Open Access News

Realizing the dream of creating a rich, openly accessible digital library requires navigating copyright. A new study, Acquiring Copyright Permission to Digitize and Provide Open Access to Books [PS: blogged here 10/21/05], examines the practical aspects of seeking open access to monographs whose rights are privately held --that is, most work published after 1923. The author, Denise Troll Covey, principal librarian for special projects at Carnegie Mellon University (CMU), describes three efforts at CMU to make books freely available on the Internet for public use. Her descriptions of this process and its results reveal an array of challenges --from locating copyright holders to defining the meaning of out of print-- but also suggest strategies for success. The studies show that obtaining legal clearances requires an enormous investment of time, money, and patience....The first of the three studies, the Random Sample Feasibility Study, was conducted between 1999 and 2001. Its purpose was to determine how likely it was that publishers would grant nonexclusive permission to digitize and provide surface Web access to their copyrighted books and to gain insight into what factors influence publishers’ decisions....The study targeted 277 titles published by 209 publishers, selected at random from CMU’s library catalog. Project staff requested permission on a title-by-title basis, mailing a separate letter for each inquiry. A follow-up letter was sent if the publisher did not respond. Publishers who no longer held the copyright in question sometimes returned the letters; in such cases, project staff attempted to find the current owner. Ultimately, 21 percent of the publishers, accounting for 19 percent of the titles in the sample, could not be located. Half of the publishers responded to the request letters, and more than one-fourth of them granted permission, thereby enabling CMU to digitize and provide Web access to about 25 percent of the copyrighted books in the sample. Most of the publishers who granted permission applied some kind of restriction, ranging from limiting access to Carnegie Mellon users to stipulating that permission did not apply to components of the work with copyright owned by a third party. Troll Covey estimates that the average cost of obtaining a single permission was about $200....[By contrast] the average transaction cost per copyrighted title in the Posner [fine and rare books] collection for which permission was granted was $78....The Million Book Project (MBP) is funded by the National Science Foundation and the governments of India and China. Its goal is to digitize and provide open access to 1 million books by 2007. The MBP is part of the Universal Library Project, a partnership of Carnegie Mellon School of Computer Science and the CMU libraries....Staff located all the publishers that they attempted to contact in the MBP....Almost one-fourth of the publishers granted permission to include at least some of their titles in the MBP; altogether, permission was given for at least 52,900 titles. Slightly more than one-fourth of the publishers denied permission. Of the publishers that granted permission, one-fourth did so for all or most of their out-of-print titles. More than half granted permission for a specific subset of their titles. The average transaction cost for obtaining permission in the MBP was 69 cents per title....

The three studies show that however painstaking the effort, it is possible to secure permission to digitize and provide open access to books. The following findings from the CMU projects may be instructive to others who plan to seek copyright permission for digitization. [1] Locating copyright holders is difficult, expensive, and often unsuccessful....[2] For very large projects, the cost of obtaining permission on a title-by-title basis may be prohibitive....[6] Publishers who deny permission may fear lost revenue, may no longer hold rights, or may be uncertain of their rights. Publishers who chose not to grant permission to digitize often did so because they feared lost revenue, even for older or out-of-print titles that were not generating revenue. Many publishers, particularly university presses, said that they wanted to participate but could not because copyright reverts to the author when their books go out of print. Many publishers noted that older contracts did not grant them electronic rights to the books or that they were uncertain of their rights in this regard. The most common reason publishers gave for not participating in the MBP was they did not have the time and staff needed to check their paper files title-by-title to determine copyright status and ownership.

(PS: The Open Content Alliance will face these or similar hurdles whenever it wishes to scan copyrighted books. It may not face exactly the same hurdles because it will offer content providers the option to restrict access to the digitized result.)