Re: locating subject of test results in EARL

>FWIW, I've seen there were discussions of trying to resolve the issue of
>locating a particular subset of a document using URIs and XPointer; I
>guess I should note this has several limitations:
>* it only works for XML documents
>* it only works for documents that have a URI

Me and Nick Kew developed a pretty reliable Fuzzy Pointer that worked well 
for even invalid HTML documents within the previous wai-er work, and within 
www-annotation.

The thread starting at 
http://lists.w3.org/Archives/Public/www-annotation/2002JanJun/0125.html 
seems to contain lots of relevant stuff and links to other threads, there 
doesn't seem to be a good single write up anywhere though, so it's not too 
convenient.

Cheers,

Jim.