English | français
The International Federation of Library Associations and Institutions (IFLA) is committed to the principles of freedom of access to information and the belief that universal and equitable access to information is vital for the social, educational, cultural, democratic, and economic well-being of people, communities, and organizations.
IFLA therefore welcomes Google’s contribution to achieving these goals. The Google Book Search programme has the potential to provide public access to a digital library of millions of books. It can be, when fully and properly developed, an unprecedented source for the advancement of learning and human development.
Google and representatives of a few libraries, publishers and authors have come to an agreement on how to settle the copyright and other legal issues in relation to the Book Search Project. This settlement is now under review by the United States District Court, Southern District of New York. IFLA hopes that the proposed settlement will serve as the beginning of a fruitful cooperation that will benefit the millions of users whom libraries serve throughout the world, bringing us closer to the achievement of our goal of providing equitable access to information.
However, IFLA believes that the following issues must be satisfactorily addressed before IFLA can support this, or any similar subsequent agreements among libraries, rightsholders and corporate partners:
Territoriality – The Digital Divide
IFLA, the International Federation of Library Associations and Institutions, is deeply worried about the territorial limits of the settlement.
The copyright laws of a country apply only within that country. Therefore, the settlement, which is based on US litigation, would apply only in the United States of America. As a consequence, the expanded services permitted under the settlementi would be provided only to users located in the United States. Users outside of the USA would have access only to the current Book Search service, which for books in copyright permits the display of only three “snippets” consisting of a few sentences of text.
The expanded services would be available to users located in countries outside USA only if Google reaches settlements with rights owner’s organisations on a country-by-country basis. Whether this is possible would depend not only on the good will of the parties involved, but also on the copyright legislation of the countries – e.g., whether their laws allow for class actions or extended collective licensing, or whether they have collecting societies or other organisations with sufficiently broad legal authority to enter into an agreement with Google. IFLA is concerned that if the Google settlement is approved in the United States and if Google is not able or willing to reach agreements with rights holders in other countries, the consequence will be an ever-widening inequality in access to books in digital format and IFLA will push for continuing discussions on this matter.
Control of information by one corporate entity
Google has not disclosed the size of the project, but independent experts estimate that it may amount to digitising 30 million books at a cost of c. $750 million. The immensity of the project, and the fact that Google has a 5-year lead, makes it challenging for other enterprises or institutions to start viable competing projects. In consequence, a large proportion of the world’s heritage of books in digital format could be under the control of a single corporate entity should the settlement be approved.
In view of the potential monopolistic nature of the project, IFLA urges the court to exercise its authority to ensure the realization of the broadest possible public benefit from the services enabled by the settlement.
When the digitisation project is concluded, it will comprise a large proportion of the world’s heritage of books in digital format. The participating libraries will have copies of “their” files for preservation or other uses. Although the Google settlement has provisions for business continuity, the settlement does not seem to include provisions for the long-term preservation of the entire database. Analyses of cost effectiveness may at some point in the future lead Google to reduce the amount of data by discarding parts of them. The importance and utility of the entire database for users worldwide requires that the agreement include provisions ensuring the long-term preservation of the database as a whole and IFLA urges the court to take this into its considerations.
The economic terms for the Institutional Subscriptions Database will be governed by two objectives: (1) the realisation of revenue at market rates, and (2) the realisation of broad access by the public, including institutions of higher education. Libraries’ recent experience has been that publishers of scientific journals have given priority to the generation of revenue at the cost of broad access, forcing many libraries to cancel subscriptions. If the beneficial societal effects of the Book Search Project are to be fully realised, it is critical that the importance of broad access be given strong weight in the settlement. In view of the potential monopolistic nature of the project, and the collaborative manner in which it must be implemented, IFLA asserts that libraries must have an integral (not merely advisory) role both in the establishment of pricing for the ISD and the manner in which revenue from it is allocated to the parties, including libraries. It must therefore be possible for any library or institutional subscriber to request the court to review the pricing of services provided.
According to the proposed settlement, Google may exclude from the database 15 % of scanned books that are under copyright, but out-of-print. This could lead to the exclusion of one million books.
Google is likely to come under pressure from interest groups and even governments to exclude books that are purported to contain “undesirable” information. If Google submits to political pressure and removes books from the database, this could lead to the suppression of these books worldwide and the contravention of censorship legislation or rights of freedom of expression. It is therefore of the utmost importance that Google be obliged under the settlement to publish lists of books that are excluded from its services, and the reason for the exclusion.
Patrons’ privacy is such a core value for libraries that a court order is usually required to force a library to disclose individuals’ use of library resources. Some of the services to be offered under the proposed settlement imply that Google will collect and retain information about users’ activities. However, the settlement does not specify how users’ privacy will be protected.
IFLA urges the court to require Google to cooperate with library associations and other representatives of users’ interests to ensure that adequate measures are taken to protect personally identifiable information.
Contracts v statutory exceptions and limitations
In copyright, contracts too often override statutory exceptions and limitations in ways that diminish users’ rights. IFLA asserts that the settlement should therefore clearly state that nothing in it supersedes legislated users’ rights, including specific and general exceptions for libraries and their users, and any existing or new approaches to making orphan works accessible.
The database containing the digital copies of the scanned books represents a unique corpus for computational analysis and research. Under the proposed settlement, Google and two institutions may host this Research Corpus for purposes of “non-consumptive research” by “qualified users”.ii The host site has the authority to decide whether a user is qualified and whether the research is non-consumptive. There is no mechanism to challenge the host’s decision and in consequence, certain types of research may be privileged. There seems to be no possibility for foreign researches to get access to the database for research purposes.
It should be possible to request an independent body to review whether the host site’s decision to refuse certain researchers or research projects is reasonable, and opportunities for research should be available to qualified researchers throughout the world. IFLA urges that this be taken into the court’s considerations.
i The expanded services consist of three primary services:
All users in the United States may search Google’s entire search database for digitized books free, and see up to 20 % text from out-of-print books. (Special rules for special categories e.g. fiction vs. non-fiction.).
Consumers may buy perpetual online access to the full text of out-of -print books. In-print books require that the copyright owner "opt in".
Users within an institution may view the full text of all the books in the Institutional Subscription Database (ISD, which will include all the books in the in-copyright but out-of-print category.
ii “Non-consumptive” means that the text is not accessed for display or reading.
August 28, 2009