Michael Fitzgerald schrieb:
> ... I am skeptical that the average person (or even the very odd
> person) would want the result of such a click to produce *every*
> author or combination of authors who have parts of the first and last
> names specified. If a system can't properly answer that simple
> question, then it needs improvement.
>
Considering all the effort that goes into name authorities and their
use, it may be a simple question, and users certainly intuitively expect
a straight answer, but that answer is not simple to produce.
> ... I imagine that
> Google is the same way. More results (whether accurate or not) are in
> their best interests. (Though I'm not saying that they think that
> more is better in every case, not at all.)
>
One nearly always gets lots more results than expected, and most of them
unexpected and unwanted. Key to success is the ranking. Catalogs can't
mimic Google ranking because they lack the necessary data elements
and structure (we would have to input all references from the books,
and in a consistent manner.) Key to the high number of results is the
high volume of data and full-text indexing. We can't put full text into
catalogs, but ToCs and maybe front of cover matter might be feasible.
Their inclusion into indexing (as is done by Ohionet) is probably
the biggest improvement that can be achieved for OPACs. Full text
includes lots of irrelevant keywords, ToCs contain many of the
relevant ones, and many more than the few LCSHs that someone might
have slapped on. With titles alone, OPACs are horribly poor in
relevant keywords.
Adding ToCs would not, I believe, make subject headings disposable.
LCSH and name authorities provide a syndetic structure the likes of
which Amazon and Google don't have. OPAC success depends on in what ways
that structure is being put to use, but its use will be in collocation
rather than in locating. IOW, before you can collocate (employing the
authority data) you have to locate something, anything, and this can
only be improved by adding more verbal substance to the bone-dry
skeletons of our records.
B. Eversberg
Received on Thu May 31 2007 - 02:03:15 EDT