Re: css draft specification

Mike Batchelor ([email protected])
Mon, 10 Jul 1995 08:51:56 -0400 (EDT)


Bert Bos once wrote...
>
> |Comments welcome.
> |
> |-h&kon
> |
> |[1] http://www.w3.org/hypertext/WWW/Style/css/draft.html
> |[2] http://www.w3.org/hypertext/WWW/Arena/0.97.html

I think the KISS rule should apply to the first style sheet specification.
Keep it very basic, very simple, and it is more likely to be implemented.
A lot of this seems too complex *for now*. An example of what I
mean:

> ** Links
>
> Setting per-element link properties can also be done with
> context-dependent addressing. If a link in a P should look differently
> from a link in an H1, one can use a syntax like:
>
> *P*A: text.color = #00F # A inside P
> *H1*A: text.color = (1.0, 0.0, 0.0) # A inside H1

This can be done more simply with classes:

a.paragraph: font.color = #00F
a.level1: font.color = #F00

<h1>This is a <a href="next.html" class=level1>link</a> inside a level
one header</h1>

This is a <a href="next.html" class=paragraph>link</a> inside a
paragraph.<p>

This is another <a href="next.html" class=paragraph>link</a> in a
paragraph.<p>

All the UA has to know is the class of the element, in order to match it
up with a style. In Bert's example, the UA would have to infer a
paragraph break at the <h1> for the first link to be matched with the
link-in-a-paragraph style, and also for all the other elements that imply
a paragraph break (<hr>, <h1-6>, <address>, etc). That makes the UA code
more complex, which will stall implementation, IMO.

Authors may try and abuse classes to assert a particular style, I know.
Perhaps my example is an abuse of classes. But I think this risk needs to
be weighed against the ease of implementing style sheets in the UAs. I am
assuming we all would like for the proposed CSS to become widely
implemented, right? :) For a first crack, then, the specification ought
to be fairly simple. Enhancements and refinements can come later. The
trick, then, is to define a style sheet specification that is simple, yet
open to extension in the future.

I realize you have all been working hard to refine these mechanisms for
many months, and I hope that as a newcomer, I haven't inadvertently
stepped on someone's toes. I'd just like to see something of these ideas
implemented soon in a widely used UA, before the world is owned by
Netscape, and HTML becomes more cluttered with PostScript-isms.

-- 
 %%%%%% [email protected] %%%%%% http://www.clark.net/pub/mikebat/www/ %%%%%%