![]() |
SUBJECT HEADINGS |
|
Alva T. Stone Florida State University atstone@law.fsu.edu |
Does the cataloger's experience with subject analysis and with the application of headings and classification result in a certain propensity for Web site construction? You bet it does! At the May 1998 OCLC seminar, Knowledge Access Management, I heard the following prediction. Librarians in the future will become less the curators of information, and more the creators of information. Already many librarians are building, for example, detailed descriptions and more widespread access to collections or documents which are unique to the library or its parent organization. If the Internet is to be the means for the access, why cannot our catalogers learn a little HTML or use a web-editing tool, and then apply our skills for description and selecting access points to this new environment? Anna Belle Leiserson made a similar point in her column in the last TSLL issue; I'd like to extend the argument, and address in particular the similarities between Subject Headings practice and Web site development.
Our cataloging department is starting to build a Web site related to the library's unique collection of United States Supreme Court Justices' autographs. We have autographs of all but two of the historical justices. Some of them are on documents, some on private correspondence, on calling cards, or on photographs. The entire collection is valued at over $25,000. It is our goal to create a Web site that will educate and inform students and public alike, all over the world, as well as bring exposure to the University and the College of Law for this unique and very special collection. In the initial planning stages, we have been noticing some overlap with traditional cataloging principles in these areas: Content, Categories, and Links.
Content. Consistency, both in form and in structure, is an important objective in the creation of subject headings. Most of the "how to" manuals on Web site construction also emphasize consistency. Particularly for a site that has numerous pages (such as one page for each of over 100 justices!), those pages will be easier to use and navigate if the format and content are uniform, from page to page. Catalogers are accustomed to using standardized formats, and today's dynamic HTML allows the design of style sheets or "library" data which resemble the workforms or the customized "constant data" macro's that catalogers use with the bibliographic utilities. Relevance is a big factor, of course. Searches should be made to ascertain whether or not the new Web site would be adding anything new to the Web world. (For example, we have noted that both Cornell Law Library and Northwestern University have already mounted some extensive data about the individual US Supreme Court justices; hence, we may not want to duplicate their data, but rather, refer to it, as appropriate.) We may use a search engine to find Web sites with content similar to our topic, in order to make decisions about the content for our own site, or possible hyperlinks we may want to add. At this point, the cataloger's skill at delineating and distinguishing between similar names or concepts comes in handy. For instance, we might find that adding the birth year of a Supreme Court justice to our search gives us more meaningful results. "Authorities" experience also lends itself to the choice of alternative access points. Although today's search engines are sophisticated enough to use the entire text of a web document for access, the cataloger might want to think of synonyms or alternative phrasing for the document's key terms, and add these in the HTML Head space. This will in effect make additional access points not unlike the action of assigning prescribed subject headings and their "see" or "search under" references to a catalog record.
Categories. Some obvious categories for the frames or headings we might consider for our Web site are: biography, portraits, and of course, the autograph(s). For the cataloger, the impulse to categorize and label is nearly instinctive. When we examine a Web page, looking for ideas or for possible hyperlinks to include in our project, immediately, almost subconsciously, we think of the category or class in which it belongs. A caption, or heading, for this class comes to mind, e.g. "Homes and haunts," or "Anniversaries, celebrations, etc." regarding individual justices, and, Voila! — the subject heading has been assigned. Of course, in the final packaging or design of the Web site, those precise labels may not be used. In any event, the cataloger's training and experience make the process of organizing the different types of Web data feel like a familiar step.
Links. Anyone who has used the Internet has surely encountered hyperlinks that allow the user to redirect her or his search to another page of the Web site, or to a different (but relatedtopic) Web site altogether. The principle behind such linking is much like that of the "see also" or "search also under" references used with traditional Subject Headings; the difference is that hyperlinks are often embedded within text inside paragraphs. When the cataloger chooses a subject heading, references from the heading's broader subjects and to the heading's narrower subjects and related subjects are added to the catalog, to help the user find the associated subjects. Likewise, when the content of a Web site alludes to a topic that is related, but beyond the scope of the present site's coverage, an active hyperlink may be added to help the user find more detailed information elsewhere. The rules and guidelines for selecting Web pages or sites for such hyperlinks are not yet standardized as are the rules for creating subject thesauri (or, a subject heading list such as LCSH). But, perhaps they should be! The guidelines for hierarchical references that catalogers use focus on three types of relationships: genusspecies (or class-class member); wholepart; and, instance. In the case of our illustrative, developing Web site, a link to a Web site on the federal Judiciary generally might represent the speciesgenus relationship; a link from the Supreme Court Web site to a site about the Chief Justices of said Court could be considered to reflect a whole/part relationship; and a link from the page for Samuel Chase (1741-1811) to a Web site on impeached officials would illustrate a kind of "instance" relationship.
This discussion of content, categories and links in the design and development of Web sites, and their relation to traditional subject cataloging or authorities practices is by no means a comprehensive study. It is only a few thoughts and considerations meant to stimulate other catalogers to get involved in creating Web sites for their libraries. If other law catalogers have already built such Web sites, I hope you will share some "do's and don'ts" with us!