Library Journal "Digital Libraries" Columns 1997-2007, Roy Tennant
Please note: these columns are an archive of my Library Journal column from 1997-2007. They have not been altered in content, so please keep in mind some of this content will be out of date.
Library Catalogs: The Wrong Solution
02/15/2003
Most integrated library systems, as they are currently configured and used, should be removed from public view. Before I say why, let me be clear that I think the integrated library system serves a very important, albeit limited, role.
An integrated library system should serve as a key piece of the infrastructure of a library, handling such tasks as materials acquisition, cataloging (including holdings, of course), and circulation. The integrated library system should be a complete and accurate recording of a local library's holdings. It should not be presented to users as the primary system for locating information. It fails badly at that important job.
Lack of content
The central problem of almost any library catalog system is that it typically includes only information about the books and journals held by a particular library. Most do not provide access to journal article indexes, web search engines, or even selective web directories like the Librarians' Index to the Internet. If they do offer such access, it is only via links to these services.
The library catalog is far from one-stop shopping for information. Although we acknowledge that fact to each other, we still treat it as if it were the best place in the universe to begin a search. Most of us give the catalog a place of great prominence on our web pages. But information for each book is limited to the author, title, and a few subject headings. Seldom can book reviews, jacket summaries, recommendations, or tables of contents be found or anything at all to help users determine if they want the material.
Lack of coverage
Most catalogs do not allow patrons to discover even all the books that are available to them. If you're lucky, your catalog may cover the collections of those libraries with which you have close ties--such as a regional network. But that leaves out all those items that could be requested via interlibrary loan. As Steve Coffman pointed out in his "Building Earth's Largest Library" article, we must show our users the universe that is open to them, highlight the items most accessible, and provide an estimate of how long it would take to obtain other items.
Inability to increase coverage
Despite some well-meaning attempts to smash everything of interest into the library catalog, the fact remains that most integrated library systems expect MARC records and MARC records only. This means that whatever we want to put into the catalog must be described using MARC and AACR2 (see "[123]Marc Must Die ," LJ 10/15/02, p. 26ff.).
This is a barrier to dramatically increasing the scope of a catalog system, even if we decided to do it. How would you, for example, use the Open Archives Initiative Harvesting Protocol to crawl the bibliographic records of remote repositories and make them searchable within your library catalog? It can't be done, and it shouldn't. The library catalog should be a record of a given library's holdings. Period.
User interface hostility
Recently I used the library catalogs of two public libraries, new products from two major library vendors. A link on one catalog said "Knowledge Portal," whatever that was supposed to mean. Clicking on it brought you to two choices: Z39.50 Bibliographic Sites and the World Wide Web. No public library user will have the faintest clue what Z39.50 is. The other catalog launched a Java applet that before long froze my web browser so badly I was forced to shut the program down.
Pick a popular book and pretend you are a library patron. Choose three to five libraries at random from the lib-web-cats site (pick catalogs that are not using your system) and attempt to find your book. Try as much as possible to see the system through the eyes of your patrons a teenager, a retiree, or an older faculty member. You may not always like what you see. Now go back to your own system and try the same thing.
What should the public see?
Our users deserve an information system that helps them find all different kinds of resources--books, articles, web pages, working papers in institutional repositories--and gives them the tools to focus in on what they want. This is not, and should not be, the library catalog. It must communicate with the catalog, but it will also need to interface with other information systems, such as vendor databases and web search engines.
What will such a tool look like? We are seeing the beginnings of such a tool in the current offerings of cross-database search tools from a few vendors (see "[124]Cross-Database Search ," LJ 10/15/01, p. 29ff). We are in the early stages of developing the kind of robust, user-friendly tool that will be required before we can pull our catalogs from public view. Meanwhile, we can begin by making what we have easier to understand and use.