[xml-h] How semantic are links?

[Gavin Thomas Nicol]

On Friday 17 January 2003 01:04 pm, Norman Walsh wrote:
> I think my mind is still open on this issue, but I tend pretty
> strongly towards the former camp, I think. Links are semantic and
> belong in the markup in a first-class way that I don't think style
> deserves. If I send you a document, you should be able to know what
> the links are, even if you have to resort to reading the content in
> notepad.

We should avoid the term "semantic" here I think.=20

The notion of having "linkedness" applied, like a style, is entirely sepa=
rate=20
from markup concerns, so we really have 2 questions:

1) Should there be a common "schema" for link markup?
2) Should applications be able to interpret arbitrary markup as links as
    process it as such?

I think the answer is fairly strongly "yes" to both to my mind. Having a=20
common "schema" is good for interchange, but being able to make arbitrary=
=20
markup "live" is extremely powerful.

For (2), the best example I can think of is DynaText/DynaWeb, where *all*=
=20
hyperlinking was ultimately "applied" to the documents, just as style was=
,=20
and you could do it in a nice visual editor. Of course, you can accomplis=
h=20
much the same thing with JavaScript in modern browsers, but the model is=20
somewhat ugly.

(2) is roughly equivalent to running an XSLT transform over a document.