On Jul 13, 2010, at 2:55 PM, Brendan Kenny wrote: > On Tue, Jul 13, 2010 at 3:21 PM, L. David Baron <dbaron@dbaron.org> wrote: >> On Tuesday 2010-07-13 07:15 -0700, Brad Kemper wrote: >>> Arguments in favor of the distance measuring the Entire blur region width (current spec language): >>> >>> 1) The entire perimeter is blurred, outer and inner, not just >>> outer, so it is logical that the width of the entire blur effect >>> width should match the authored value. >> >> However, for 'text-shadow' the value is called the "blur radius", >> not the "blur diameter" (and has been called "blur radius" since >> http://www.w3.org/TR/1998/REC-CSS2-19980512/text.html#text-shadow-props ). >> >> (I'm not sure when the definition of 'box-shadow' changed from using >> the commonly-used term "blur radius" to using the new term "blur >> distance".) >> >> I'd also note that blurring is implemented as a generic >> transformation of images; it's not just something applied to edges. >> In that form, I think measuring in terms of the radius >> (approximating the concept of how far away from its original >> location can the color of a point can get) makes more sense than >> using the diameter (approximating the distance between the two >> farthest points in opposite directions that the color of a pixel can >> reach). >> >> So I also favor switching the definition of box-shadow to match >> text-shadow and to describe a blur radius rather than a diameter. > > > "blur distance" was added very recently, I think just a few messages > before it was realized that the change would be an issue: > > http://dev.w3.org/cvsweb/csswg/css3-background/Overview.src.html.diff?r1=1.232&r2=1.233&f=h > > "blur radius" had been replaced with "blur value" the revision before that. The reason that it was changed from a radius is that Brad thought that it should be described, and defined in terms of its appearance, not in terms of the underlying implementation. However, I'm with David. Blurring is a common graphical technique which is always based on filtering via a gaussian blur, for which the blur radius is an input parameter. I see nothing wrong with calling it a 'radius' in the spec. SimonReceived on Tuesday, 13 July 2010 22:32:00 UTC
This archive was generated by hypermail 2.4.0 : Monday, 23 January 2023 02:13:48 UTC