Re: When and how should tesuji be a link?

At 9:41 AM -0500 6/16/97, Paul Prescod wrote:
>David G. Durand wrote:
>> Single-ended out of line links can be used to implement external tagging of
>> documents.
>
>Links with only one end seem like a logical impossibility. What are they
>"linking"?
>
>> This can be useful for marking up read-only unmarked documents,
>> or for recording alternative markup of a single document. Conventions for
>> expressing that semantics for links jhave not yet evolved, but are
>> certainly possible and useful.
>
>Okay, so they are "linking" semantics and data. I believe that the
>"semantic" should be considered a resource in which case there are two
>resources and two ends.

This is an old debate in the hypertext data model discussion. The question
is whether links should contain type information or other semantic
annotations, ir simply note correspondences between items. I call this the
distinction between lightweight and heavyweight links. Since heavyweight
links with n endpoints can be modeled as lightweight links with n+1
endpoints, the debate is a bit moot.

Because of the architecture approach we have chosen, there _is_ type
information associated with each link -- given that, 1-ended links make
perfect sense, and has (at least) the applications that I have suggested.

Whether we call such semantics link data, link meta-data, or an implicit
semantic endpoint is irrelevant. In fact, I think that Hevyweight links
have a bit more power since the semantics is a universal attribute of all
links, and it is distinguished on all links as well -- whereas it just
becomes another endpoint in the lightweight case.

   -- David

I suspect that this we are in raging agreement, but I want to make sure
that that is the case by making my position more explicit.


_________________________________________
David Durand              dgd@cs.bu.edu  \  david@dynamicDiagrams.com
Boston University Computer Science        \  Sr. Analyst
http://www.cs.bu.edu/students/grads/dgd/   \  Dynamic Diagrams
--------------------------------------------\  http://dynamicDiagrams.com/
MAPA: mapping for the WWW                    \__________________________

Received on Tuesday, 17 June 1997 16:36:57 UTC