Retention policies: the joys of de-accessioning

Mike Bruer said that his question about de-accessioning has several sources, including Nicholson Baker's article in the New Yorker on withdrawals from the San Francisco Public Library and a letter to the President of NYPL from a book-dealer about material getting out of that library and into the hands of others such as himself (material that had been withdrawn after being microformed). The question has grown into a larger one: Is there a difference between accessioned and owned? A member of Congress (who shall remain nameless) is understood to be preparing legislation that would make it almost impossible for research libraries to get rid of anything. Under what circumstances is it legitimate to dispose of material? What means of disposal are legitimate? He emphasized that this is something we may all need to deal with.

In response to a question from Brian Schottlaender (UCLA) Mike said that NYPL routinely discards originals after micro reproduction unless the original has intrinsic artifactual value. In response to another question from Brian Schottlaender he said the potential legislation would apply to gifts as well as purchases.

Brian Schottlaender asked what grounds the potential legislation was predicated on? What was the logic behind it? That a research library is supposed to keep everything?

John Lubans (Duke): Does the public think we should keep everything? In his opinion most people don't think much about that question but a few individuals have a reactionary view of the matter. Brian Schottlaender said the cultural heritage argument says that libraries in the AGGREGATE are responsible for preserving the aggregate heritage, not that each library has to preserve everything.

Judith Nadler (Chicago) suggested that collection development policies should state the library's policy for removing material. Mike responded that NYPL has a new 25 or so page policy that is much more conservative than its predecessor. It will need to be accompanied by a statement of Policy and Principle, and then by specific procedures.

Sharon Clark (Illinois) asked if there is New York state law that governs such matters as there is in Illinois. The answer was Yes; institutions in New York State must first offer material to all other state institutions that might want it.

Bob Wolven (Columbia) suggested that the feeling behind the proposed legislation probably is that libraries are wasting public money when they discard material that might have been sold. Mike said NYPL is doing a survey of a sample of material discarded in the past and having Christie's or Sotheby's value it. Early indications are that most of the stuff that has been discarded, as we would have assumed, is of a routine and essentially valueless nature.

Michael Kaplan (Indiana) and Brian Schottlaender (UCLA) were the presenters of this topic. Brian said the major issue in his view has to do with input streams. He spends inordinate amounts of time coordinating 'acquisition' of such material, reviewing licenses, talking to the right people about IP addresses, getting resources on webspace, publicity, etc. Many decisions are now being made by those external to university; more and more people are getting involved in the process and there is need to keep them all informed.

Michael Kaplan asked who in your organization is responsible for electronic publications? One or several people? Do you maintain a legal review? By whom, with what background? What is meant by 'cataloging' for electronic resources? Indiana University has a project to acquire about 1100 ejournals. What if you later decide not to keep all of them? The Peak Project which was predicted to provide input about user behavior on full-text journals has not provided the level of information that had been hoped, at least to date. In general, he feels the need to put one person in charge.

Sally Sinn (NAL) said this question is taking more of her time than anticipated. NAL uses a team of public and technical services staff. They are taking a detached view towards cataloging. Users are having no problem finding ejournal text without catalog records. NAL's primary interest is in connecting Agricola database (which indexes the material) to the electronic texts.

Beacher Wiggins (LC) said that LC is developing a model. LC had internal staff work with a contractor to see what all the pieces of the question are. The final report from the contractor and group, submitted in late 1997, included a list of recommendations. One major recommendation was to create a position for an 'electronic czar' to have oversight over all aspects of dealing with electronic materials. Questions to be addressed include: Who will report to the czar? What will be the copyright implications? Budget implications? Collection development guidelines? Cataloging issues? LC is currently drafting a budget initiative for FY1999 to encompass these identified needs.

Judith Nadler (Chicago). At Chicago they think e-publications activity falls into four aspects or categories:

1. Selection and licensing is done by Collection Development; the Head of Reference is also involved.
2. Organization and access, including metadata is the responsibility of technical services.
3. Interpretation and distribution falls to the Head of Reference.
4. Local digitization projects are suggested by individuals and departments.

They need good coordination among these.

Carol Diedrichs (OSU) said that OSU has been heavily involved in the licensing aspects, with much of the work delegated to the Head of Serials who is working with University Counsel. The Head of Serials has educated the University Counsel about libraries. The University Counsel has indicated that the Head of Serials now has the skills of a paralegal in this area. University policy is that if you sign a contract you are responsible for your own legal fees.

Jeffrey Horrell (Harvard) described Harvard's digital initiative infrastructure. There is a Digital Acquisitions Coordinator who will be responsible for licensing and renewals and the position was just filled by Ivy Anderson. Robin Wendler is serving in the position responsible for metadata and Steve Chapman holds the reformatting position. A position devoted to digital archiving will be developed and recruited for in the not to distant future.

Lee Leighton said that Berkeley set up an organization structure similar to Chicago's but after a reorganization it didn't work so well. People have taken it upon themselves to learn as much as they can.

Beth Warner said that Michigan has had a Digital Library for several years. They have moved from having everyone involved to having things funnelled through a few teams to more decentralization through creation of teams with technical services representation. The team decides what level of cataloging is needed for various categories of electronic materials. Michigan decided to put everything in catalog as well as to create a special database of e-materials for people who want only e-materials and who need to be informed of local web pages.

Joan Swanekamp said that Yale has had a fair amount of experience over the last several years. As to notification: Ann Okerson and her staff inform library staff of all new electronic resources through an electronic discussion list; they are also listed on a webpage. Yale has been entering their records for web cataloging for several years.

Sally Sinn (NAL) said that one problem is dealing with purchasing officers.

Larry Alford that at the University of North Carolina at Chapel Hill (UNC) they make vendors compete with each other. This is done more by discussion and negotiation rather than through formal purchasing agreements.

Duane Arenales (NLM) said that they have been concentrating on the input stream. Perhaps for the next Big Heads meeting (Philadelphia) we could combine this with Mike Bruer's retention issue to discuss questions relating to access and preservation. Perhaps we need to revive the old idea of a National Periodicals Center and to create a National Electronics Center.

John Lubans , in noting how rapidly things change in the electronics area, offered this definition: A geek year is 90 days.