F e d e r a l    D e p o s i t o r y    L i b r a r y    P r o g r a m

Council's Comments on the
Collection of Last Resort Draft Dated April 6, 2004

Council applauds GPO for developing the Collection of Last Resort draft plan. At this stage of the plan Council believes that the draft provides a clear vision for the future as it makes strong commitments to:

• Collect and preserve all formats, including born-digital objects;

• Reinforce the long standing principle of no-fee public access;

• Create cataloging and metadata records for access and preservation;

• Safeguard government information in times of national emergency.

The draft begins to articulate a new vision for long-term access to federal government information. For such a vision to become a reality, however, additional details must be provided to the library community and other stakeholders. The following recommendations suggest specific areas where additional details are needed.

1. In Spring 2003, Council recommended the creation of the "United States Library of Public Information." The recommendation is as follows:

United States Library of Public Information

Council recommends that as the Government Printing Office pursues its initiative in creating a United States Library of Public Information, the Government Printing Office, along with members of the depository library community, develop a cogent, flexible collection that is:

1. Comprehensive in scope and content

2. Fully cataloged

3. Widely accessible

4. Permanently archived

Rationale: In today’s increasingly electronic environment, the need for a United States Library of Public Information providing permanent public access, full cataloging records, widely accessible and comprehensive in scope, becomes more of a national need. GPO’s pursuit of this library will address the current and future needs of the new depository environment.

Council conceptualized the United States Library of Public Information as a widely accessible collection including appropriate service to that collection. As articulated in the preliminary CLR plan, there does not appear to be a clear statement regarding how access and service will be provided. Council strongly recommends that a plan for the shared or "light" repositories be developed and vetted with all stakeholders as soon as possible so that the larger context within which the CLR functions will be clear.

2. Council would like to see a more complete articulation of what GPO envisions regarding the roles of both FDLs and GPO in subsequent versions of the plan.

3. Assumptions #11-13 provide estimates of the time and money needed to digitize the legacy FDLP collection. Although not explicit, these assumptions appear to allude to the agreement between GPO and the Association of Research Libraries to collaborate on this effort. Council recommends that GPO also encourage, support and include the efforts of smaller depository libraries and those who are not members of ARL to be partners in this worthwhile endeavor.

Council would again like to emphasize our belief that born-digital government information should be of the highest priority in the GPO strategic planning process. See Spring 2004 recommendation on Born Digital Information is at Risk.

4. The CLR draft states that "Tangible copies of 'born digital' products will be produced for the dark archive." Council recommends that GPO devote appropriate resources to ensure the goal of creating and storing print copies of all electronic items, including those on floppy disks, CD-ROMs and DVDs. A tangible collection of current and retrospective U.S. government information would better ensure the preservation of the entire collection, particularly during and after a national emergency. Council suggests that GPO work with a variety of partners to print the digital collection, just as it is working with partners to digitize the print collection. For example GPO should begin to print 1 or 2 copies of information currently being added to the FDLP Electronic Collection. Printing a set of the pre-existing born-digital collection could be achieved through the print-on-demand service envisioned by GPO. Such printing will require the establishment of standards. Council also wishes to emphasize the importance of establishing "light archives" as locations for the additional copies of printed items. "Light archives" could include equipment for onsite printing and might be able to serve as distribution centers for libraries requesting hard copy from digital images.

5. Table 2 indicates that GPO will acquire fugitives to add to the CLR. Council recommends that GPO outline specific ways in which fugitives will be identified, added and preserved. Council further recommends that GPO partner with depository libraries and professional organizations in acquiring fugitive materials.

6. Table 2 also indicates that GPO will acquire tangible information products from depository

7. The plan discusses access and preservation in two separate sections. Council recommends that the plan articulate the long-standing view among preservation librarians that preservation and access are inter-related. Preserving information, regardless of form or format, helps ensure accessibility in the future, even though, as in the case of the dark archive, the physical items themselves are relatively inaccessible.

8. Assumption #14 states: "After digitization copies of the original publication, even if disbound, will be retained and preserved in case the item must be digitized again in the future." Council recommends that GPO not disbind its only copy of a publication.

9. Council recommends that a timeline for the development of the CLR, including a target date indicating when its existence begins, be included in subsequent versions of the plan. The timeline should include the preservation plan when the CLR comes into existence and should also include phases in the progress toward full digitization and cataloging/metadata creation.

10. Council recommends that a detailed cost analysis and/or budget should be included in subsequent versions of the plan.

11. Council recommends that GPO partner with OCLC and RLIN, perhaps through their respective members who are also in the FDLP, to acquire, use, and adapt pre-1976 cataloging for legacy collection items.

12. Council assumes that a "Dark Archive" containing master copies of the digital CLR will reside on digital storage devices that are not networked.

13. The CLR plan should answer the following questions:

• To what does Table 3's mention of GPO processing "CLR boxes" refer?

• What is ONIX and how will its use be of benefit?

• What does the statement "GPO is responsible for providing expertise in interpretation, access, and service for the publicly accessible portions of the CLR" mean in practice?

• Does the GPO plan to try to collect two copies of every legacy publication, or only one?

• Will the CRL be located in one facility or geographically distributed?

• What are GPO’s contingency plans for access in to the CLR in times of a national emergency?

• How will GPO authenticate digital masters?

• How can GPO guarantee that the digital masters are official and have not been tampered with?

• Will source files be made available to FDLs?

• Does GPO envision using LOCKSS (or similar technology) to safeguard the integrity of geographically distributed digital objects that are housed by partner institutions?