Re: [CSSWG] Minutes and Resolutions 2009-08-12

Sent from my iPhone

On Aug 13, 2009, at 2:45 PM, "Robert O'Callahan"  
<robert@ocallahan.org> wrote:

> On Fri, Aug 14, 2009 at 9:34 AM, Brad Kemper <brad.kemper@gmail.com>  
> wrote:
> On Aug 13, 2009, at 1:43 PM, "Robert O'Callahan"  
> <robert@ocallahan.org> wrote:
>
> Being able to specify an angle is reasonable. However, if you want a  
> linear gradient to go from one corner of the box to another, for  
> example, and you don't know the size of the box, then you can't use  
> an angle, or if you do know the size of the box, then you have to do  
> some trigonometry.
>
> I don't see how that is true. The corner is implied by the angle,  
> and once you have that it is the same math as defining a stop from  
> keywords like "top left" or "bottom right" or whatever.
>
> Suppose you don't know the width or height of the box, and you want  
> the top-left to have the start color and the bottom bottom-right to  
> have the end color.
>
> If you want the angle of the gradient to be 45deg, then you need to  
> specify an angle, you can't do it with start and end positions.
>
> If you want the angle of the gradient to follow the line from top- 
> left to bottom-right, then you can't use a fixed angle.
>
> So I think we need both.

I see what you're saying now. I agree.