A serious detail point

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
independent justification) want the standard to change so my usage is

Here's the problem: I want to have tens of thousands of links in a
single document, all pointing to (spans in) ONE target document.
Without a way to specify a default containing resource OTHER THAN the
document contain the linking element, I will have to reproduce the URL
for my target document in the specification of EVERY ONE of those

Or have I missed something?

Proposal:  another attribute (groan :-) for SIMPLE, EXTENDED and
LOCATOR, e.g. 'CONTEXT CDATA #IMPLIED', which defaults to the
empty string, meaning the containing document, but which once given a
value (a URL) is sticky, i.e. the resource thereby identified is the
default containing resource for all subsequent locators, until
overriden explicitly.



[I wasn't going to say anything about implementation efficiency, but I
will because it stands as a counter-argument to the best work-around
I've come up with, namely

<!ENTITY c "http://www.library-of-congress.gov/index/">
<myptr href="&c;#ID(p324)..&c;#ID(p334)"/>

This will work, of course, and it cuts the overhead down to 6
characters per locator, but a) it's obnoxious and b) it's likely to be
seriously slow if implemented without caching, which is I claim much
more likely to be the case for resources in general than it would be
for a (reasonable to expect to be small) set of containing resources.]