[
Lists Home |
Date Index |
Thread Index
]
It's a little more. It is a control rendered to the
screen in some form, eg, the text link http:// which
turns blue and is clickable, or the button one presses
that then activates the hyperlink. In some past life,
I made the comparison of links to functions. A declarative
link is as you say, a relationship declaration, but then
we want to provide events and behaviors and are then
out of links and into controls. It seems to me that
on the web, a link is what http://something does and
that everything else is application design. What one
can share across application designs depends on the
framework, so when one is standardizing links and linktypes,
one is standardizing the implementation framework, and
the markup language design should fall out of that.
BTW, Bob, while we're here: why were you asking about
the use of abstract types for schema languages? Reply
offline if you like, but the topic came up in the HumanML
WG meeting yesterday as part of a discussion contemplating
an RDF ontology to complement the primary.
len
From: DuCharme, Bob (LNG-CHO) [mailto:bob.ducharme@lexisnexis.com]
>Hypertext is a particular application of linking, a more basic notion.
>It's a mistake to think of links solely in the hypertext context.
Hear,hear. I can complain about XLink with the best of them, but I do like
the Rec's definitions, particularly its definition of a link: "An XLink link
is an explicit relationship between resources or portions of resources." A
footnote, a URL on the side of a bus, and a guy holding up a "John 3:16"
sign at a basketball game are all links. A hyperlink is an interactive
presentation of a link on a screen--a rendering of a link.
|
