Re: [css3-background] Default shadow color

That the shadow would change based on the colour of the text surprises me too. Defaulting to a set colour (like black) seems to make the most sense. Nobody should HAVE to specify a colour, but they should be shown a default that they can change whenever they want. The colour of the text and the colour of the shadow should be seen as two completely separate styling options. Just as setting the background of a div (or any of the new HTML5) elements shouldn't change the colour of the border as they are also two seperate styling options.

Authors may also want to put a shadow in while making mock-ups where they may not have settled on a specific colour theme yet. In this instance its best to have a default colour not based on the colour of the text. Imagine having the colour of text set to white just for demonstration purposes. You then set a shadow and don't define a colour. The shadow is now white, which is undesirable even for a draft. You would then have to think about the specific colour of the shadow long before intended. Having a default colour would, at the very least, give authors a starting point.

(I am from Australia so please excuse my spelling of the word "colour". Any time this word arose, I was of course talking about "color" in a CSS sense.
-----Original Message-----
From: Xaxio Brandish <xaxiobrandish@gmail.com>
Sender: www-style-request@w3.org
Date: Sat, 5 Mar 2011 23:53:13 
To: Sylvain Galineau<sylvaing@microsoft.com>
Cc: Brad Kemper<brad.kemper@gmail.com>; Brian Manthos<brianman@microsoft.com>; fantasai<fantasai.lists@inkedblade.net>; Estelle Weyl<estelle@weyl.org>; www-style@w3.org CSS<www-style@w3.org>
Subject: Re: [css3-background] Default shadow color

Good evening,

Let's not pick on somebody for using the term "surprising".  Fantasai also
used the term "surprised" earlier, and I do agree that's an accurate choice
of wording.

In everyday life, if one sees letters made of black construction paper
suspended above a page, with a light source casting a shadow on that page,
the color of the letter shadows does not change if the color of the
construction paper changes.  If the color of the shadows did change, I'd say
it would be surprising.

However, the digital display isn't and cannot always be modeled after such
print media.  Making the color a required component seems to make sense
because only the author knows how their shadows should be displayed.

*Unfortunately*, not all authors have a sense of where to start with
shadows, and forcing them to specify a color/opacity for shadows would lead
to quite a bit of trial-and-error before they figured out what seems correct
to them.  Having the UA default this seems like a reasonable way to avoid
hassle.  What makes sense for a default for on-screen rendering DEFINITELY
may not make sense for projection media.  The UA can provide a valuable
starting point (default) for authors.  Devices with limited color profiles
may not be able to render author-specified preferences in much accuracy,
which provides another case for defaulting the color to the UA.

Unless the media for this property becomes more specific than "visual", I
would vote for the UA to be able to provide a value for this color (and
opacity!) if one is not explicitly specified.

However, specifying currentColor does not seem like an interoperable default
except among browsers, which is very limiting to other types of visual
media.

Although the group is trying to determine the specification [ by specifying
things, nonetheless! ;) ], there are sometimes parts of the specification
that need to be open enough to interpretation that they become *useful*
parts of the specification.

--Xaxio


On Sat, Mar 5, 2011 at 10:41 PM, Sylvain Galineau <sylvaing@microsoft.com>wrote:

> It's been like this for quite some time and it wasn't 'surprising' until
> you and others here heard of it. As far as I can tell from looking all over
> including public bug databases, it has simply not been an issue. I wish all
> 'surprising' behaviors were like this.
> ________________________________________
> From: Brad Kemper [brad.kemper@gmail.com]
> Sent: Saturday, March 05, 2011 10:55 AM
> To: Sylvain Galineau
> Cc: Brian Manthos; fantasai; Estelle Weyl; www-style@w3.org CSS
> Subject: Re: [css3-background] Default shadow color
>
> On Mar 4, 2011, at 6:11 PM, Sylvain Galineau wrote:
>
> > I don't get how black is better than currentColor given that it is the
> default color for text in most browsers; why pick a default that's will be
> wrong the most often ?
>
> Black or translucent black would be wrong less often. Shadows that stayed
> black when the author changes the text color to red would be less
> surprising. Having the box shadow turn red just because you turn set the
> color of the text to red would be totally surprising to almost anyone.
>
>