Technology for Small Libraries

by Karen Coyle

Column: Managing Technology 

PREPRINT. Managing Technology column, Journal of Academic Librarianship v. 34, n. 4, July, 2008. pp. 369-370

Technology has allowed us to greatly increase the scale of library services. We can reach out to users anywhere in the world by placing our services on the Internet. We can bring a larger library to our users by forming consortia that expand the library’s holdings and by presenting users with a union catalog of consortium members. Not only is this larger scale good for our users, it has economic advantages for the library because the relative cost of growth, once the technology solution is established, is low.

The ultimate in large scale technology for libraries is the vast union catalog established by OCLC, WorldCat. With over 60,000 libraries and 100 million records, WorldCat is the largest library catalog in the world.[1] What we cannot measure, however, is what WorldCat does not contain: the many small libraries or less connected libraries that cannot afford to be part of WorldCat or that simply have no means to produce the data that WorldCat requires for participation.

The needs of small libraries was brought home to me in April of 2008 when I visited libraries in Kosovo as part of a U.S. State Department program. Although the National and University Library (Bibliotekës Kombëtare dhe Universitare të Kosovës, or BKUK)[2] is impressively well-appointed, local and regional libraries of one thousand to three thousand books are the norm. Most of these libraries have no catalog, neither in paper nor electronic. Clearly these libraries could benefit from the kind of infrastructure that would allow them to share information about their holdings, and the world could benefit from the information that these collections exist, especially because the libraries often have some unique archival materials.

You don’t need to go as far as Kosovo, however, to see small libraries that are not connected to the global library network that WorldCat represents. Many small libraries in the U.S. are unable to make the financial and technology investment that such participation requires. Some of these small libraries are in our own backyards: departmental collections that provide commonly used materials and ready reference just down the hall from the teaching staff. These libraries often have no catalog and do their lending on an honor basis.

As a person who focuses on technology, my immediate thought is that technology should be able to serve these small libraries as well; that although technology supports large solutions well, it should also be able to support small ones. With that in mind, I went looking for that technology. Unfortunately, my optimism was not borne out.

Catalog Software

In the United States and in some areas of the Western world, the world "library" is synonymous with "lending library." We have grown so accustomed to the fact that our libraries lend books and other materials that we forget that in much of the world materials never leave the library. Books are rare and precious. The concept of browsing the shelves is also very much a Western notion; the closed stack is common beyond our shores. Libraries with closed stacks absolutely must have at least an inventory of their holdings, as nothing will be discovered by users through any other means. Discovery also requires topical access, not just access to known items. So the minimum requirement for these libraries seems to be the ability to create a library catalog. For large and medium-sized libraries, the library catalog is one of the functions of the integrated library system. Smaller libraries can make use of stand-alone catalog software designed for libraries or analogous software designed for document management in businesses, schools or churches. This software needs some particular features, especially if it is to be used in situations like those in developing countries.

First, it needs to import and export standard records. This can be MARC, UNIMARC, or any other library standard format. It is only through these standards that small libraries will have the opportunity to share cataloging copy and form union catalogs with peer institutions.

Next, it needs full Unicode support so that cataloging can be done in the language of the library. Not only the catalog records but also the user interface must be customizable for the local language. (This may not be an issue for some small libraries in the Western world, except to the extent that they carry materials in non-Roman scripts.)

From the technical point of view, the system must require very little technical expertise and must run on common hardware and software. Librarians at these institutions will not have the knowledge to tackle technical problems and the software company’s technical support may not be available at all locations. This requirement to run on common software argues against Linux-based systems, even though Windows and Macintosh systems may require more expensive hardware than a library system that runs on Linux. It is the case that the most popular current Open Source library systems, Koha[3] and Evergreen[4], run exclusively on Linux.[5]. Without local Linux expertise, however, libraries will not be able to make use of those solutions.

The systems should have a cataloger view and a user view. It must be possible to run the user interface on a networked computer. The user interface could query against the same database that the cataloger accesses, or it could be based on a stand alone copy of the data. This would be similar to the user interface solutions of some of our "next generation" catalogs, such as Endeca[6], Aquabrowser,[7] and the locally-developed and Open Source program Scriblio.[8]

Now we need to talk about money. There are small cataloging systems available for only a few hundred dollars, although I haven’t found any in this price range that support library data standards such as MARC21.[9] Many of these appear to be designed for office or even home use and do not have a public catalog view. These very low-end systems generally do not have mechanisms to import copy cataloging from other catalogs. In a situation where copy cataloging is not expected to be available, it might be acceptable to do manual input of the library records. But the lack of support for record standards on output places a roadblock on the library’s future capabilities to become part of a local, regional or global network of libraries.

The systems that do support library standards are priced in the thousands of dollars, not the hundreds. They also provide many features that the smaller libraries will not need but that require the software and the support to be reasonably complex. Unfortunately, removing features does not reduce the price nor the complexity in most cases.

A Network Solution

For libraries that have reasonably stable Internet access, a remotely hosted system may be the best solution. The small library would not have to take on the responsibility of running the software, and costs could be shared. By being on the network the library’s user view would be available from any computer connected to the internet. The online service should support copy cataloging, original cataloging, and consortial sharing. It should adhere to known library standards.

You may realize at this point that the above paragraph bears a canny resemblance to OCLC’s cataloging services and the WorldCat interface. The solution I am imagining would, however, be scaled down in relation to OCLC. The records, although standard, would tend to have a smaller number of fields than full MARC21 cataloging. The cataloging interface would be online, not in a client, and would be Initially, it may not be necessary to provide services beyond the catalog. Interlibrary loan, record export, and much of the structure of OCLC could be deferred. It may turn out that the best solution would be to clone copies of the site for different groups or geographic areas, somewhat like the different language versions of Wikipedia.

Ongoing Work

There is at least one group working to bring technology solutions libraries throughout the world who otherwise would not be able to participate in the global library community: Electronic Information for Libraries (eIFL).[10] The group negotiates access to online resources for its members and has a program that promotes the use of Open Source Software. This type of effort, combined with some simple online and offline solutions for sharing, could go a long way toward opening up an entire new world of culture and information. Not only would the areas that are currently underserved benefit from a connection to libraries in the developed world, the latter would gain access to materials that have not been seen before.

We can use technology to make this happen. It is, however, going to take some new development.

[1] WorldCat facts and statistics. (Accessed April 24, 2008)
[5]There is an open source solution that runs on Windows, Kitlib4. The software and its documentation are in Croatian only, but hopefully translations will begin to appear.
[9] For a list of software designed for small libraries, see (Accessed April 25, 2008)

The copyright in this article is NOT held by the author. For copyright-related permissions, contact Elsevier Inc.