Re: [ARSCLIST] Cataloging Sound Recordings

[George Brock-Nannestad <pattac@xxxxxxxx>]

From: Patent Tactics, George Brock-Nannestad

----- in response to Karl's discussion that I wholeheartedly support:

some of my best finds of relevant literature (and something no-one else has
found, which is a useful attribute in some situations) have appeared after
searching for deliberate misspellings. So bibliographic control would be
quite necessary to serve the commonweal (new term I have just added to my
vocabulary. I am not going to use it much).

The whole history of Information Science is extremely interesting, because it
gets down to the roots of how we organize information so that it is
retrievable. The hierarchical decimal classification systems (Dewey, and I
suspect LC - Denmark uses UDC for scientific libaries and a simplified
version for public libraries) were basically pigeon-hole systems that
permitted you to put things somewhere with others that you considered
similar. However, queries that went across the hierarchies, that looked for
similarities other than those appreciated at the time of filing, were system-
hostile. Knowing that new developments frequently occur at the interface
between subject areas (cross-fertilization), would indicate that it would be
very difficult to find material in cross-overs that were not realized at the
time. Coordinative indexing was very promising, because using a controlled
vocabulary (a Thesaurus) permitted the creation of lists of numbered
documents (access numbers) belonging to the terms used, as many terms as
needed being used to index an entry, and the answers to searches being
obtained by comparing lists of those permitted terms that you combined (by
means of logical operators) during your search. This meant that for logical
AND a search would never take longer than comparing to the shortest list.

Some present systems are based on coordinative indexing, but sequential
reading (string-search or "free text") has become so fast that some of the
cheap database systems are not at all well organized. Instead we are given
"search aids", such as wildcards. In other words, we are in many instances
forced to use quite unintelligent tools. From personal experience I know that
a good Thesaurus (with suitable broader terms BT and narrower terms NT, as
well as preferred terms PT) works well and reliably. A Thesaurus must be
revised from time to time, but it can be made backwards compatible. And using
indexing terms interactively from a Thesaurus will ensure correct spelling -
it is easier to click on a term than to type it all over again.

Based on a simple DOS programme from the early 1980s called MicroFind I had a
license to create databases, and it was very simple. Some would probably
consider it a disadvantage that the search tables needed to be updated with
new entries (say, daily), before efficient searching could be performed, but
I imagine that possible updated versions would do this in the background and
take a short time only.

This looks like one of those entries where you can read the whole thread -
another way of making information easily retrievable.

Kind regards,


George

> On Mon, 25 Apr 2005, Thom Pease wrote:
>
> > A noble thought.  In the meantime, we have decades of
> > unique recordings in the OCLC database, which are
> > linked to LC authority data.  This is useful
> > information for discovery, once you figure out how to
> > use it.
>
> Which is, for me, one of the major problems.
>
> Researchers of musical scores have many of
> > the same problems, esp. with manuscripts and early
> > editions.  How have they gotten around it?  With RISM,
> > thematic catalogs, and other indexes that supplement
> > their library catalogs and First Search.
>
> As with other finding aids. But my question is, why have they seen a need to
> circumvent the fundamental of the single system. I believe it has to do with
> several factors, limitations of the system, the complexity of data entry as an
> end unto itself and the funding limitations which have libraries devoting much
> of their energies to those items which are likely to be frequently requested.
>
> > Libraries that own sound materials could not afford to
> > re-catalog all these materials, and it is for
> > libraries (and their patrons) that the OCLC
> > Bibliographic File is often valuable (WorldCat/First
> > Search for patrons).
>
> I agree, but that is only because the information is not valued enough.
>
> > Often times, it isn't MARC or AACR2 or even OCLC
> > that's failing us...but rather the integrated library
> > systems that we're using in our local institutions.
> > Not naming systems here, but there's one that just
> > won't index uniform titles with main entries
> > correctly--it just drives librarians crazy.
>
> And can you imagine the frustration it provides patrons!
>
>   The good
> > news is that there are many systems out there, so you
> > just have to pick a good one.
>
> Yes, but that is not an ultimate solution. Of course, one question is, is
> there an ultimate solution...a question which we ask every day in
> preservation.
>
> > What you're proposing is something similar to what
> > Wikipedia is about.  That might be useful too, but how
> > could experienced people trust all of the information
> > in there.  Ever is it the conundrum between providing
> > access and our desire to have authorized control over
> > names--to have the right name, the right title, etc.
>
> For me, that is also a fundamental concern. However, as I like to say,
> when you consider the increase in information, one has to question the
> basic notion of bibliographic control. The production of information has
> already exceeded our ability and/or resources to control it. Why not
> concentrate on the development of technology to navigate
> information...which is one of the basic notions of most search engines. I
> believe that our current notions of control have already "lost control" as
> they are predicated on the basic notion of the linearity of the printed page.
>
> > I think improving/influencing LC cataloging
> > policy--maybe even a redesign of the sound recordings
> > work format--would be a better use of resources than
> > trying to build the next database.  The problem is the
> > material that we've already put in the catalog for
> > decades.
>
> Yes, there is sense to that, but then, I remember when libraries were
> making the conversion from Dewey to LC.
>
>  For that reason and others, I'd like to
> > paraphrase a thought from Rob Ray and Chuck Haddix's
> > March 2005 ARSC pre-conference workshop that shared
> > catalogs are better than individual databases.
>
> Yes, but I would respond that the development of good naviation tools are
> closer to the solution than reinforcing the notion of control. Control is not
> only a psychological issue for people, but a systemic issue for institutions
> and professions. Under the current notion of bibliographic control, it seems
> to me that the only hope for libraries is the reduction of acquisitions and
> the rechanneling of monies to bibliographic control.
>
> You can't expect a grant to do cover the basics of operations. What do you do
> when the grant money is gone...you apply for another...and then you hire staff
> looking for grant money and divert more of your resources to raising money,
> branding, publicity, and all of it going to help you look for the cheese
> (bibliographic control and the money to support it) in the same places...and
> it has been moved.
>
> Just my thoughts.
>
> Karl