[xml-h] Linkbases and "Document Enrichment"

[Simon St.Laurent]

ldodds at ingenta.com (Leigh Dodds) writes:
>Here's another use case for link bases, I'm afraid I'm behind on 
>this list, so apologies if someone's mentioned it already.

There's always room for more discussion of these things!

>I've always assumed that when people talk about link bases 
>they're usually assuming some form of coupling between the 
>source material and that link base. i.e. the link base is basically 
>the links from the source material factored out into a separate 
>document thereby allowing the links to be described in a richer 
>fashion.

That's one aspect of it, yes, but you're right that there's more.

>A twist on this is where the link base and document are authored 
>independently. The document can then go through an "enrichment" 
>process to tie it into the link base, allowing the reader to navigate 
>to more resources than were originally specified in the document.

Yes, definitely.  In polite usage this is called annotation; in impolite
circumstances it's called grafitti.

>This might involve textual analysis of the original document e.g 
>to pick out identifiers, key phrases, etc.
>
>PubMed do something similar to this with their Linkout feature:
>
>http://www.ncbi.nlm.nih.gov/entrez/linkout/doc/linkoutoverview.html
>
>This is effectively a link base that can be used to enrich scientific 
>papers with links to related articles, journals, even gene sequences 
>and the like.
>
>In a way I suppose this is using a link base as an annotation feature.

For me, the use case I'm looking forward to is the combination of
pristine historical primary sources with secondary perspectives.  I
wrote a piece on some of this long ago:

http://simonstl.com/articles/hypehist.htm

I'd like to make it more possible for readers to interact directly with
original sources rather than relying on an authoritative secondary
voice, while still preserving the value (ugh! who wants to read all
those primary sources!) of the explanations in the secondary sources.

The scientific case is probably a better story overall for reaching
people, though.

>Do people commonly do this kind of thing?

Lots of people do it in practice - biblical concordances and legal
citation work come readily to mind.  I don't know how many people are
doing it with XML per se at the moment.  I'm a long ways from having
even the basic engine I want for my historical work.

>p.s. stable linking in general is a bit of a hot topic in the STM area, 
>with features like DOI (http://www.doi.org) and CrossRef 
>(http://www.crossref.org) 
>being commonly used. These are probably off-topic here though.

Might not be off-topic.  What's STM?

-- 
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com -- http://monasticxml.org