Statement on Use of Digital Archives
November 14, 2005
In November of 2002 Google and the University of Michigan began discussing the possibility of digitizing the entire University library collection of approximately 7 million bound copies. The motivation behind the talks was a desire on the part of the University to preserve these works for future generations and to make the Library’s resources widely available for education and research. The University of Michigan had been digitizing works for years, but at a substantially slower pace than what was anticipated with the Google project. An agreement was reached on April 19, 2004, under which Google would digitize certain collections in the Library and in return provide a digital copy to the University. This statement of principles addresses the Library’s use of this digital copy.
What the Library Intends to Do
Preserve Its Copy in a Digital Archive
First, and most importantly, the Library will store and preserve its copy in digital form for future generations. Although our immediate plans are to keep the in-copyright materials “dark,” or inaccessible to any use, we know that as a result of book acquisition patterns among research libraries and the quality of paper used in printing through much of the 20 th century, these digital copies may be the only version of many copies that survive into the future. It is important to keep in mind that only libraries are charged with preserving the published record for future generations without regard to business considerations.
Define Any Use by the Nature of the Work
How the digital files will be used depends upon the characteristics of each work. Public domain works will be widely accessible. In-copyright works will be accessible only to the extent that there is a lawful use. Although there are few lawful uses of the in-copyright works, archiving them ensures that they will be accessible when they are no longer in copyright or when such uses are clearly established.
Secure the Archive
The University Library has extensive experience with securing, storing, and providing access to significant bodies of published material with attendant intellectual property issues. In all cases, the Library has employed a combination of stringent system-level security and rights management control to ensure that access to the files takes place only in sanctioned, appropriate ways. The same measures are being applied to the files created as part of this project.
Examples of Potential Use of the Archive:
Natural disaster recovery
Imagine what we could have done in the wake of Hurricane Katrina. For those libraries affected by the hurricane, we could have provided digital access to the public domain works. We could have worked with copyright owners to make available printed hard copies of out-of-print (but in copyright) works held by libraries such as Tulane but damaged by the storm. After all, publishers typically do not have copies of out-of-print works and they cannot be replaced any other way.
The archive may also expedite our ability to provide authorized students who have disabilities with access to works they have purchased for their studies. If we possess a digital copy of a work that a student has purchased for class, a visually impaired student who uses a screen reader for instance, will be able to access that work the day she purchased the book rather than days or weeks later if we must scan the work for her.
Computer science research
Other legitimate uses of the in-copyright works can be imagined. Consider the example of computer scientists working on automated ways to translate text. With multiple translations of the same work in the archive, they could run their programs against those works to test their algorithms. In this case, no user reads works from the archive, though the archive is used to further science and research. Linguistic and lexicographic studies would benefit from the ability to perform full-text searches over large bodies of relevant material. These sorts of analyses can locate relevant works, eliminate dead-ends, and shed light on problems that could only be solved by a lifetime of careful reading. For in-copyright materials, these searches would lead researchers to print versions of the works in question. The work of projects such as the Middle English Dictionary at the University of Michigan would benefit significantly from searchable access to such large bodies of text.
While we have not yet determined uses to which we will put in-copyright materials in the archive, it should be noted that our determinations will be appropriately conservative and guided by the law; moreover, without the existence of the archived copies, no possibilities for legitimate uses exist.
What the Library Will Not Do
The existence of the archive will not reduce acquisitions.
Current acquisitions are a lynchpin of our fundamental work to support scholarship at our institutions. In fact, none of the libraries approached by Google would have been considered had those institutions been anything but voracious in their acquisition of library materials. Because digitized in-copyright materials are not a surrogate for reading and use, they are not a substitute for the physical acquisition of books. The University of Michigan spent $16 million on new acquisitions and electronic licenses last year, and despite declining budgets we have budgeted $16.4 million for new acquisitions and electronic licenses this year.
The existence of the archive will not eliminate the need to replace worn/damaged works or the need to purchase additional copies of works.
The existence of the archive will not affect the Library’s practices concerning book replacement, nor will it affect our practices related to purchasing multiple copies of works. We will continue to replace books and purchase multiple copies as traditional library practices dictate.
The existence of the archive will not be a way of providing classroom access to in-print works.
The University will not give access to digital copies in lieu of purchasing text books. In our opinion, no such use would be sanctioned by the law, and doing so undermines the very economy that makes scholarship possible at our institutions. It is important to realize that universities and their libraries depend upon a healthy publishing environment. Scholarship requires publication and we have a vested interest in supporting a healthy publishing environment and a system that supports authors.
Some have speculated that with a digital copy of the Library, students, faculty and staff will no longer need to visit the Library because the Library will make the entire archive accessible to the University community. Providing such access to materials in copyright—particularly those that are in-print—would be unlawful. Merely because the Library possesses a digital copy of a work does not mean it is entitled to, nor will it, ignore the law and distribute it to people who would ordinarily have access to the hard copy.
The Google U-M partnership was undertaken with strictest attention to the rights of all stakeholders. As Universities and libraries fully engage in the digital world, this project is the right way to preserve the world’s collective knowledge and ensure its discoverability today and in the future.