Re: A serious detail point

I am sympathetic to Henry's problem as I also encounter something very
similar (possibly for different reasons).  Henry's motivation sounds like
the annotation of a document (e.g. Shakespeare) while mine is the provision
of links from a document to a glossary.  I had also come up with the 
entity mechanism as the simplest and (hopefully elegant) mechanism and
(because my glossaries are usually small files under a common root) decided I 
would need a caching mechanism.

In message <1242.199704171551@grogan.cogsci.ed.ac.uk> "Henry S. Thompson" writes:
> Some of you may have heard my talk(s) at WWW6 about standoff markup.
> They seemed to strike a chord with a number of folk.  I'll be doing a
> similar talk at SGML Europe next month.  I use XML-LINK in these
> talks, but it's actually bogus, and I (arrogantly? No, I think there's
Is this because you cannot accurately address the target document (i.e. spans)?
Otherwise it seems possible using the existing syntax as below:

in commentary.xml (targetting Jon's markup):

<NOTE XML-LINK="SIMPLE" CONTEXT="http://some/where/i/forget/macbeth.sgm" 
    HREF="#DESCENDANT(4,ACT)(1,SCENE)(1,LINE)" TITLE="Simple recipes"/>
<!-- 20 more notes -->
    TITLE="Care of Trees"/>

However, if the initial document (the link container) is likely to change,
so is the context of any XML-LINK.  (In my hands, this would be a recipe for
disaster.)  It would be similar if *part* of the link container were abstracted
(e.g. via TEI Xptrs) so that the context were again lost.  
For example I could search for 
which returns a well-formed fragment which the reader might assume contained 
valid HREFs, but no longer does.  In the latter context it could be possible
for an XML-LINK processor to add the CONTEXT info during initial traversal,
but it would be easy to foul up.  


Peter Murray-Rust, domestic net connection
Virtual School of Molecular Sciences