- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Thu, 20 Aug 2009 15:30:37 -0500
- To: David Perrell <davidp@hpaa.com>
- Cc: www-style@w3.org
On Thu, Aug 20, 2009 at 12:47 PM, David Perrell<davidp@hpaa.com> wrote: > Tab Atkins Jr. wrote: > | linear-gradient(left / green 30px, wheat 20%, mirror) > > The mirror idea sounds good, but I don't like this form. 'Mirror' isn't a color stop. It would make more sense to me if 'mirror' is a keyword in the first part: > > linear-gradient(left mirror / green 30px, wheat 40%) > > where 'mirror' divides the gradient length in two and percentages refer to half the full length of the gradient (100% = 50% of the gradient length). This would degrade symmetrically. Alternate idea: mirror-gradient(). Other possibilities: mirrored-gradient(), half-gradient(), semilinear-gradient(), half-linear-gradient() Benefit of splitting it out as a new type of gradient is that we can eliminate parts of the syntax that don't make sense, like the angle construction, and the second bg-position. We'd just say that 0% is at the specified point, and 100% is the center of the box. Then the colors get reversed on the far side. > | > Actually, neither color-stop positioning solution is > | satisfactory inasmuch as the implicit position of color-stops > | (subdividing the span between explicit color stops) doesn't jibe > | with either. A default value is normally one of several > | specifiable options. > | > | Not sure what you're saying here. > > First and last color stop positions are specified as 0% and 100% by default. Default position for other color stops is a percentage of the space between stops with specified position. So, I'm thinking of a hierarchy of position types: percent <- relative length <- pixels <- default. Relative lengths refer to spans within specified percentages, pixel lengths to spans within specified percentages or relative lengths, default lengths within any specified lengths. > > Make sense? > > (BTW, since computed percentages for stops without specified positions are percentages of a span *between stops*, having percentages in general refer to spans between <length> stops is not totally inconsistent.) All right, makes sense. Still way too complicated, I think. ^_^ ~TJ
Received on Thursday, 20 August 2009 20:33:15 UTC