[xml-h] How semantic are links?
Fri, 17 Jan 2003 13:04:16 -0500
-----BEGIN PGP SIGNED MESSAGE-----
Say you've got this small set of schemas for your content. This is the
web, you've probably got links between different pages and different
sections of individual pages.
You've probably also got some stylesheets. I expect the blog pages and
the vacation pictures, and the technical white papers are presented a
little differently. Being a good web citizen, you're probably doing it
Ok, now the question is, in your mind, are the links in your documents
more like elements and attributes, things that are "baked in"
semantics, or are they more like style?
There's two approaches you could take here. You could have a little link
and you could use that in each of your schemas. Or you could use different
elements for linking in each of your schemas:
In the former case, you might expect the user agent to "just know"
that "a" is a link, the way HTML browsers know about them today.
In the latter case, you'd be relying on transformation or some
external link description to tell your user agent what elements are
links and what their semantics are.
In the latter case, you might even look at different things as links
at different times, or on different pages even though they're written
in the same schema.
I can see that this isn't a forced choice, you could have both, but
I'm curious about whether there's a strong preponderance for one
answer or the other.
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
Be seeing you,
Norman Walsh <firstname.lastname@example.org> | Why in our youth does the life we still
http://nwalsh.com/ | have before us look so immeasurably
| long? Because we have to find room for
| the boundless hopes with which we cram
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.7 <http://mailcrypt.sourceforge.net/>
-----END PGP SIGNATURE-----