Conference Report: MARC 21: Experiences, Challenges and Visions

Rowena Griem, Yale University

The “MARC 21 – Experiences, Challenges and Visions” workshop was organized by the Deutsche Nationalbibliothek (German National Library) and supported by the Deutsche Forschungsgemeinschaft (German Resource Foundation) or DFG, and the Andrew W. Mellon Foundation. There were over seventy participants representing Austria, Croatia, the Czech Republic, Germany, Latvia, the Netherlands, Norway, Poland, Russia, Slovenia, Spain, Sweden, Switzerland, the United Kingdom, and the United States.

There were three speakers from the United States: Julianne Beall and Sally McCallum from the Library of Congress (LC), and Glenn Patton from OCLC. Their presentations covered the history of Machine Readable Cataloging (MARC), the Machine Readable Bibliographic Information Committee (MARBI), and treatment of authority and classification data in MARC 21. Trond Aalberg from the Norges Teknisk-Naturvitenskapelige Universitet (Norwegian University of Science and Technology) spoke about MARC 21, the current version of MARC, and the Functional Requirements for Bibliographic Records (FRBR). All of the presentations are available on the workshop website.

This report is limited to the European speakers who described their experiences converting to MARC 21. The most common reason for adopting MARC 21 was to facilitate the sharing of records internationally, but the change was often accompanied by a sense of loss that a national tradition/practice was being abandoned, as well as some loss of features allowed by earlier locally designed formats. Despite these potential drawbacks, no one regretted adopting MARC 21. A significant contributor to this satisfaction is no doubt that MARBI has been willing to consider requests for adjustments to the format.

Paintin’ the Fence, the Long Way From MAB to MARC 21:
Experiences from Germany and Austria

Reinhold Heuvelmann, Deutsche Nationalbibliothek

Individual libraries and regional library systems have long used a variety of local systems in Germany. Maschinelles Austauschformat fuer Bibliotheken (MAB) was developed in 1973 and revised in 1994. From 1998–2004, the process of internationalization began with the introduction of Unicode, Dublin Core, FRBR, XML and the translation of MAB into English. The internationalization of library standards, i.e. the cataloging rules and formats, began in 2001 and is expected to continue until 2012. However, it should be noted that German libraries plan to retain existing data and functionality. For example, they will use MARC 21 but retain their current treatment of multipart monographs (MPMs), rather than adopting the AACR2 guidelines.

Before deciding to adopt MARC 21, an extensive feasibility study was conducted. This assessment was followed by lively discussions within the German library community. Ultimately, the proposal to convert to MARC 21 as the prescribed exchange format was successful; the changeover began in 2004 and should be completed by 2012.

The average cataloger will not be expected to learn MARC 21, and it is not taught in German library schools. So far, very few libraries use systems that are MARC-based, although the number is increasing. Understanding MARC 21 is relegated to the information technology professionals responsible for converting data to and from MARC 21. Records are created in one format and transferred into another. While this may seem awkward to us, German librarians have long been accustomed to having their data converted from MAB to local/internal and regional formats, and vice versa, so it does not seem unwieldy to them and is not overly problematic once the correct mapping has been worked out.

Heuvelmann pointed out the advantages of adopting MARC 21: the expansion of the community of cataloging expertise, the benefits of importing international data, the increased visibility of Austrian and German data, a better choice of systems, and the opportunity to participate in future MARC developments. To that end, representatives of German and Austrian libraries have presented a wish list to MARBI, requesting several adjustments and additions to current MARC 21 capabilities.

The disadvantages to adopting MARC 21 were that the documentation needs to be translated into German, an international framework makes it more difficult for experts to meet and discuss issues, the loss of functionality and ability to satisfy local needs, the inevitability of loose ends left by converting data, and the melancholy associated with change. Conversely, Heuvelmann pointed out that Germany has undergone other significant conversions recently in monetary systems, orthography, zip codes, etc. He noted that cooperation and trust are worth the struggle and converting to an international system is a way of bringing together the entire globe.

Changing the Record: The Transition from UKMARC to MARC 21

Alan Danskin, British Library (BL)

The inspiration to convert to MARC 21 came from rising costs and quantities of materials coupled with flat or reduced budgets. Danskin claimed that the United States is the largest source of copy cataloging and that the overlap in U.S. and U.K. publishing makes shared cataloging advantageous. The negatives were that the supposed cost benefits were not demonstrated and the U.S. version of MARC seemed “retrograde” to British eyes, since adopting it meant losing features allowed by UKMARC. However, this predicament may be temporary, since the British Library has proposed changes to MARBI for implementation into MARC 21.

A survey of UKMARC users to decide on the future of that system was undertaken in June 2000. This study showed that 57 percent of the respondents wanted to switch to MARC 21, despite the fact that 65 percent of the respondents were using UKMARC at the time.

The BL undertook the transition to MARC 21, including offering workshops countrywide. They continue to support UKMARC for a limited time to accommodate the needs of libraries that would be slower to make the change. British librarians made a conscious decision to convert all at once, which Danskin describes as “the big bang movement,” rather than undergo a slow transition or wait until XML or another system is adopted. British academic libraries have already switched to MARC 21. Public and small libraries were not really affected by the conversion, since their cataloging work is outsourced.

There have been some problems. The British National Bibliography is still compiled in UKMARC, and all new data must be converted from MARC 21 to UKMARC. Records created in MARC 21 have a different structure from those that were converted en masse from UKMARC. There are also some problems with punctuation. On the other hand, the BL successfully proposed that some elements of UKMARC be introduced into MARC 21, so some of the original disadvantages of adopting MARC 21 have begun to be addressed.

Migration to MARC 21: Swedish Experiences

Anders Cato, Kungl. Biblioteket (National Library of Sweden)

In the past, research and public libraries in Sweden used two different library formats, LIBRIS MARC and BTJ MARC, both based on UKMARC. In 1998, SWEMARC was developed. It was compatible with USMARC while addressing local needs, but communication with other library systems was complex, requiring the data to be converted from one format to the other. In December 1998, Dynix/Horizon was asked to develop a new national cataloging system for use with SWEMARC. The results were unsatisfactory, so the contract was canceled in September 2000, and the SWEMARC format was abandoned.

A week later, the Swedish National Library and research libraries decided to adopt MARC 21 and signed a contract with Endeavor/Voyager. The national library’s database was converted in 2001, and research libraries began using MARC 21 the following year. Public libraries have retained the local format, complicating the exchange of data nationally. Furthermore, even libraries that adopted MARC 21 for cataloging purposes continue to use their own subsystems for non-cataloging tasks, such as interlibrary loan. MARC 21 is not currently taught in Swedish library schools, but there is a continuing education program offered by the national library.

Swedish librarians were bothered that MARC 21 seemed old fashioned and required data to be entered more than once. This was not to be a significant problem due to the cut/paste function, but they still found it difficult to understand why coded information from the fixed fields also had to be written out. In addition, the necessity of entering ISBD punctuation manually seemed outdated.

Swedish libraries have been using a Swedish version of AACR2 since 1983, and research libraries plan to eventually abandon local practices wherever possible. AACR2’s treatment of MPMs is particularly problematic, since there are currently five different MPM models in use in Sweden. The necessity of proposing changes through MARBI makes change slow moving and formalized. However, with other MARC 21-using European countries as allies, there is hope that their concerns will be heard in the future.

The benefits of international record exchange, ease of use, existence of documentation (albeit in English), and support of the Library of Congress outweigh these disadvantages. According to Cato, since switching to MARC 21, Swedish catalogers who work on foreign materials have increased the amount of materials they catalog.

The Croatian Experience

Mirna Willer, Nacionalna i Sveuèilišna Knji�nica, Hrvatska (National and University Library, Croatia)

Croatian libraries have used UNIMARC for bibliographic data since 1980 and authorities since 1991. They adopted MARC 21 between 2005-2006 and began production in MARC 21 in October 2006. MARC 21 was adopted to meet international standards, despite a feeling that it was also important to maintain national traditions. Croatian libraries follow their own cataloging rules, developed by Eva Verona, as much as possible, unless MARC 21 commands differently. Croatian librarians noticed some loss of coded data, for example for records describing rare books, but new resources appear to be well accommodated. Although librarians were affected by the adoption of MARC 21, Willer doubts that users were even aware of the change.


The main reason for adopting MARC 21 was to foster global cooperation and to exchange data more readily. The previous national codes worked sufficiently well, but they were developed to satisfy local needs. Although these local practices were often a source of national pride, and there was some nostalgia at the thought of abandoning their old systems, the national systems also left the countries feeling somewhat isolated. After much discussion, each of the countries ultimately decided that adopting MARC 21 was the best solution. Despite some problems, most of the migrations went more smoothly than expected. As Cato concluded, MARC 21 may not be perfect, but it is the international standard and, with input from new users, it is getting better. And the speakers are gratified that MARBI seriously considers their suggestions and concerns. In the end, all of the speakers agreed that the adopting MARC 21 was the right decision.