[
Lists Home |
Date Index |
Thread Index
]
At this point, it is simply useful to compare what
Hytime suggests about linking as a study of what has
been tried and defined. It saves time. That's the point.
Out of line linking can work similarly to the way a link database
works. It is a set of named records (in a relational
implementation) for keeping the link (a control data type)
and the metadata about the relationship denoted by
the complete information (not just the link) in one
identified record. We use such all the time for concepts
such as involvements (in what circumstances has this
named individual or vehicle been encountered).
Let me ask: what should the information that activates
traversal be the same as information that assigns a
relationship?
Are hypertext controls the best way to get the job done?
Why? Because it is common isn't a sufficient argument.
So were handcranks for starting automobiles. Unless we
are quite clear about what a link IS before we decide what
it commonly is used for (a hypertext GUI control), we
will keep circling overlapping but non-isomorphic abstractions.
len
-----Original Message-----
From: Adam Turoff [mailto:ziggy@panix.com]
On Mon, Jan 06, 2003 at 07:23:09PM -0500, Simon St.Laurent wrote:
> SkunkLink's approach makes far more sense to me for in-line linking than
> does XLink's approach, and seems likely to be far more to attractive to
> browser developers deciding whether or not to invest the time needed to
> incorporate such mechanisms in their browsers.
I like the SkunkLink approach as well. Linking is such a fundemental
component to information processing that it's surprising there isn't one
universal method to link two pieces of information out there.
<snip />
I've heard the praise and hype for HyTime and out-of-line linking,
but I can't say as I've seen a single practical need for it. What
would a sensible out-of-line linking language provide that
SkunkLink+RDF (or the moral equivalent) *can't* provide?
|
