Foskett, A.C. (1996). Chapter 8: Alphabetical subject headings: Cutter to Austin (p. 123-146). The subject approach to information, 5th ed. London: Library Association Publishing.
Describes approaches to creating subject headings. Cutter advocated natural language headings with cross-references to create multiple access points. Kaiser created rules for separating a complex subject into standardized headings. Metcalfe attempted to combine the ideas of Cutter and Kaiser. Coates extended Kaiser's rules using a psychological approach. PRECIS, COMPASS, and NEPHIS are computer-based systems for automatic creation of a subject index.
Foskett, A.C. (1996). Chapter 23: Library of Congress Subject Headings (p. 336-347). The subject approach to information, 5th ed. London: Library Association Publishing.
Discusses the structure of the Library of Congress Subject Headings (LCSH). The 1995 edition has grown to 213,800 headings. Recent changes have removed some ambiguity in the headings, but there are still many anomolies present.
3/22/00 abstract and discusison
Dykstra, M. (1988). LC subject headings disguised as a thesaurus. Library Journal 113(4), 42-46.
The Library of Congress Subject Headings (LCSH) have adopted codes commonly used for thesauri to indicate relationships between subjects. Since subjects are not the same as thesaurus terms, this action has violated several thesaurus standards and caused major problems for LCSH.
(response to Shirley's comment that since subjects are not terms, borrowing the thesaurus codes doesn't actually violate the standards) If LCSH were to continue using the old codes, users wouldn't have a chance (I know I wouldn't). At one point, she states that the new codes "have resulted in chaos rather than conformity." Would she have us create a completely new vocabulary for every slight difference in usage? Wellisch showed us an impressive list of uses for the term "index". Does Dykstra want us to define a standard meaning for "index" and create a new word for all of the other uses? Probably not, but this is where her subject vs. term argument would lead us. I suppose she could make an argument for changing BT/NT to BS/NS, but there isn't much to be gained from this.
On another note, towards the end of the article, she states "The future text analysis systems will be expert systems, and by definition an expert system is a rule-based system." Well, the second half of this statement is correct. But she doesn't provide any evidence for the first half. Some current text analysis sytems are expert systems, but many (most?) are pattern matchers and statistical systems, which are far from rule-based.
Taylor, A. G. (1995). On the subject of subjects. Journal of Academic Librarianship 21(6), 484-491.
Librarians used to think that subject searching was seldom used. (Strange, since card catalogs usually had subject cards.) They were surprised when recent studies indicated that a majority of searches were subject searches. With the advent of the internet, subject searching is becoming even more important, although there is a tendency to replace it with keyword searching. A controlled vocabulary (subjects) can produce better searh results than keywords, but keywords can be assigned more quickly and inexpensively to documents. In order to use the full power of subject searching, the principle of ``specific entry'' must be obeyed. Some people still think of classification (or subject assignment) as a way to locate items on shelves, rather than a way to group related items. Current OPACs do not assist users in the use of subject headings as well as they could. For the forseeable future, the Library of Congress Subject Headings will continue to be the dominant set of subject headings.
3/22/00 summary and discussion
Levy, D. M. (1995). Cataloguing in the digital order: Paper regarding the future of cataloguing, from the Digital Libraries 95 conference. Available on the web.
Cataloging is the process of organizing a collection of documents to facilitate their location and use. The cataloger's job is to create a refined representation of a document regardless of any irregularies. The catalog must be maintained as cultural thought changes and new types of documents are created. Digital documents pose a challenge for cataloging, since documents are created much more quickly than they can be cataloged, and the content of a document may be dynamic.
This article was very helpful. I wish I'd read it earlier... Until I began this class, I had always assumed that someone (the publisher? LC?) gave a book a classification, and it was placed on the shelf. It sounded like a fairly simple process. I never thought about people in each library doing the classification/cataloging. With the advent of the web, classification is becoming more visible, mostly through its absence.
This leads me to some questions about cataloging:
Is there a difference between a cataloger and a classifier? Or is it that classification is just one of the jobs of a cataloger?
How many libraries employ their own catalogers/classifiers?
How often does a library acquire a document that hasn't already been given a classification by LC or someone else? Since IU uses LC numbers, what do the local catalogers actually do? Do they make changes to the LC catalog entries? Or do they only work with things that LC hasn't dealt with?
Bush, V. (1996/1945). As we may think. Interactions, 3(2), 35-46. Originally published in Atlantic Monthly, 176 (1), 101-108.].
Discusses future directions for science, concentrating on the storage and retrieval of knowledge.
Manufacturing technology is improving to the point where machines can assist in storing and retrieving knowledge. Photographic technology will soon be able to fit the entirety of the Encyclopedia Britannica on a single page. ``Memex'' is a theoretical machine that allows for rapid storage and retrieval of documents. Documents may be viewed sequentially or by following user-defined trails.
science, photography, dry photography, future, machines, punched cards, selection, memex, microfilm, trails
electronic publishing, micropublishing, information explosion, computer equipment, data storage, library automation, innovation, forecasting, machine aided indexing, photographs
Possible LCC Classes
Q 350-390 Information theory Z 681-681.3 Reproduction of library materials. Storage media of library materials Z 699-699.5 Machine methods of information and retrieval. Mechanized bibliographic control
Possible DDC Classes
(the OCLC resource no longer exists) After poking around in (http://www.tnrdlib.bc.ca/dewey.html) I found:
The problem with using the classification systems is that this article (being speculative) doesn't really fit into any of the available classes.
3/26/00 group work
We discussed the group's answers for the above, and came up with some consensus answers. This was fairly difficult, since there was little overlap in the answers that the four of us brought to class.
3/26/00 class notes
Official Lecture Notes
The Library of Congress uses the system of literary warrant -- create a scheme that represents only the documents in the current collection, not all possible documents.
LCSH is now an international standard. Everything from LC is alphabetical.
Check out Andy Clark's Being There.
Cutter's principles for subject headings
IUCat folds related terms and narrower terms together into a ``see also'' page.
``Knowledge is generated by discourse in an attempt to reach consensus.''
Nothing is required reading for session 15; we should be working on debates and portfolios.