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

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

From: Chris Lilley <chris@w3.org>
Date: Fri, 24 Apr 1998 21:36:45 +0200
Message-ID: <3540E9CD.3A018834@w3.org>
To: Ian Hickson <exxieh@bath.ac.uk>
CC: Daniel Glazman <Daniel.Glazman@der.edf.fr>, www-style@w3.org
Ian Hickson wrote:

> And Daniel did answer:
> >Right. Does it mean you can define let's say the SPROTCH element in a XML
> document
> >with no attribute list and say _through CSS 2_ it should act as a hyperlink

No, but then, you would rarely want to. Lets use the right specs for the
right jobs.

To indicate that something is a link, use the XLINK specification. To
indicate the styling of a link, use a stylesheet language.

> That's the idea (but I would say CSS3). Although SPROTCH would then be an
> element of limited use, to be fair! using
>    <!ATTLIST SPROTCH  uri  >
> and
>    SPROTCH { personality: hyperlink;
>              href: attr(uri); }
> would be much more useful (pardon the glaring errors in my pseudo XML DTD
> thing here, I've only skimmed the XML spec).

Right. I suggest skimming section 3 of the XLink spec, instead [1] and
then noticing that any conforming XML parser immediately know that
SPROTCH is *not* a link because it does not contain the xml:link

Wheras the WIBBLE element is, quite clearly, a link:

<!ATTLIST WIBBLE xml:link CDATA #FIXED "simple">

and thus

:visited { color: red }

will turn all visited WIBBLE elements red. Simple, really.
> >Hmmm. Is it really the purpose of a **Style** Sheet ?
> I would say, yes.

No. Its semantics. Deciding what the link looks like, or sounds like, is
the purpose of a stylesheet.

[1] http://www.w3.org/TR/WD-xlink.html#3

Received on Friday, 24 April 1998 15:57:54 UTC

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