I think Karen's point about the authority files was not that they were
designed to help cataloguers and the creation of authoritative headings,
rather than to achieve what you describe - appropriate cross-referncing.
So, she is not arguing that we should do away with cross-referencing,
but rather that the tools we currently have (authority files) are not
the best way to achieve this.
I'm not sure I understand your point about searching for "wwi" on
Google. However, one of the interesting things to note is despite the
lack of authority files, how well Google helps the searcher who uses
alternative terms. Searches for Tolstoi and Tolstoy turn up relevant
information, linking to pages with both spellings. Although obviously
the results sets have their flaws, trying to do a similarly useful
search against the Library of Congress catalogue is not straightforward.
Now, if when I searched for tolstoi (under any index) the LoC prompted
me "Are you looking for works by or about Tolstoy, Leo, graf, 1828-1910"
then I'd be impressed.
Finally when Karen says "if we were to re-design our data with systems
in mind" I don't think she means we should pick a specific system, and
then work out what the data needs to look at to exploit the functions in
that system. My own interpretation would be that we need to design data
in order that it can be effectively utilised in a computerised
environment. What we have is data designed to work well in a
non-computerised environment. I also think we need to think about
designing data to optimise finding it, rather than filing it (i.e. think
about the searcher, not the cataloguer).
Just my Wednesday morning thoughts...
Owen
Owen Stephens
Assistant Director: e-Strategy and Information Resources
Imperial College London Library
Imperial College London
South Kensington
London SW7 2AZ
Tel: 020 7594 8829
Email: o.stephens_at_imperial.ac.uk
-----Original Message-----
From: Next generation catalogs for libraries
[mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Weinheimer Jim
Sent: 22 August 2007 08:40
To: NGC4LIB_at_listserv.nd.edu
Subject: Re: [NGC4LIB] Ceci n'est pas un catalogue
Karen Coyle wrote:
> But there are definitely some
> problems with the data in our current environment. As an example, one
> thing that we have all over others is our use of controlled
vocabularies
> and authoritative forms of names. But our authority data is designed
to
> aid the *creation* of authoritative forms, not the use of those forms
in
> a system. The LCSH authority records are virtually useless (ok, maybe
I
> exaggerate) because they are based on instructions for catalogers and
> don't include the authoritative headings nor the proper references for
> an online catalog. I think that if we were to re-design our data with
> systems in mind, rather than basing our data on the cataloging rules,
> then we could greatly improve the catalog itself.
I'm not sure I understand about the uselessness of authority records. I
certainly won't argue that the way they have been implemented in current
OPACs (that word again!) is very poor indeed, but I can't believe that
matters have changed so fundamentally that someone who is searching for
Léon Nikolaevitch Tolstoi does not want to know that the works by
this person can be found by searching under Tolstoy, Leo, graf,
1828-1910. It doesn't matter whether they happen to "agree with the
heading" or not. This is how the system works. I think people still want
and expect this level of access. The authority record provides this and
many other cross-references, and no other tool in the world does this,
that I am aware of.
In fact, when I have mentioned these sorts of issues to non-specialists,
they are absolutely shocked that Google does not--and cannot--provide
similar access. This should not be surprising since it seems to be
common sense that when someone searches for "wwi" in Google, the result
they are looking at is everything on World War 1. Of course, this is not
at all true and when you point it out to them, it's not that complicated
of an issue to grasp and they begin to see what is happening. I think
one of our tasks is to demonstrate this to people so that they
themselves can see precisely how a Google result differs from the result
in a traditional bibliographical tool.
Certainly, this system could be, and should be, improved . Authority
records are some of the most underused tools in the library. They should
have many more cross-references, especially the subject headings. They
should be included with keyword searches. The browse capability is very
poorly implemented in our current catalogs, and I would say, almost
useless. One bright spot is OCLC's Worldcat Identities which experiments
with some of this information.
Another issue is where you say, "I think that if we were to re-design
our data with systems in mind, rather than basing our data on the
cataloging rules, then we could greatly improve the catalog itself." I
am not sure I agree with this. It is my belief that we should look first
at the task, then design the information necessary to achieve that task,
and finally to design the computer system. I have been involved in
several projects where we wanted to include certain information, but it
was nixed because "the system can't utilize it." Of course, in just a
year or two, the system was modified or a new system came along that
could do what we wanted. It is my experience that systems will do what
you want today, so the focus should be on expanding our imaginations and
not on the capabilities of current computer systems. The system can be
built and we should insist on it.
In this case, the current cataloging rules are based on the tasks set so
long ago in Cutter's time. I personally believe that all of those tasks
("Objects" in his terminology) are still just as valid today as ever.
It's hard to believe, but all of the voluminous cataloging rules are
based on those few tasks! The way he suggests to solve those tasks (his
"Means") may not be the way to go in many cases today and should be
rethought.
Does anyone object to this? Is there any evidence that people today do
not want to know what a collection has by its authors, titles
, and subjects? Do they not want help in deciding which version
(edition) to choose? Of course, they want more than just this today (for
example, what is "a collection" today?), and that is fine, so this list
needs to be expanded. It has been my experience that people believe they
already have these capabilities in Google and Yahoo, but they do not.
Without evidence that people no longer need or want this level of
access, we should not eliminate what is there now.
James Weinheimer j.weinheimer_at_aur.edu
Director of Library and Information Services
The American University of Rome
Rome, Italy
Received on Wed Aug 22 2007 - 02:36:47 EDT