By David Dorman
American Libraries Columnist
Library-technology consultant based in Middletown, Connecticut.
Column for December 2003
A recent onslaught of articles coming out of California newspapers reported on the use of Radio Frequency Identification (RFID) tags in public libraries and how they will strip away the reading privacy of library patrons. The concern has even been picked up and professed by some librarians who should know better. The devil, as always, is in the details. And the critical detail with respect to library use of RFID tags is that only two types of information are contained in these tags. One is a single bit (on or off) indicating whether the book has properly left the library. Pretty harmless, that.
The other is a set of fields designed to hold information about the book itself. Library RFID tags do not contain any patron information whatsoever. And most RFID implementations contain only one field of information about the book—the bar code number.
Forgive me for being skeptical of all the hullabaloo, but I just can’t get too worked up about a library book broadcasting its bar code for several feet to any and all who care to lug around portable RFID readers and eavesdrop on the reading habits of passersby carrying library books. However, if that thought bothers some folks, I recommend they campaign for encrypting the barcode number in the RFID tag rather than foment (or buy into) vague apprehensions about a very useful and rather benign library inventory control tool. For a well-written, thorough, and balanced treatment of RFID and Patron Privacy, check out the paper "Personal Privacy and the Use of RFID Technology in Libraries" by Vinod Chachra and Daniel McPherson.
“There are not good digital library systems out there. Everyone worries so much about reinventing the wheel, but if you don’t have a good wheel to begin with you aren’t reinventing anything.” This sentiment, expressed by Thornton Staples, director of digital library research and development at the University of Virginia, is the reason why four years ago Virginia teamed up with Cornell to develop a scalable digital-object-repository management system. With seed money from the Mellon Foundation, they began work on FEDORA (Flexible Extensible Digital Object and Repository Architecture). Today, the FEDORA repository system is still the only digital library management software in its class.
But Staples does not think Virginia and Cornell should be in the library software business: “We would really like to buy a system from someone.” The fact that no established library vendor has yet been willing to commit resources to, and market, this open source software project is an indication that the library software vendor community does not yet realize the potential of open source software to increase market share and revenues.
Old ideas and outdated verities die hard, but I am certain it is only a matter of time before a farsighted library vendor takes the risk to invest in a well-thought-out and well-organized open source initiative such as FEDORA. That vendor will, I believe, achieve a market success that will demonstrate the value a well-executed open source initiative can have on the bottom line. For information on the FEDORA project, see www.fedora.info.
The long-awaited effort to develop a standard scheme to represent and reference standard identifiers on the Web has received a boost in the form of a draft submission to the Internet Engineering Task Force. The proposed standard would allow numbering schemes such as CODEN and PMID, as well as vocabularies such as MeSH and LCSH, to be integrated into the Web suite of addressing standards.
Dubbed the “info URI scheme” (URI stands for Universal Resource Identifier), the proposal has two basic elements: a standard way of expressing “information assets” (for example, records from a database, terms from vocabulary schemes) and the creation of an official Web registry that describes the scheme. The registry would enable any organization to easily register its information asset scheme as well as enable any searcher to learn about the characteristics of the scheme by going to the registry. These “info” URIs are expected to be especially useful in conjunction with the OpenURL Framework, the development of which was a catalyst for the creation of this proposed standard.
Using the example of LC call numbers, here is how the standard would work. The Library of Congress would be required to register the LCCN numbering scheme at the officially designated info URI Registry website. (Translated into metadata jargon, LC would register LCCN in the information registry of public namespaces.) Maintaining a web-accessible database of actual values would be optional. LCCNs in bibliographic records would then be marked up in a standard way. For example, the LCCN 51002474 would be expressed as: info:lccn/51002474. This would be an official URI that would identify 51002474 as an information asset belonging to the LCCN scheme. Any searcher coming across the LCCN would then be able to readily go to the info URI registry to find out what LCCNs are, what organization creates and maintains them, and any other LCCN-related information required by the standard or volunteered by LC.
What is the value of this new proposal? Unlike current Universal Resource Number requirements, it has the virtue of easy compliance for creators of numbering schemes and vocabularies. If the proposal gets adopted and its use becomes widespread, it will be one more step in the creation of the semantic Web. Thucydides would be pleased. The Internet Draft can be viewed at: www.ietf.org/internet-drafts/draft-vandesompel-info-uri-00.txt.
Information Transform (a.k.a. MITINET/MARC Software), which markets the cataloging and catalog-maintenance software program MARC Magician, has become a pre-eminent MARC record clearinghouse for K–12 libraries. The company is using its 5–6 million MARC record database to create low-cost cataloging and bibliographic cleanup services for small not-for-profit libraries that find OCLC unaffordable.
For $79, existing MARC Magician customers can obtain two additional services that take advantage of the company’s extensive bibliographic database: AccessMARC enables a library to use the database for both adding to—and upgrading records in—its local system; the Magic Recon program uses the database to automatically replace a library’s brief non-MARC records with fully cataloged AACR/MARC records.
MARC Magician, including these new functions as well as card- and label-printing software, is being marketed under the moniker of Dream Suite. The package is sold to new customers for $150 to $200 per station, depending on the number of stations purchased. For additional information, see www.mitinet.com.
Feedback on reporting Library Management System sales monthly or on a quarterly basis was unanimous. AL readers say that the quarterly review gave a more useful perspective. Therefore, this month’s column and the January column will not record LMS sales. A fourth quarter report for 2003 is scheduled for February.
Docutek—with the Michigan Library Consortium, for Docutek’s VRLplus virtual reference system, which will be available to all of MLC’s 650+ member libraries
of all types.
Swets Blackwell will change its name to Swets Information Services as of December 2. The Blackwell name will no longer be used for subscription services, but only for publishing and bookselling.