Printer Friendly

MARC and INMAGIC: tools for developing an in-house catalog; retrospective conversion is achieved with INMAGIC.


As consultants to special libraries, we have helped a number of our clients bring their collections into online catalogs. The process, known as retrospective conversion, while widely covered in the library literature, is still poorly understood by many librarians.

This article attempts to shed some light on the subject by illustrating the steps involved in converting MARC cataloging data into a local catalog using the INMAGIC family of software products. We pay particular attention to the needs and concerns of small libraries.

An Overview of MARC

Thanks to the Library of Congress (LC), there is a strong likelihood that bibliographic records for most commercially published items in a library's collection already exist in computerized form. MARC (Machine Readable Catalog) started in 1966 as an experiment by LC to determine if it were possible to produce a standardized machine-readable catalog record that could be manipulated and reformatted by libraries to serve local practices and needs.

Implemented in June 1968, MARC is now widely accepted by the English-language library community as the standard for machine-readable cataloging and has formed the basis for many cooperative cataloging programs and services including LC MARC, OCLC, the Research Libraries Information Network (RLIN), the Washington Library Network (WLN), and Utlas CATSS (Catalog Support Service).

As a direct result of the wide availability of MARC records, original cataloging is being supplanted by "copy cataloging," the process of matching items in a library collection against a MARC database, and copying the matches to a local system or otherwise tagging them for local use. MARC records not only obviate the need to keyboard cataloging data into a computer system, but also reduce the problem of bibliographic quality control and verification.

A library whose holdings consist primarily of commercially published monographs and serials has few compelling reasons to undertake original cataloging and data entry, unless its accumulated collection is extremely small. Even libraries with "home-grown" cataloging data requirements can successfully use MARC records, as they can easily be adapted to reflect in-house needs.

Sources of MARC

The most widely known sources of MARC records are not necessarily the best choice for small libraries preparing for a retrospective conversion. Online union catalogs, CD-ROM MARC databases, and tape subscription services are not only costly, but also difficult for small libraries to justify in the long term owing to their modest ongoing cataloging needs.

OCLC, RLIN, Utlas CATSS, WLN, and other national, regional, and local bibliographic resource sharing services offer member libraries online access to extensive union catalogs in MARC format, from which catalog data can be captured for use in local systems. However, the price of membership is often discouraging to small libraries, entailing hardware, software, and database access fees starting at several thousands of dollars annually.

MARC databases on CD-ROM, such as BiblioFile (The Library Corporation, Inwood, West Virginia) and Laser-Cat (The Western Library Network, Washington) are somewhat less costly to purchase and maintain, and offer the benefit of greater local control over the conversion process, but they are also far more limited in size and scope than most online union catalogs.

Contrast BiblioFile's three million record LC MARC English database, consisting mainly of books, with OCLC's twenty-one million record Union Catalog of monographs, serials, audio-visual materials and other nonbook formats. Tape subscriptions, such as LC's Copyright Cataloging Tape Service, are intended for use by large library systems and consortia and are priced accordingly -$50,000 for the LC retrospective file, excluding annual updates.

Online bibliographic reference files such as OCLC EPIC, OCLC EASI Reference on BRS, LC MARC on DIALOG and WILSONLINE, and Utlas REMARC on DIALOG, also contain millions of catalog records, but were never intended as vehicles for retrospective conversion. Most of these files are protected by vendor copyright restrictions prohibiting downloading records for local manipulation. Moreover, records captured in this manner are costly and may require extensive editing, as the formats in which they can be written to disk are incompatible with most

commercially available MARC translation programs.

When considering these sources of MARC records, cost is only the most obvious issue. Staff resources are another. In-house conversion is a labor-intensive task, requiring one or more skilled persons to search for, match, and download MARC records for each item in the collection. While this procedure is still preferable to original cataloging, we know of few small libraries willing to devote personnel to the task, given a reasonable alternative.

MARC Copy Cataloging Services

A mote practical alternative for small libraries (and in many cases, for larger ones as well) is to purchase MARC records from a retrospective conversion service or copy cataloging vendor.

Vendors offering batch copy cataloging services come in several shapes and sizes. They include the Library of Congress MARC Distribution Service (Washington, D-C), MARCIVE, Inc. (San Antonio, Texas), SOLINET Inc. (Atlanta, Georgia), The Catalog Card Corporation of America (Burnsville, Maryland), and Retro Link Associates (Provo, Utah). OCLC offers several services to nonmember libraries - RETROCON, MICROPRO, and MICROCON*PRO.

Vendor-contracted copy cataloging is relatively straightforward. The client library submits to the vendor a copy of its shelf list or equivalent, e.g., inventory list, hard copy citation list, bibliographic worksheets, or list of standard bibliographic numbers. The vendor matches each item in the collection against its own MARC database. The MARC records are then downloaded to disk or tape and delivered to the client for local processing. Additional services may include entry of local data, original cataloging, deduplication, authority control, and the production of spin-off products, such as spine and pocket labels, barcode labels, and book or fiche catalogs.

Basic costs, excluding add-on services, set-up fees, and media processing charges, averaged less than $1.00 per record at the time this article was written.

A thorough discussion of criteria for selecting an appropriate vendor for a retrospective conversion would require a separate article. Briefly stated, they include: size and scope of the vendor's database; quality assurance and technical support; availability and pricing of add-on products and services; flexibility in conversion scheduling and turnaround, and data delivery options (e.g., disk vs. tape).

One of the most important criteria in choosing a vendor is record-matching requirements. Several vendors - for example, LC and MARCIVE - can only identify items by LC Card Number (LCCN), International Standard Book Number (ISBN), or International Standard Serial Number (ISSN). Vendors with more flexible searching options will attempt to match records on author/title and other access points when standard numbers are lacking.

MARC Communications Format

MARC records from copy cataloging vendors are usually delivered in US-MARC communications format. Figure 1 illustrates a typical record.

Records in MARC communications format are not intended to be deciphered with the human eye. However, they do contain a wealth of bibliographic information that can be understood by many local library automation systems.

Local Software Issues

Because the majority of library automation tools do not use MARC as a native format, the software you select to develop the online catalog should have a MARC interface, i.e., a translation program to convert MARC records to a format recognizable by the local software. MARC interfaces are included as standard or optional features in a variety of PC-based library automation products, many of which can be found in the Directory of Information Management Software for Libraries, Information Centers, Record Centers. [1]

The software product used in this article to illustrate the migration of MARC records to a local system is INMAGIC, among the most popular library automation and bibliographic information management tools available in today's market, with over 7,000 users at 5,000 sites worldwide.

Version 7.2, released in April 1990, expanded the software's capabilities to include a number of features that greatly facilitate online catalog processing, including a combined search/modify environment, global record modification, extended batch processing features, userdefinable stop words, and nine sorting options. The basic system sells for $975 for a single user license ($125 to upgrade). INMAGIC is also available for PC-based networks and the Digital Equipment Corporation VAX.

The INMAGIC family features three "add-ons" particularly applicable to online catalog development. The first, the Biblio Guide ($145) is a set of models for computerizing cataloging, serials management, acquisitions, and loans. Tbe online catalog model includes about two dozen hard copy and screen reports, including four book catalog formats, three varieties of catalog cards, and several types of spine and book labels.

SearchMAGIC (from $395) is a menu-driven "point-and-shoot" interface to INMAGIC that fulfills the searching requirements of both the unskilled patron and the veteran reference librarian.

A third product, the MARC Adaptor ($125), is a MARC interface as described above. It will translate records in USMARC communications format as well as OCLC or RLIN screen-dump formats (the latter using the PASS command).

The Process of Conversion

The process of conversion entails several discrete steps, as outlined below. With the exception of the first step, the tasks involved are roughly the same for all PC-based library automation tools.

* adapting the online catalog structure

* defining conversion requirements

* converting the MARC records

* post-processing

Adapting the Online Catalog Structure

Unlike many turnkey library automation tools, INMAGIC allows the user to customize an online catalog in response to local library needs. So, for example, a library wishing to adapt the MARC record structure by adding fields of local interest (unique codes, local descriptors, or internal notations) can easily do so without the assistance of a programmer. Serials, analytics, audio-visual materials, and other nonbook formats happily co-exist with monographs in the same catalog.

The annotated catalog data structure shown in Figure 2 was adapted from the IMAGIC Biblio Guide. Each field is of unlimited length and can be repeated an unlimited number of times.

Defining Conversion Requirements MARC interfaces allow the user to determine which MARC fields to capture and how they will be processed by INMAGIC. This is accomplished by means of a conversion map. A sample INMAGIC MARC Adaptor conversion map is shown in Figure 3.

Predefined conversion maps are often included with the MARC interface software. However, in order to modify the defaults, a librarian must be familiar with MARC field tags and their meanings. Detailed information about MARC can be obtained from copy cataloging vendors, the Library of Congress, or from a variety of independent sources, such as Crawford's MARC for Library Use. [2]

The conversion map shown in Figure 3 lists each INMAGIC "target" field as defined in the catalog data structure, with the equivalent MARC field or subfield and appropriate processing instructions. For example, the line "CALL 050" instructs the MARC Adaptor to place the contents of MARC field 050 (Library of Congress Call Number) into the INMAGIC field CALL. Similarly, TI 245 $a" means to place the contents of subfield "a" of MARC field 245 (Title Statement) in the IMAGIC field TI.

The librarian can choose to capture as many or as few MARC fields as desired. Multiple MARC fields can be concatenated into a single INMAGIC field. Subfields of a single MARC field can also be split into two or more INMAGIC fields. The map also allows for automatic sequential numbering of records (ID @NUMBER) and the entry of global default values, for example, the text "Main Library, 4th Floor" in the LOCN Location) field.

Converting the MARC Records

Converting MARC records to the local software's native format is the simplest aspect of the process, often requiring just a few keystrokes to instruct the program to begin the conversion. This step is also relatively quick. Records processed by the INMAGIC MARC Adapter, for example, convert at the rate of less than a second each. MARC records received from a vendor on floppy disk are usually processed in batches of several hundred at a time.

Some MARC interfaces channel the records directly into the catalog database. Tbe INMAGIC MARC Adaptor creates an interim ASCII file, which can then be edited (e.g., spell-checked) or otherwise manipulated before the records are batch added to the catalog.

The MARC record shown in Figure 1 is shown again, after conversion to INMAGIC, in Figure 4. The record is displayed in a book catalog format.

Post-Processing Phase

The requirements of post-processing will vary from library to library. Depending on the services provided by the copy cataloging vendor, they may include one or more of the following:

* record deduplication

* normalization of names, uniform and series tides and subject headings

* addition of local data

* identification and resolution of mismatches and substandard records

A library's selection of local software will affect post-processing productivity. With INMAGIC, for example, it is easy to view or print the contents of any field in the catalog to determine if the field values are valid and consistent. Necessary changes can be made recordby-record or globally across the database.

A Word on PACs

A public access catalog (PAC) is often the ultimate goal of converting a library catalog to machine-readable form. As the American work force becomes more comfortable with computerized tools, small and special libraries face increasing pressure to provide their patrons with online access to internal information resources.

Figure 5 illustrates a screen from a public access catalog developed under SearchMAGIC.

If your online catalog is intended for public access, the software selected to develop the catalog should feature a search interface simple enough for a casual searcher to use, while also providing advanced search capabilities for finding the "needle in the haystack." Multi-user search access across a network is also important as a means of accommodating a growing user base.


1. Pamela R. Cibbarelli and Edward John Kazlauskas, Directory of Information Management Software for Libraries, Information Centers, Records Centers. Felton, CA: Pacific Information, C1990.

2. Walt Crawford, MARC for Library Use, Understanding Integrated USMARC, Second edition. New York: G.K Hall & Co., 1989.

(Figures Omitted)
COPYRIGHT 1990 Information Today, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Velardi, Marie; Lafferty, James
Publication:Computers in Libraries
Date:Dec 1, 1990
Previous Article:INMAGIC version 7.2; it may not be sleek, but this sturdy package can handle special needs.
Next Article:Book catalogs using INMAGIC and WordPerfect; with INMAGIC and WordPerfect, producing a book catalog is easy.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters