Re: [css3-images] 2011/12/01 ED section 4.2 review notes

On Wed, Dec 7, 2011 at 10:37 AM, Brad Kemper <brad.kemper@gmail.com> wrote:
> On Dec 7, 2011, at 10:15 AM, Tab Atkins Jr. wrote:
>> On Wed, Dec 7, 2011 at 10:09 AM, Brad Kemper <brad.kemper@gmail.com> wrote:
>>> Aside from disagreeing that about allowing less-readable order, I still have a problem with having to write out 'ellipse' much more often. I used to be able to write this:
>>>
>>> radial-gradient(closest-corner, yellow, blue)
>>>
>>> But now I have to be longer and more explicit, like this:
>>>
>>> radial-gradient(ellipse to closest-corner, yellow, blue)
>>>
>>> That is a big downgrade, IMO. It could be solved with this:
>>>
>>> radial-gradient(
>>> [
>>>   [ circle?  <length>? ] | [ ellipse?  <length>{2}? ] |
>>>   [ [ circle | ellipse ] to]? <closest/farthest-corner-side>
>>> ]?
>>> [ at <position ]?
>>> , <color-stops>)
>>>
>>> I'd rather this be settled now, before last call, than to be objecting during last call, or waiting until there are more implementations using yet another syntax that could change again.
>>
>> Huh?  I don't understand.  That's not true at all.  There's no "to" in
>> the grammar, and you can always omit 'ellipse'.
>> "radial-gradient(closest-corner, yellow, blue)" is definitely valid
>> and does what you expect (the same thing as it did in the old WD
>> grammar).
>
> Well then, I really don't know where the latest grammar came from. I never discussed it. The last one I discussed amongst us four, before you presented it to the WG 4 days after Thanksgiving was from fantasai's e-mail two days after Thanksgiving, which had this:
>
>>> If we have 'at' disambiguating the <position>, we don't need a keyword
>>> specifically to disambiguate the size. So here's a suggestion to address
>>> that feedback:
>>>  - specify circle/ellipse keyword || explicit size OR
>>>  - specify circle/ellipse keyword with "to <keyword>"
>>>
>>> radial-gradient(
>>> [  [ circle || <length> ] | [ ellipse || <length>{2} ] |
>>>   [ circle | ellipse ] to <closest/farthest-corner-side> ]?
>>> [ at <position ]?
>>> , <color-stops>)
>>>
>>> You get:
>>>
>>>  radial-gradient(5em circle at 25% 25%, yellow, blue)
>>>  radial-gradient(ellipse 50% 2em at center, yellow, blue)
>>>  radial-gradient(circle to closest-corner, yellow, blue)
>
> That was the one I thought had some promise, but still needed a little work. So apparently, shortly after that, you came up with a grammar that was different, and presented it to the WG as something the four of us had agreed to that best addressed the comments of the survey. No wonder I am confused. The WG voted on something I wasn't even aware of, and which had not been adaquately discussed beforehand.

Do you hate the current grammar?  Do you see any mistakes or killer
problems with it?  It looks like it addresses your concerns, so what
is the problem?  It doesn't seem useful to be talking about the merits
of a grammar that you don't like and which doesn't exist in any
published draft.


>> (This wasn't true in the November ED grammar, either - you could write
>> "radial-gradient(to closest-corner, yellow, blue)".)
>
> What day in November was that?

I'm referring here about the grammar that we came up after TPAC, and
presented in the blog post.

~TJ

Received on Wednesday, 7 December 2011 19:07:14 UTC