Kyle Banerjee wrote:
> What causes the uniqueness isn't the specificity. It's the virtually
> endless permutations for assigning subdivisions. For example, let's
> suppose you have a book on salmon genetics that just happened to focus
> on fish that were in Oregon. The rule say that salmon can be
> subdivided geographically, but genetics can't. So the cataloger
> wanting to make everything specific creates
>
> Salmon -- Oregon -- Genetics
>
> which is kosher by the rules but actually hinders useful collocation
> and meaningful browsing -- particularly when you add in the fact that
> the "aboutness" of works is frequently scattered across multiple
> subject headings. Precoordination is useful ONLY when the components
> are hierarchical or order can be predicted. It is frequently the case
> that neither of these apply.
I don't understand this without some additional examples. It just seems that we should never hobble our subject analysis because in my experience, people want more specific subjects, not less specific. And if there is a problem with pre-coordination, the catalog interface should deal with it. But making our records less specific, i.e. "worse" in my opinion, should not be done.
>
> > This is one reason why I am concerned over the current linear structure in
> the way LC has let out the subjects. We can't screen out geographic subdivision
> or display by time period.
> >
>
> It's actually pretty easy. LCSH is useless for this purpose, but MARC
> coding practice for the past decade or so has been to put geographical
> fields in |z, temporal in |y. Even when this is not the case in older
> records, geographic and temporal headings would be easy enough to
> detect.
Agreed. But the SKOS headings put out in id.loc.gov lack this subfield information, and is exactly the point that I (and I think Karen) were trying to make. If the subfields were in there, we would have no problem at all, but as it is, while we might be able to break the strings at the hyphens, we could not recreate the semantics, i.e. the subfields v,w,x,z. At least, not without a lot of work.
Jim Weinheimer
Received on Mon May 18 2009 - 17:49:20 EDT