Re: [css3-images] simplifying radial gradients

On Fri, Oct 7, 2011 at 9:08 AM, L. David Baron <dbaron@dbaron.org> wrote:
> I think it would help other WG members (at the very least, this one)
> to follow and respond constructively to the discussion if you
> provided better links to things that you are referring to.  For
> example, in this message, you've referred a number of times to your
> proposal for simplifying radial gradients, yet you have not provided
> a link to it.  (Nor, in his reply, did Tab provide a link to his
> response.)

I apologize, David, you're totally right.  Here you are:

The original thread from Brad, where he argued for simplifying the
syntax: <http://lists.w3.org/Archives/Public/www-style/2011Sep/0019.html>

The email is fairly long, but it culminates in this suggestion for an
altered radial-gradient syntax:

radial-gradient(
 [circle,]? <color-stop>[, <color-stop>]+
)

Brad argues that the cut functionality can be emulated sufficiently by
the existing background properties, possibly augmented by some future
properties that have been discussed in the past.

I replied at <http://lists.w3.org/Archives/Public/www-style/2011Sep/0313.html>.
 Here, I go through his suggestions and find several of them
potentially valid, proposing a slightly less cut-down simplified
syntax.  Brad, Brian, and I talk back and forth about some of the
details of it.

Later, I wrote <http://lists.w3.org/Archives/Public/www-style/2011Sep/0562.html>,
explaining that after further consideration, I don't really want to
simplify the syntax at all.  I outline every category of
simplification that Brad suggested, and explain why I think it's
worthwhile to keep the current syntax.  The only thing without strong
justification in the syntax are the four side/corner sizing keywords
(I think 'cover' and 'contain' are pretty worthwhile).

~TJ

Received on Friday, 7 October 2011 22:31:39 UTC