On Fri, Apr 8, 2011 at 5:02 AM, Owen Stephens <[log in to unmask]> wrote:

> Then obviously I lose the context of the full heading - so I also want to
> look for
> Education--England--Finance (which I won't find on id.loc.gov as not
> authorised)
>
> At this point I could stop, but my feeling is that it is useful to also look
> for other combinations of the terms:
>
> Education--England (not authorised)
> Education--Finance (authorised! http://id.loc.gov/authorities/sh85041008)
>
> My theory is that as long as I stick to combinations that start with a
> topical term I'm not going to make startlingly inaccurate statements?

I would definitely ask this question somewhere other than Code4lib
(autocat, maybe?), since I think the answer is more complicated than
this (although they could validate/invalidate your assumption about
whether or not this approach would get you "close enough").

My understanding is that Education--England--Finance *is* authorized,
because Education--Finance is and England is a free-floating
geographic subdivision.  Because it's also an authorized heading,
"Education--England--Finance" is, in fact, an authority.  The problem
is that free-floating subdivisions cause an almost infinite number of
permutations, so there aren't LCCNs issued for them.

This is where things get super-wonky.  It's also the reason I
initially created lcsubjects.org, specifically to give these (and,
ideally, locally controlled subject headings) a publishing
platform/centralized repository, but it quickly grew to be more than
"just a side project".  There were issues of how the data would be
constructed (esp. since, at the time, I had no access to the NAF), how
to reconcile changes, provenance, etc.  Add to the fact that 2 years
ago, there wasn't much linked library data going on, it was really
hard to justify the effort.

But, yeah, it would be worth running your ideas by a few catalogers to
see what they think.

-Ross.