Re: Stretchy backgrounds? (background-width,height properties)

On Sun, 8 Jul 2001, Dylan Schiemann wrote:
>>
>> So you're saying that if you object, we should take it out, but if
>> we object, we should leave it in because you want it? :-) (There
>> are many reasons to have these colours, including promoting
>> interoperability).
>
> The argument as I remember it was why add something that is
> considered to be an arbitrary system and with which no one seems
> happy. The fact that Dodger Blue is one of the 140 colors says it
> all, in my opinion.

Like I said, one of the many reasons to include these colours is
interoperability. At least four browsers support these colours exactly
already. Better to have everyone agree that "orange" means a
particular shade of red rather than have different authors reverse-
engineer different browsers differently, ending up with incompatible
renderings of the same document.


>>> - the ability to use text as a background and to specify how it
>>> would be positioned and repeat.
>>
>> There were valid objections made to this proposal (namely, it was
>> underspecified and nobody could see a way to improve it).
>
> I don't know if this information ever made it to the public eyes.

The discussion only happened in the public eye as far as I recall.

The main problems, from memory, were that the concept was
underspecified, there was only one clear use case, and implementation
would have been non-trivial.


> But these things always start out underspecified. I think there were
> some good ideas that could have led to something.

If you have an idea for the CSS spec, I would strongly recommend fully
specifying it and having it well discussed on www-style. If an idea
surfaces but isn't well thought out, then working group is not likely
to spend much time on it. Don't forget that every member of the
working group has a separate full time job in addition to working on
the CSS specs, and each has their own pet property or value that they
want added. If there isn't strong support for an idea, it is unlikely
to be considered further.

(These comments are based on my personal experience and are not the
official position of the W3C or the CSSWG, whatever that may be.)


>>> - better definitions for opacity
>> This is being worked on, but is not a top priority.
> The main point I had a while back was that setting opacity on the
> element rather than on properties of an element seemed like a bad
> idea. For example, it makes sense to be able to set the opacity of
> the background independent of the foreground color.

You have to balance power and ease of specification, use and
implementation. Also, it is possible to get the same effect both ways,
so the argument for more power in this case is not very strong.

However, like I said, the issue is still open. (Personally I hope that
the way opacity is done in SVG is what we use for CSS [1].)

[1] http://www.w3.org/TR/SVG/masking.html#ObjectAndGroupOpacityProperties

-- 
Ian Hickson                                            )\     _. - ._.)   fL
Invited Expert, CSS Working Group                     /. `- '  (  `--'
The views expressed in this message are strictly      `- , ) -> ) \
personal and not those of Netscape or Mozilla. ________ (.' \) (.' -' ______

Received on Sunday, 8 July 2001 23:29:18 UTC