Re: [ARSCLIST] cataloging sound recordings

[Eric Jacobs <ericj@xxxxxxxxxxxxxxxxxxx>]

Wow - that post certainly cuts through a wide swath of topics
and ideas.  But it did inspire a few comments...

With respect to search tools, Google et al are good tools for
unstructured searching through heterogeneous data.  The
question is, do sound recordings lend themselves better to
structured or unstructured searching?  In an ideal world,
metadata for sound recordings would be captured with some
consistency, making structured searching the way to go, and
make "finding a needle in a haystack" more feasible.  Otherwise
unstructured searches may still leave you with thousands of hits
to sort through.  Google and its ilk are not necessarily the
answer to all of our search problems - you need the right tool
for the job, and it all starts with the data.

As for digital libraries, I couldn't agree more that they are
the future.  But when?  Widespread digital access across all
socioeconomic groups is an important prerequisite, otherwise
the content becomes available to only the more privileged.  But
digital libraries need to become a reality.  And like sound,
they must grapple with issues of copyright as well - another
issue that must be addressed to make the digital library a
reality.

Eric Jacobs
The Audio Archive


---- Original message ----
>Date: Tue, 26 Apr 2005 09:31:48 -0500
>From: Karl Miller <lyaa071@xxxxxxxxxxxxxxxxx>
>Subject: Re: [ARSCLIST] cataloging sound recordings
>To: ARSCLIST@xxxxxxxxxxxx
>
>On Mon, 25 Apr 2005, Steven C. Barr wrote:
>
>> Also, as I have commented before, I think that ARSC might try and define
>> a basic core set of fields (in the sense of field types and mimimum sizes
>> as well as names) for phonorecord cataloguing. Each user could, of course,
>> augment this core set of fields as he/she/it saw fit...but it would at
>> least guarantee data interchangibility for the core fields, which would in
>> turn make it possible to create an "uber-catalog" comprising as many
>> collections as possible. This could be used to:
>> 1) Collect data on as many sound recordings (I'm thinking of 78's, but
>>    others could be included) as possible. This could be used as a data
>>    source on recordings.
>>
>> 2) Establish, for researchers and other interested parties, a guide to
>>    which sound recordings still exists and where they can be accessed.
>>
>> There are all sorts of other possible uses which suggest themselves,
>> or will if this is discussed!
>>
>> I know of at least present or proposed institutional collections of
>> sound recordings aimed at collecting either all such recordings or
>> different subsets of them. There were/are other collections of data,
>> Not standarizing the digital portion of the process(es) means at
>> the very least these projects may wind up working at cross-purposes
>> with no practical means of sharing or comparing information on
>> activities.
>>
>> Thoughts?
>
>
>The MARC record is marketed as the answer to most of these considerations.
>I sometimes make the comparison to the CISC and RISC computing. While I
>will never market myself as an expert on anything, and perhaps I just
>don't know or understand enough, but it has seemed to me that RISC is a
>more efficient computing environment. Yet CISC is still with us. I wonder
>if that is because of its overwhelming share of the computing marketplace.
>I think also of how long it took companies like Nikon to accept the
>realities of the use of the computer for both the control of exposure and
>then digital imaging...they had always promised your Nikon lenses would
>never be obsolete.
>
>There is such a huge infrastucture that has evolved around the MARC
>record...There is an interesting similarity in the notion of shelving
>books with Dewey decimal versus the Library of Congress systems. Only when
>the Dewey system became ridiculously encumbered, did most libraries
>convert.
>
>For me, declining library budgets are the only hope for libraries in that
>if they are to address their mission, they will be forced to embrace the
>technology, and make their cataloging systems more friendly to both users
>and those who are employed to input data. This would reduce the time and
>level of training required and save money. Right now, subject knowledge is
>almost on equal terms (or so it seems to me) to knowledge of the
>methodology of data input. If you can simplify the system, you could get
>more done without the need of those cross trained in both disciplines
>(subject and the methodology of cataloging).
>
>Then I consider, where is one going to find someone conversant in both
>matrix numbers, etc. and the methodology of the MARC record. Further, with
>librarians not often being users, how is one going to convince them that
>things like matrix and take numbers are of significance. Some are
>convinced, yet others don't know what you are talking about.
>
>The solution has seemed to be, informed curators setting up their own
>files with things like keyword searches for finding aids. While these
>sorts of solutions are helpful, they don't address the fundamental issues.
>
>For me (an I know I am rambling, but I figure only those interested in
>bibliographic control are still reading), how would one find if there
>existed a surviving copy of Koussevitzky conducting Copland's Third
>Symphony. Well, OCLC and RLIN aren't any help. However, you might think, I
>wonder if Copland had a copy. His collection went to Library of Congress.
>So you look on SONIC and there it is.
>
>On the other hand, you may be able to find your friend's Master's recital
>cataloged on OCLC. It is not just a question of the methodology of
>locating something, but the priorities for cataloging. One institution may
>have more money available so they can catalog you friend's
>recital...actually something often required by the certification process
>for a graduate degree.
>
>For ten points, where can one find some recordings of music of piano
>improvisations by Schillinger? The route to that information is not an
>easy one...hint...try the Library of the Performing Arts. (New York
>Public)
>
>It is a complex problem and from my perspective, a problem encumbered by
>the legacy of OCLC and the nature of libraries and library education. I
>guess only when it collapses under its own weight,
>or when google becomes an accepted part of the research methodology, will
>the library establishment once again, as it did when the MARC record was
>developed, do a thoughtful reexamination of the methodology of cataloging.
>
>I am reminded of an article I read some years ago...I think the title was
>"The end of libraries." The author suggested that, in time, information
>would become too valuable to be entrusted to libraries which provide free
>access and by their nature, were unable to charge for access and existed
>as a public service. Perhaps the leadership of ARSC might like to sit down
>with the folks at Google. OCLC has and has developed a relationship, and
>perhaps Google might be able to bring libraries into the 20th Century (yes
>I mean 20th Century, for, from my perspective they are still in the 19th
>Century as they are only now beginning to recognize electronic
>information, the recording being the first, as worth considering in a
>serious way.) For all of the hype about libraries embracing technology,
>it seems to me that much of their application of the digital
>technology is still be caught up in the linear modality of the printed page.
>
>From my perspective "the cheese has been moved" and libraries don't know
>it yet. They seem to be looking in the same places. As long as their
>cheese is limited to books, they are fine. They seem to look for books in
>the digital environment. Maybe libraries are just about books.
>
>
>Karl