Peter S. Graham, Rutgers University Libraries
PROLOGUE: Access to information is cast in entirely new terms with the advent of the internet. The ability to move complex information transnationally more quickly than from one's own floppy disk drive will change the face of scholarly communication. Libraries must no longer think electronically only in terms of bibliographic access, but of full-text (and full-image and full-data base) access; the recent Mellon Report provides an eloquent statement of the need.# Guiding users to information that is not local must transform what we think of as our catalog.
It used to be a truism that the day an OPAC was opened to the public was the day the library began the search for the next system. Perhaps that should no longer be assumed while the library community determines what kind of access tools our clienteles need and what we want to provide. The next tool must take into account local and remote information, text and images, access to data and actual data, information itself and information-handling tools, data and programs. Our staffs need to reconfigure themselves to provide expertise in the tools both for our print heritage and for our new electronic heritage.
CONTINUITY AND INTEGRATION: The important point to remember is that, from the users' point of view, the catalog is only one of a repertoire of tools. (They don't expect one-stop shopping; we're on their mall.) We who provide this tool must remember the necessity to integrate it into users' other resources.
What follows is drawn from experience in the academic environment, that is, where catalog use is typically for a specific purpose: to locate information for use. Thus the catalog, as in the past, is a bridge between the recorded information sought and the end product of the information's use (whether in the user's head or as another recorded product). The catalog is a means to an end.#
The next stage of the catalog's evolution is possible to envision from the building blocks now available and in use. The mid-decade catalog will be different from the existing online catalog in several important respects:
The present catalog (generalizations; changes are already occurring):
The next catalog additionally:
The catalog tool, as opposed to many other access means becoming available, will continue to provide added value by organizing information in advance of need. The best example is authority control, which brings together names, titles and subjects under established headings with a cross-reference structure from other headings possibly of use. The catalog will continue to be a pre-coordinated, consistent, and syndetic structure requiring skilled intellectual effort from people known as librarians.
THE MID-DECADE CATALOG IN ITS ENVIRONMENT: The following definitions are keyed to the attached diagram of elements comprising the next catalog environment (see Figure [not yet available]). A dotted line surrounds those elements and portions of elements that comprise what we can think of as the Mid-Decade Catalog (and the end-of-decade tools will be different again).
A document, in this discussion, is loosely defined as a recorded quantum of information, whether electronic or not: a book, an article or other text, a graphic, a sound passage, or an archival record. It is the search for documents that librarians wish to serve by providing catalogs and other access mechanisms. A workstation is a user's networked computer (whether personal or institutionally located), what we now often call a PC or microcomputer. Its environment may be Windows, Macintosh, OS/2, Unix, PowerOpen or another network-sophisticated operating system. For many purposes the new catalog will support dumb-terminal functions as well.
The user will "see" as she or he approaches the catalog a common interface for several information-seeking uses. Librarians have with good reason come to value highly the consistency of presentation, as users then need less training and can more quickly shift fruitfully from environment to environment. (The common interface will be common to an institution, not more globally, as it will depend upon a local system implementation which itself is dependent upon local vendor choices, budgets, and local systems talent.) A GUI (graphical user interface) will be a desirable option where it is practical for user workstation client software to be made available.
The common interface will be driven by the Public Access Computer System (PACS),# of which the online public access catalog search engine (OPAC Engine) will be a component. The OPAC engine will provide the interface that most users now think of as the catalog: the prompts for searching, the display of search results, locally-provided guides to usage, and context-sensitive help.
Other PACS components may well include gopher client software, a WAIS client, and clients for searching vendor-provided bibliographic data bases such as Medline, the Wilson indexes, UnCover, PAIS, and ChemAbstracts. Note that the location of such data bases (on local computers or remote) is irrelevant to this model. The Z39.50 information query standard will be a basic component of the PACS and will underlie the common interface.
The user will have other access modes available from the workstation, such as FTP tools, Archie, World Wide Web, and the like.# The PACS may not provide these directly, as their interfaces don't lend themselves to the commonality of the other tools.
The local catalog is created at the institution, using catalog system tools much like today's: record sources include utilities, local creation, other institutions, batch loads (e.g. GPO or CRL), and local acquisitions information. The definition of the records to be included will expand to include those for artifactual holdings located elsewhere, and for electronic documents located locally or elsewhere on the network. Authority control remains an integral part of the catalog system and is automated. Allied catalogs, available for search directly from the OPAC engine in the PACS using the Z39.50 standard, include the institution's cataloging utility (OCLC, RLIN, WLN, etc.) and catalogs of other institutions, geographically or otherwise proximate, which have agreed to accept the traffic likely from such convenient access.
Documents, from the user's point of view, will become of two kinds: those the user will have immediately available using links provided by the PACSÊtools, and those that require an additional effort by the user to obtain. The former comprise electronic documents that are located on network servers (locally or elsewhere) and are linked to catalog entries. The user will invoke the linkage, following the successful search, to download the document (text, image, etc.) into his or her workstation for further use. This function will be dependent upon implementation of the Universal Record Locator (or something similar) now being constructed within the Internet Engineering Task Force#. It will also depend upon a political, structural development outside the purview of the catalog tool and the local environment. That is, electronic document linkage and retrieval will depend upon the establishment of electronic repositories prepared to make organized, selected electronic information available, in protected, refreshed and authenticated form, with a commitment for the long term (i.e. longer than human careers). This is of course the library mission.
Documents that a user must obtain through other means will themselves be broadly of two kinds: other electronic documents and artifactual materials. Other electronic documents will include those made available by vendors or others in non-standard or ephemeral environments, or through tools that are otherwise not appropriate for linkage through the PACS, and that require specific retrieval action by the user.
Artifactual materials include of course what we've all been familiar with up to this time: books, journals, manuscripts, recordings and other information resources which are inseparably linked to the objects that are their medium, and therefore exist in space and require specific physical handling to use. In the mid-decade these will remain by far the largest category of document of user interest; by end of decade we cannot be sure.
Gopher sources, local and remote, will include both documents and further information locators provided at other gopher servers. Sometimes these locators will themselves be library catalogs or other finding tools that for various reasons (choice or constraint) may not be made available through the OPAC engine. Sometimes the gopher sources will be documents. Some gopher documents will be found through use of the OPAC engine but will only be retrievable through the gopher tool itself. Other tools will no doubt be available by mid-decade.
CONCLUSION: As is evident, the definition of what should fall within and without the dotted line is open to discussion to a degree much greater than in the past. Local implementations will reflect different views, pragmatic decisions and local capabilities. To make these complex local decisions, the need will be ever greater for more and more able catalog librarians -- those who design catalog environments, those who implement them and those who provide their intellectual contents.
In the 80s we breathed a sigh of relief that we were beyond the homegrown bibliographic systems of the 60s and 70s (who remembers the quadraplanar structure?) and could rely on the vendors for orderly system development and support. We may now be back to a stage in which we cannot expect vendors to lead, and individual institutional experimentation and development will be inevitable and desirable (look at gopher, WAIS, electronic text centers at Virginia and Georgetown and Rutgers/Princeton, and the plethora of CD-ROM library networks). At the same time some coordination of goals and development is essential if we are not to flounder for years and end up with an unsystematic, unstandardized set of tools.
PG's Home Page -http://aultnis.rutgers.edu/PGHome.html.