RE: [css4-color] #RGBA

I don't have the link (and didn't see it in the original mail), but I'm referring to the extension of adding the #rrggbbaa syntax (CSS4) *not* the extension of adding rgba (CSS3).

If you add a "non-functional" syntax to RGB, and don't add it to HSL that suggests that HSL isn't worthy of spending the time to provide a tight non-functional syntax so that it has parity with RGB.

So with the CSS4 change, RGB is now two steps ahead of HSL -- it has rgb(), rgba(), #rrggbb, and #rrggbbaa whereas HSL only has hsl() and hsla().

-Brian

-----Original Message-----
From: Eduard Pascual [mailto:herenvardo@gmail.com] 
Sent: Friday, September 10, 2010 11:45 AM
To: Brian Manthos
Cc: Patrick Garies; mollyh@opera.com; Tab Atkins Jr.; Christoph Päper; www-style@w3.org list
Subject: Re: [css4-color] #RGBA

On Fri, Sep 10, 2010 at 8:37 PM, Brian Manthos <brianman@microsoft.com> wrote:
>
> Note: I didn't really suggest anything on this front.
>
> I was and am curious if HSL is becoming a first class citizen in CSS or taking a back seat to RGB.  When RGB is extended and HSL is not, it suggests that RGB is primary and HSL is secondary.

Actually, the extension that is being added to RGB (the alpha channel)
was already included for HSL in Colors3. So this is bringing RGB back
to "first class citizenship" after it had slipped a bit behind for a
while.

Regards,
Eduard Pascual

>
> Also, in the gradient discussions as of late there's been talk of RGB (premultiplied and not) but HSL hasn't been discussed much.  When interpolating HSL and RGB, you tend to get different results with "simple" algorithms so I was kind of surprised not to see HSL enter the discussion.
>
> -Brian
>
> -----Original Message-----
> From: Patrick Garies [mailto:w3c.www-style@patrick.garies.name]
> Sent: Thursday, September 09, 2010 7:37 PM
> To: mollyh@opera.com
> Cc: Tab Atkins Jr.; Brian Manthos; Christoph Päper; www-style@w3.org list
> Subject: Re: [css4-color] #RGBA
>
> On 2010-09-06 1:38 PM, Molly E. Holzschlag wrote:
> > So even if there were a way to shorthand HSL it doesn't make sense from
> > a design standpoint to do so. It's a shorthand in and of itself.
>
> While I don’t think an |#RGB|‐ or |#RRGGBB|‐like syntax is a good idea
> for HSL (as Brian apparently suggested), I can see a good reason from a
> “design standpoint” to abbreviate the HSL syntax further. As someone who
> uses all‐gray or nearly all‐gray palettes, I find it to be somewhat
> tedious and messy to have to use |hsl(<H>, <S>, <L>)| when the first two
> arguments are completely irrelevant.
>
> Something like |hsl(<L>)| or |l(<L>)| where the hue and saturation are
> assumed to be zero (and which I’ve previously proposed as |rgb(<RGB>)|
> for the same reasons) would make this easier.
>

Received on Friday, 10 September 2010 19:06:52 UTC