W3C home > Mailing lists > Public > www-style@w3.org > June 2008

Re: Box-shadow : Why not follow the standardized OpenXML specification ?

From: Brad Kemper <brkemper@comcast.net>
Date: Wed, 18 Jun 2008 08:17:27 -0700
Cc: Frode Børli <frode@seria.no>, Francois Remy <fremycompany_pub@yahoo.fr>, Henrik Hansen <henrikb4@gmail.com>, CSS 3 W3C Group <www-style@w3.org>
Message-Id: <D8295D68-92CF-4841-92BC-959AD9C217DD@comcast.net>
To: Alan Gresley <alan@css-class.com>


On Jun 18, 2008, at 4:49 AM, Alan Gresley wrote:

>
> Brad Kemper wrote:
>> On Jun 17, 2008, at 1:03 PM, Frode Børli wrote:
>>>> Inner Shadow: Because it is so similar to box-shadow, and because  
>>>> it is a
>>>> type of box shadow, most would probably agree that it is better  
>>>> to add this
>>>> as a key word to box-shadow (or a sub-property of a
>>>> box-shadow-as-shorthand), than to create its own new property that
>>>> replicates most of what is already present in the box-shadow draft.
>>>
>>> How would we add both inner and outer shadow at the same time?
>> One div inside another, I suppose, But since inner shadow creates  
>> the illusion of a hole cut in something, and outer shadow creates  
>> the illusion of that shape floating above something instead, I  
>> really don't think there is going to be that much demand for both  
>> inner and outer shadows on the same rectangle. Its much simpler to  
>> have a single key word on a single compound property.
>
>
> This is what we spoke about [1] a month ago Brad. Why can't we have  
> an inner and outer shadow at the same time?

I think that is the question I just responded to, and I think I gave  
roughly the same answer then. If it is a shadow on an object or glyph,  
then how can it (as Henrik eloquently points out) be "carved into the  
plane it's levitating over"?

> Not being one for my graphic arts expertise, here's a few demos of  
> what I see.
>
> http://css-class.com/test/images/text-shadow3.png

Yes, that is a nice embossing effect (with a shadow under it), in  
which the emboss has a specular highlight, but I don't see an inner  
shadow there. I think that if you want to do bevels and embosses like  
that, or the "pillow emboss" that Flode suggests, then it should be  
its own property. If it is worth doing, then it should be done  
properly. Perhaps:

	 < text | box >-emboss:<angle><bevel-width><blur-radius/ 
sharpness><light-color><dark-color><specular-highlight-width><specular- 
highlight-radius><specular-highlight-blur-radius><specular-highlight- 
color>

But the syntax for bevels is a separate conversation. It is not a  
shadow. I do not want to see box-shadow or text-shadow become the  
dumping ground for every possible PhotoShop layer effect or lighting  
effect, when those effects are not shadows per se. Currently, box- 
shadow and text-shadow create drop shadow effects, and with the  
addition of <inner | outer> and <spread> [1] they would continue to be  
drop shadow effects, without too much extra bulk.

What you are suggesting is 2 new separate properties of "inner-text- 
shadow" and "inner-box-shadow" (instead of the simple addition of  
<inner | outer> to the existing properties), in order to achieve  
embossing effects and drop shadows at the same time. I'm saying that  
if you want 2  new properties for embossing effects, then they should  
be "text-emboss" and "box-emboss". Then we can debate whether they are  
worthwhile, without sullying the discussion of drop shadow effects,  
and without hacking drop shadow properties into something else.

> This is what I mean by having a shadow and highlight and the reverse  
> with a glow.
>
> http://css-class.com/test/images/text-shadow3.png

That's the same picture (by mistake, I assume). But my views on glows  
are the same as for embossing. Just because you can create a glow-like  
effect with a drop shadow by specifying zero offsets, does not mean  
that we should craft drop shadows with that hack as a goal.

I would much rather see blur radius added to "outline" in order to  
create a glow, since:

1.  a glow is a type of outline,

2. the outline width can be thought of as spread (especially if it  
could handle negative numbers),

3. Safari already has the concept of outline-as-glow.

> Thus why I suggested text-shadow and text-highlight. Text highlight  
> is layered above the text and the shadow is layered under. Both can  
> be declared on the same line of text.

Yes. But highlights are not shadows, and certainly not the drop  
shadows that text-shadow and box-shadow currently represent.

[1] http://lists.w3.org/Archives/Public/www-style/2008May/0222.html
Received on Wednesday, 18 June 2008 15:18:09 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:55:07 GMT