Librarians seem caught in ambivalence these days about Google Book Search (http://books.google.com), the beyond-mammoth book digitization project currently rolling up to (or past) 8 million books. The next major event in the project's history-the court's approval or disapproval of Google's settlement terms with authors and publishers over copyright issues-will decide whether millions more of those books will become available to all or part of the public. The American Library Association and the Association of Research Libraries have already sent a "hot/cold," "yes/no," "go, but carefully" recommendation to the court. However, the University of Michigan Libraries-one of the first and always the most generous and supportive of the Google Book Search Library Partners-has already made its approval clear by signing a contract that presumes the settlement agreement will go through as planned. In that agreement, however, the university has addressed some of the concerns expressed by librarians, which could set a standard for future agreements and allay concerns between librarians and mighty Google.
Just to recap the history here: The Google Book Search program, particularly the library contributions that dominate the collection and include both in-copyright and out-of-copyright books, was challenged in the courts by both authors and publishers. Before courts could reach decisions on the matter, a settlement agreement was made by all parties. However, the settlement agreement needs the approval of the court, particularly since part of the settlement involves releasing millions of in-copyright/out-of-print-possibly orphan-works to the general public under subscription arrangements with institutions and a limited, free access route for public libraries. (For details on the settlement agreement, read the NewsBreak, "The Google Book Search Settlement: ‘The Devil's in the Details,'" Nov. 3, 2008, http://newsbreaks.infotoday.com/NewsBreaks/The-Google-Book-Search-Settlement-The-Devils-in-the-Details-51429.asp.) When the settlement was announced, parties involved hoped the court would have made its decision by June 2009. But a 4-month postponement means that the court probably won't begin hearings until October.
Some of the additional input made to the court has come from libraries. In particular, the American Library Association (ALA; www.ala.org) and the Association of Research Libraries (ARL; www.arl.org) have submitted a set of comments, available as a PDF file (http://wo.ala.org/gbs/wp-content/uploads/2009/05/googlebrieffinal.pdf). The document seems to blow hot and cold, complimenting and congratulating Google on its magnificent gift to the world, while at the same time raising fearsome doubts about Google as a commercial monopoly. Representing one of the parties to the settlement, Allen Adler, vice president of legal and government affairs for the Association of American Publishers (AAP), considered the document as proof that "librarians are not opposed to the settlement, they have just asked the court for vigorous oversight. So that's not a bad sign. The library community has had diverse views on Google Book Search, especially the library portion, from the outset. So it's not surprising."
The concerns expressed in the ALA/ARL document regarding the settlement are arranged in six sections:
- Creates an essential facility with concentrated control
- Could limit access to the ISD (institutional subscription database)
- Will heighten inequalities among libraries
- Does not protect user privacy
- Could limit intellectual freedom
- Could frustrate the development of innovative services
An eighth section-This Court Can Address The Library Associations' Concerns Through Rigorous Oversight of the Implementation of the Settlement-advises the court on how to protect the country from the problems enumerated while enabling the public and libraries to enjoy the world of benefits.
On May 19, 2009, the University of Michigan expanded its long-standing Cooperative Agreement, which defined its participation in Google Book Search (www.lib.umich.edu/mdp/um-google-cooperative-agreement.pdf), with a lengthy Amendment (www.lib.umich.edu/mdp/Amendment-to-Cooperative-Agreement.pdf), which states that the university agrees to participate in the Google Book Search Settlement arrangements under certain terms and conditions. The Amendment presupposes the court will agree to the overall settlement, although it provides for renegotiation if the court disapproves or significantly changes the proposed arrangements.
Paul Courant, dean of the University of Michigan Libraries, indicated that both parties had been working on this agreement for some time. Though much concern by librarians has focused on possible pricing pressures, Google has not yet set any prices. However, it has indicated a tiered pricing system, based on the size and resources of institutions and/or content selections. In Courant's view, "Google wants eyeballs much more than cash, but authors and publishers want the cash." Pricing issues will arise, but Courant doesn't see them as insoluble. As one of the largest contributors ("fully participating") libraries in the Google Book Search program, the University of Michigan will not pay anything for access to the new millions of books scheduled for release under the settlement agreement. "Most of the other libraries contributing as Google Book Search partners won't contribute enough to get a substantial discount on items," said Courant. "We have tried to represent libraries broadly, but the fact is that only two to three dozen libraries are participating and the majority contribute way less than a million books. The contributing libraries do not get their own digital copy back while the fully participating do. But Google wants people to contribute and, once the settlement is in place, maybe it will go back to the ‘firehose' approach toward digitization [it] used initially with us. On the other hand, [it] still [doesn't] want to do duplicates."
The Amendment does provide detailed procedures for biennial pricing reviews by "Initiating Libraries" to see that the pricing supports both revenue for rightsholders and broad access. However, the Amendment does focus on Google Book Search library partners. Unless the court establishes a system for price review by institutional subscribers in general, which it could do, the provisions might serve more as a guide for other libraries in considering what Google may offer.
The Amendment also provides for conditions where Google might block access to public domain books. It also set a 2050 date for release of nonsettlement digital copy and a 2024 date for evaluating the business conditions of Google Book Search in general. Courant explained his reasons, "Call me a Nervous Nellie, but I'm a student of history. The University of Michigan is 192 years old; Google was eleven years old when we started this discussion. I would be very surprised if 90 years from now there is still an entity called Google. All kinds of things could go wrong. If they do, it's essential that libraries reclaim that corpus for re-use and shopping around. By the way, in arguing about those dates, our strongest argument for 2050 was that none of us would still be alive. We figured that if Google couldn't recoup its investment by then, it never would. The 2024 date was [soon] enough to see how they were making money and, if they weren't, then we could cut loose earlier. It's a very gentle tripwire." Speaking of revenue, under the Amendment, Google can allow the presence of advertising to affect the prices charged for institutional subscriptions.
Other provisions touched on quality issues such as revised or redacted copies. Courant indicated this did not mean small changes, such as individual page fixes, but only a guarantee that if there were significant technical improvements in the quality of scans and the associated optical character recognition (OCR), the university would get sets of the best available copy.
One of the pet projects at the University of Michigan Libraries is the consortium of Google Book Search participating libraries called the HathiTrust. (For details, read Beth Ashmore's NewsBreak, "HathiTrust: A Digital Repository for Libraries, by Libraries," Oct. 23, 2008, http://newsbreaks.infotoday.com/NewsBreaks/HathiTrust-A-Digital-Repository-for-Libraries-by-Libraries-51225.asp.) The Amendment has extensive policies for hosting research corpus laboratories, plus some Google funding, to support and authorize this and other forms of collaboration. Courant stated, "From the beginning we thought it essential that we construct real library collections in parallel, but not the same thing as Google Book Search. The structure of this agreement lets us grow that way. We can't freely read in-copyright works, but we could build concordances and search algorithms and linguistic tools that could recommend works."
Once again, the leaders of the University of Michigan Libraries are pushing forward into the brave new world, working to solve the problems of progress. And according to another Michigan librarian, Google still has at least a million more of its books to digitize.