Here are some things that need to change:
1) The "authorized heading" is also the identifier for the thing or
person being identified. That means that if you change cataloging rules,
you change the identifier. This is a very bad practice, because it
breaks all of the links between bib records and authority records.
2) There is nothing to tell you, on a field by field basis, what
language is represented by the heading. So you can't algorithmically
select the languages that are important to your community for
cross-referencing.
3) The name authority record contains virtually nothing that could
interest a user -- no bio about the author, nothing to help the user
know who it is we are talking about. Giving an author's dates as a way
to differentiate between authors may be fine when you have only two
names and one has a 18th century date and the other has a 20th century
date. But when you have authors who are contemporary (with each other) a
few years difference in date of birth isn't useful because users are
highly unlikely to know if an author was born in 1956 or 1958.
OK, there are just 3 examples. and I havent' even touched on subject
authorities. Nor have I touched on the issues of non-unique entries in
the name authority 4xx's (which cannot be resolved algorithmically with
the data we have today). Or lots of other things.
Believe me, anyone who has worked to make use of authority records in a
system has a litany of issues.
kc
Weinheimer Jim wrote:
>> James, it's not that authorities *shouldn't* be used, it's that the
>> authority records, and their contents, make that use very difficult.
>> That's why they aren't used well in our systems. With some re-design, I
>> think they could be much more useful.
>>
>> And the problem isn't with the cataloging rules (e.g. what is a title?)
>> it's with how the results of those rules are turned into data elements.
>
> Sorry Karen, I still don't understand the problem here. The authority records are difficult to use, I agree: finally trash the old ISO 2709 format, we could do away with much of the fixed field information and nobody would know the difference, except catalogers and computer programmers, who would breathe a sigh of relief. I think that many of the subfields could be done away with as well, but now I am bordering on heresy with many people....
>
> But the contents of the authority records is something else. What would you change other than some fixed field information? The heading is needed (although this could be expanded as well. I gave a paper in Norway on this a few years ago). The references are needed, with see and see-alsos, and broader terms and narrower terms. Then there are scope notes and sources of the information, plus a very small amount of technical information here and there.
>
> I know that some people want more information in authority files. For example, at one place I worked, the big thing was "ontologies" which attempt to expand the current cross-reference structure, but that becomes very tricky, not from the system point of view, but from the clarity point of view. For example, you may want to have "cause-effect" relationships, but then you have a specific example of the relationship between "deforestation" and "desertification." Either one could be a cause or effect of the other. And if you have the cataloger determine that relationship at the time of cataloging (which one causes which?) that would take substantially more time as the cataloger would have to actually read and digest the document. Even experts might not agree after reading it. I always felt it wouldn't be worth the effort. But....
>
> Still, the authority records could be improved and utilized far more effectively, but I really do not believe that it is a problem with their contents, other than simply that more information would be very useful.
>
> Finally, I also don't see the problem with turning information in the present bibliographic records into data elements, other than format, and that is a technical computer problem, easily solved. I admit there are crosswalks between formats that may not work 100% perfectly. MARC21 is also deficient in many areas. But you can always find a field somewhere, and I have hopes for a "super format" someday. Still, this doesn't have to do with the contents.
>
> Cataloging and catalogs must change, and change profoundly, but I think they should change in other ways. Catalogers must stop caring about the fine points of punctuation and focus on the real issues of access and description in a networked world.
>
> Regards,
> Jim
>
> James Weinheimer j.weinheimer_at_aur.edu
> Director of Library and Information Services
> The American University of Rome
> Rome, Italy
>
>
>
>
>
--
-----------------------------------
Karen Coyle / Digital Library Consultant
kcoyle@kcoyle.net http://www.kcoyle.net
ph.: 510-540-7596 skype: kcoylenet
fx.: 510-848-3913
mo.: 510-435-8234
------------------------------------
Received on Wed Aug 22 2007 - 08:40:03 EDT