W3C home > Mailing lists > Public > www-style@w3.org > April 1998

Re: Functional notations (was: How to tell the browser that <A> is a link.)

From: Ian Hickson <exxieh@bath.ac.uk>
Date: Thu, 16 Apr 1998 21:21:17 +0100
Message-ID: <000401bd6b77$3e987120$cd20268a@hpxu>
To: Daniel Glazman <Daniel.Glazman@der.edf.fr>
Cc: Rasmus Kaj <kaj@interbizz.se>, www-style@w3.org, ij@w3.org, Daniel.Glazman@der.edfgdf.fr
>This idea is VERY interesting and I am clearly supporting the concept
>even if I'm not supporting the way it is done.

Hi Daniel!
Hey, not getting picky or anything, but are you going to say that to *all*
my ideas? ;-) [1]

What I was really proposing was the *personality property* to describe what
an element is, e.g. normal stuff or a link or a check box or a button. Read
my original post again, not the reply, because the reply to Rasmus' comment
(Hi Rasmus!) was a gut reaction, not well thought out like my original
My original proposal requires no knowledge of the ATTLIST part of the DTD
(only the actual element names, but then that is the whole point of CSS).

>The href property adds to CSS knowdledge of the DTD of the document,

Err... no. Again, reread the original post.
The idea is that this applies to *any* XML based markup language. The attr()
bit is simply to allow the {personality:hyperlink} declaration to work with
HTML. You could do this:

  ACRONYM.W3C { personality: hyperlink;
                href: url(http://www.w3.org/); }

  Visit the <ACRONYM CLASS=W3C>W3C</ACRONYM> web
  site for all the answers.

Which requires no knowledge of the DTD any more than P{color:navy} does.

Of course, it also begs the following rule:
ACRONYM.W3C:tooltip { content: 'World Wide Web Consortium' }

Regarding one of your other comments,
>        IMG.logo1 {
>                attr(src) : url(http://somewhere.com/wherever/logo1.gif)
>                }
>        <IMG CLASS=logo1>

While the effect that your description of the idea would have is quite
interesting, it is IMHO even worse. You can already do what you want with
the current state of CSS2:

IMG.logo1 { width: 100; height: 100; /* width and height must be given, like
they should be in HTML */
            background: url(http://somewhere.com/wherever/logo1.gif)
            border: none; }

IMG.left { display: block;
           float: left; }

<IMG CLASS="logo1 left">

In fact, you can do most of this in CSS1. Of course, it won't work in any
non CSS aware browsers, so you won't use it for months, but at least it
doesn't invent new technologies like your idea...

>Conclusion : I see immediate difficulties to access
>to element's contents for URL definition.
That was mentioned during the big regexp discussion. Basically, I would
agree that accessing the *content* of the markup is generally a Bad Thing.

[1] you said the same about relaxing the rules regarding pseudos.
Ian Hickson
Version: 3.12   Info: www.geekcode.com
GIT/M/S d->-- s+: a--->? C++(+++)>$ U>*++++ P L+>+++++ E(+)>+++ W+++ N(+) o?
K? w@ O- !M V- PS+ PE- Y+ PGP>+ t 5+++>++++ X- R+(+++) tv b++(+++) DI++
D++(---)>++++ G>+++ e(*)>+++++ h!()(--) !r y?
Received on Sunday, 19 April 1998 05:41:40 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:26:47 UTC