Berkeley Digital Library SunSITE

Redesigning Library Services: A Manifesto

6: Collections Reconsidered

The Problem
The Roles of Collections
Two Classes of Material
Electronic and Paper Collections Compared
From Local to National Collections


The shift from the Paper Library to the Automated Library, as discussed in Chapter 4, has raised interesting basic questions about the design and role of the library catalog. It is not simply that the appearance and form of presentation of records in printed book-form catalogs and on three-by-five-inch cards needs to be changed for on-line display. A much more fundamental reconstruction of the definition role of the catalog and, especially, of the relationship between bibliographies and catalogs will eventually follow from the radical change in the underlying technology. The widespread practice during the 1970s and 1980s of using sophisticated computer systems and telecommunications networks to generate three-by-five-inch cards to be filed by hand in card catalogs made sense only as a transitional stage. Designs and approaches that are optimal with one technology are unlikely to remain optimal with a quite different technology.

Do similarly interesting basic questions arise concerning library collections when the documents are electronic rather than on paper? We have already stressed the importance of local collections for the Paper Library. Further, developing a library collection is an attractive activity. How else do you get to spend lots of money buying and keeping interesting things ... using other people's money?

Collection development is also unusually interesting because so much money is involved. It is not only a matter of the funds used to pay for materials, but also the substantial proportion of a library's labor that is devoted to selecting, purchasing, cataloging, and processing the materials, plus associated administrative overhead and space needs for these people and for housing the collection. In the nine campuses of the University of California, for example, two-thirds of the libraries' operating budget and two-thirds of the libraries space needs can be attributed to the assembling of collections and preparing them for use. Libraries are chronically short of operating budget and of space, so any activity that accounts for two-thirds of both operating budget and space usage ought to be of great interest. When trying to solve a murder mystery, the traditional advice was "Cherchez la femme," look for romantic interest. When studying organizations it is useful to look at where the money flows.

Our perspective on library service is based on experience with libraries in the form of large collections of books. Perhaps we should be better off if we retreat a bit toward first principles. A good starting point is to ask why libraries develop collections. What role, or roles, do collections play?


Why do libraries spend so much of their operating budget and space assembling collections. Collecting material does not create material. It only affects where copies are located. Library collection development is a matter of "file organization," concerned with where copies of documents are to be located and for how long. Collecting cannot be justified for its own sake but only as a means for the role and mission of the library. The role of library service is to facilitate access to documents and, by extension, to provide service based on the availability of documents. The mission of a library service is to support the mission of the institution or population served. These are, however, general statements that need more detailed interpretation in each case. To justify the investment we ought to have a coherent, explicit, and convincing explanation of why we devote so much effort, money, and space to the assembling and refining of collections of documents and of how the "value-added" benefit of collecting compares with the contributions of other claims on library resources, such as bibliographies and assistance to readers. These basic questions have been rather neglected and, perhaps, taken for granted.

A start can be made by isolating the specific purposes that collections, hence collecting, support:

  1. Preservation role: Any document that is not collected and preserved is likely to be lost, unavailable both now and in the future. It is difficult to predict what might be of interest to someone in the future. When in doubt it is prudent to preserve non-renewable resources.

  2. Dispensing role: The principal reason for most investment in collection development is not preservation but the need to provide convenient access to materials that people want to see where they want to see them. If someone asks to see a book, it is not entirely satisfactory to answer that a copy exists and is being carefully preserved in some foreign national library. The need is for a copy here and now.

    The difference between the dispensing role and the preserving role can be imagined by considering the difference between the size of library collections as they are now (or, better, as large as people would like them to be) and the size of libraries as they would be if, nation-wide, only two or three designated copies of each edition were retained for preservation purposes and all other copies were vaporized. Imagine how dramatically libraries' space problems could be solved--and how detrimental it would be for library service--if only two or three preservation copies were retained. The difference is an indication of the importance and high cost of the dispensing role in relation to library materials on paper.

  3. Bibliographic role: The use of materials depends on identifying and locating what exists. Bibliographies and the bibliographic superstructure that is built up in our catalogs are created for this bibliographic role. Remember, however, that the bibliographic records that we create, arrange, and search in so many ways are surrogates for the (more unwieldy) original documents. Yet bibliographic records are only small, incomplete representations of the real thing. There is no reason why the original documents should not also represent themselves, sitting on the shelves for all to see the choice that exists. Browsing shelves of documents arranged by subject (or any other) order may not be an entirely satisfactory guide to what exists, but it is one way and, unlike the examination of catalog records, has an enormous advantage: the document itself is at hand.

  4. Symbolic role: The three roles already mentioned--preservation, dispensing, and bibliographic--do not seem sufficient, even taken together, to explain collection development behavior adequately. Collections also have a symbolic role. Large collections, particularly of special materials, bring status and prestige whether the materials are used or not. The symbolic value of collections and of buildings to house them are, perhaps, more marked in the case of museums, but should not be ignored in the case of libraries.

If these are the four purposes of collections of library materials, we need to inquire how the change from paper to electronic media may change how we seek to effect these roles.


What would an extraterrestrial have to say about the way library services are provided on Earth? To have reached Earth and to be able to send messages back, the extraterrestrial must be familiar with sophisticated technology and telecommunications, but might, however, have been unfamiliar with, and very intrigued by paper as a form of information technology.

Paper is energy-efficient and is fairly robust, but does have two attributes that limit and dominate the way it is used. First, paper is generally a solo technology: Like a telescope, it is usually best used by only one person at a time. It is frustrating when two or more people try to use the same reference book simultaneously. Second, paper is a localized medium: Paper can be read only if the reader and the paper are in the same place at the same time. True, one can travel to a library at any distance or interlibrary loan can bring a copy of a document. Either action involves inconvenience and delay in order to achieve a situation in which you and the paper are in the same place.

Electronic documents, in sharp contrast, can be used by many people at the same time. Who knows or would care how many other people are using an on-line library catalog at the same time as you are? Further, unlike documents on paper, users don't even have to be where the database is. Users do have to have a telecommunications connection to the database, but they don't even need to know the physical location of the database.

This discussion is in terms of paper documents and of electronic documents. Other media such as microforms and clay tablets can be seen as inconvenient variations on paper. In the following discussion we will distinguish between materials that are localized (notably paper) and materials that are not localized (notably electronic documents).


What does the distinction between localized media (e.g. paper) and non-localized media (i.e. electronic) signify for the development of library collections when each role is considered?

  1. Preservation role: For the preservation of nonrenewable resources, it remains prudent to retain two or more copies designated as archival copies and carefully stored at different locations under suitable conditions. The specific techniques for preservation vary with the differing physical media (paper, magnetic tape, microfilm, etc.), but a broadly comparable pattern emerges for the preservation of paper and of electronic documents.

  2. Dispensing role: The substantial difference in transportability indicates a major change in the dispensing role: a much reduced premium on local storage compared with storage at a distance. In contrast to the localized media of paper and microform, local storage is no longer a necessary condition for convenient access with electronic collections. As a preview of this situation we can consider on-line catalogs. Filing catalog cards locally is no longer necessary for the catalog to be consulted locally. The reader need not know or care where the disk-drives are physically located so long as records appear on the screen. (Local storage may sometimes be desirable on grounds of economy or reliability, but that is a technical matter, varying as telecommunications, storage, and other costs change and is a concern for the service provider rather than the service user.) There is, therefore, a fundamental change. To the extent to which materials capable of remote access are used, the historic necessity for local collections ceases to apply. Material will not need to be stored locally, will not be out on loan (because it is copied rather than lent), and will be available wherever the user's workstation is located. Local storage becomes optional rather than necessary. Decisions about what to store locally will depend on several changing, technical and economic factors.

    Since local collections can account for two-thirds of operating and space costs, there would appear, in theory, to be substantial potential scope for investment in remote access as an alternative.

  3. Bibliographic role: It is not yet clear how access to machine-readable text will be provided. However, there is no obvious reason why bibliographic records would not also be stored and linked with the text. If this is the case, then the combination of on-line bibliographic data and on-line electronic documents would appear to have all of the advantages of on-line bibliographies and catalogs combined the advantages of having immediate access to all of the texts for searching, browsing, serendipity, scanning, and reading. This best of two worlds simply cannot be achieved with localized media on paper and microform.

  4. Symbolic role: The symbolic and status-bringing role of large and impressive local collections of paper documents cannot be denied. Status and useful function tended to coincide. The prestige of having extensive access to electronic documents is less clear, especially as access to electronic documents will probably be a lot more easily and a lot more equitably achievable than has been the case with paper documents, which have always very unevenly distributed among groups and geographically. Perhaps the wisest course is to emphasize what is functional and hope that prestige will be achieved as a by-product or in other ways. The alternative approach--major investment is what is prestigious but no longer the most functional approach--becomes questionable.

Of these four roles, it is the dispensing role that is stands out as being different when electronic documents are compared with paper documents. The difference is two-fold: It is the dispensing role that accounts for great preponderance of libraries' operating costs and space needs in the Paper Library and in the Automated Library; and it is in the dispensing role that electronic documents promise to be particularly advantageous because local storage, which so dominates the operating and space expenses with paper documents, ceases to be necessary.

So, in principle, it appears that it is the dispensing role in which there is the most to be gained and it is also the dispensing role in which change appears most feasible. How and how far such a change would be acceptable for the purposes of library users deserves very careful attention because the stakes are so high.

It becomes important to remember that library users fall into two quite different cases with respect to the dispensing role. The obvious case is when the choice is between use of local collections on paper or access to electronic documents. In practice, this is the limited case of the privileged few. For most people, for most documents, for most of the world the effective choice is likely to be between remote access to collections of electronic documents or the costs and delays of obtaining paper documents on interlibrary loan because so much material is not in conveniently local collections. This is the case even with those few countries that currently have the best library collections. Taking a global view it seems difficult to imagine that the vast preponderance of cities that do not now have splendid public library collections and the vast majority of universities that do not now have excellent collections have much prospect of ever being able to achieve excellence by assembling sufficient local collections of paper documents. For rural areas, smaller institutions, and poorer countries the prospect is hopeless.

It would seem very foolish to expect any scenario involving an "either/or" dichotomy--only paper collections or only a paperless library. Some balance between selected materials on paper, presumably the more heavily-used and less volatile material, and selective recourse to electronic documents for much of the rest as available. On such a view one might expect core collections of materials in relatively high use to be collected and held locally ("dispensed") on paper, even if also available as electronic documents. Core collections, however, are not where the heavy costs arise. The demand for library materials is very unevenly distributed over collections and so most of the cost of acquisition, processing, and especially storage is attributable to the larger quantities of relatively little-used material.


Just as the change from the Paper Library to the Automated Library, in conjunction with the rise of on-line bibliographies, changes our perspective on the catalog, so also the rise of the Electronic Library changes our perspective on collecting and local collections. Instead of our thinking being dominated by local collections, as is unavoidable with the Paper Library and the Automated Library, the effect of having electronic documents is to make local storage optional rather than necessary. This means that we can realistically begin to think nationally or, in more technical terms, network-wide--as broadly as there is networked access. Since electronic documents do not need to be locally held and since the needs of users in different locations vary, the most plausible approach is that of viewing the totality of electronic documents as being, in effect, one great big distributed collection, analogous to the great big collectivity of bibliographies and holdings records discussed in Chapter 4.

There is much that has to be studied in these issues: Relatively little attention seems to have been given to comparing the costs of maintaining paper collections compared with the costs of maintaining electronic collections. If hardware costs and computing costs continue to come down, then presumably the costs of storing electronic documents must also be trending downward relative to the costs of storing paper documents. Similarly studies will need to be made of the costs and difficulties of accessing electronic documents remotely compared with accessing both locally-held and remote paper documents. Here again the same assumptions about cost trends would suggest a long-term cost trend favorable to electronic documents. For both types of documents the underlying problem is an "owning versus borrowing" trade-off, but the costs, costs trends, and acceptability appear to be different and to be changing.

Here again the issues are broader than technology, costs, and the preferences of library users. Publishers, booksellers, authors, and, probably, intellectual property rights are also moving into a changed situation as electronic documents permit, in principle, a significant change in library collecting practices and in the use of library materials. A changed "industry" and new policies are to be expected.

 Go to Supplement

Notes on Chapter 6: Collections Reconsidered

    1. The estimate on operating budget derived from campus reports of actual expenditures from State funds for "Library Materials" and "Acquisitions and Processing." The estimate for space is based on university-wide projections of space needs rather than actual space use. Both are probably underestimates. The assistance of Dennis E. Smith and Sue Plezia is gratefully acknowledged.

    2. This section is based on a more detailed discussion: Michael K. Buckland, "The Roles of Collections and the Scope of Collection Development," Journal of Documentation 45, no.3 (September 1989):213-226.

    3. M. K. Buckland, Book Availability and the Library User. (New York: Pergamon, 1975).

Copyright © 1997 Michael K. Buckland.
This text may be used for non-commercial personal and non-commercial education purposes provided that copyright ownership by Michael K. Buckland is acknowledged. All other rights reserved.
Document maintained at by the SunSITE Manager.
Last update July 2, 1998. SunSITE Manager: