[
Lists Home |
Date Index |
Thread Index
]
>So, you're suggesting a single RDF syntax to define hypertext
relationships?
>I'm confused.
Sorry. I agreed with Michael Kay that we don't need any new standard to
define *hyper*linking (or hypertext), because hyperlinking is the use of UI
widgets to present different kinds of linking, and each medium has its own
particular UI capabilities that evolve over time.
The more general concept of linking, which is the representation of a
specific relationship (often, a typed relationship e.g. weblog entry X
praises/criticizes/misrepresents weblog entry Z) can be modeled well enough
with RDF. No existing syntax to represent the RDF model is getting great
traction. Among the efforts to come up with new syntaxes, the RDF-in-XHTML
project is working with some ideas that can do much of the job of expressing
resource relationships without scaring too many people off, which I find
encouraging. They're mostly thinking in terms of metadata for now, e.g. how
to represent statements like "1. this web page has a rights profile
described at this Creative Commons web page, 2. this page was authored by
that person, 3. that person has a home page at http://whatever."; If someone
wants to write a stylesheet that loads these facts into a database, or turns
them into a/@href links or javascript popups, and it serves the needs of
their users, great.
I've felt for a while that linking is about expressing relationships and
attaching metadata to those expressions, and RDF is good for that. What a
developer does with those expressions, like what the developer does with
<title/> and <xref/> elements, is up to his or her imagination and
familiarity with the options available in the output device at hand.
Bob
|
