> comparing against that? >> >> By "forms a gradient" I mean "creates a monotonic transition from one >> color to another". That's easy to verify by just walking the pixels. > >More accurately, it is from one _opacity_ (the shadow color's Alpha component) to another (transparent). It does look like 'one color to another' if it >is a single shadow against a solid-color background, with no other elements overlapping it. But that isn't always the case. Is it also required that the algorithm be symmetric? Must the rates of change be consistent in the X and Y dimensions? Further, is it required the algorithm be consistent in the positive and negative directions. If I put two boxes next to each other horizontally that use the same blur, should they produce a consistent effect on the underlying background along the linear part of the shared left/right border (because combine they provide a consistent opacity)? -Brian
Received on Saturday, 17 July 2010 20:23:06 UTC