W3C home > Mailing lists > Public > www-style@w3.org > August 2009

Re: Gradient syntax proposal

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Wed, 19 Aug 2009 14:04:45 -0500
Message-ID: <dd0fbad0908191204j5cd0f80ay16ebd4710fcf9d89@mail.gmail.com>
To: David Perrell <davidp@hpaa.com>
Cc: www-style@w3.org
On Wed, Aug 19, 2009 at 1:50 PM, David Perrell<davidp@hpaa.com> wrote:
> Tab Atkins Jr. wrote:
> | Obviously you *can* do that with two gradients, but it might be useful
> | to make that just part of the normal syntax.  That would eliminate
> | completely the primary cause of misordered points, and might indeed be
> | better in general.  Hrm.
> |
> | I'll think on this.  In the meantime, any *more* examples?
>
> Maybe. I have one in mind comparing degradation.

Kk.  Note that I talked again about that image earlier this morning.

> I agree a slash shouldn't be doing double duty on one line. A single calc() suffices to set a point from end-of-gradient and would typically be the only one needed. Other lengths would be relative to that (and, if percentages applied to the span between lengths, percentages would be relative to that as well).

Agreed, but even with a different separator I don't like it.  It's not
a separator at all, but a toggle switch that says "after this, all
measures are from the end point instead".

I find it difficult to justify making a % mean two different things on
two different sides of a slash when they are used on the *same
construction*.  That seems a lot worse than having % refer to box
dimensions when detailing a point at the beginning of the rule, but
refer to distance from the starting-point later in the color-stops.

> Please, no 'end()'!!

Heh, ok.  Just an idea.  I didn't like it much either - I'm fine with calc().

~TJ
Received on Wednesday, 19 August 2009 19:05:48 GMT

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