Questions Raised About Google Library Project’s Impact On Knowledge Access

By William New
What has been heralded as a breakthrough in the digitisation of human knowledge is also raising questions about how most humans will access that knowledge, according to an expert in copyright and the public interest.

Fred von Lohmann, senior staff attorney at the Electronic Frontier Foundation, recently raised concerns about Google’s new settlement with publishers allowing the search engine to continue borrowing millions of books from libraries and scanning them to make a digital library.

His remarks were made to an international library copyright event in Chisinau, Moldova on 13 November where he spoke on the subject of “copyright’s ever-expanding empire” addressing digital rights management (technologies for controlling copyrighted content), licences and the privatisation of public information.

The key concern is that the Google project, likely to go into effect in 2010, will be in the private sector, which has different implications than public libraries, which von Lohmann described. The settlement itself does not affect scanning (digitisation) of public domain books. It allows scanning of copyrighted, out-of-print books for free viewing of up to 20 percent, with a charge for the remaining 80 percent. For copyrighted, in-print books, Google must receive authorisation from rightsholders before making any part available. In all cases, viewing must take place through Google’s website.

Google will fund creation of a book rights registry to collect royalties for copyright holders, for which it has set aside $34 million, von Lohmann said. The deal gives 63 percent of all revenues to the rightsholders and the rest to Google, which plans to develop new value-added features to encourage use of its library (IPW, Access to Knowledge, 30 October 2008).

The benefit to libraries is that all of their books will be scanned, and new research capabilities will emerge by making works searchable, but every library in the United States only gets one terminal with free access to every book in the Google collection and will have to pay for more, von Lohmann said. Google also will want to scan books in other languages besides English, he added, and presumably will use this settlement as the basis for those negotiations.

On works in the public domain, Google is allowed to scan them and make them fully available. But in the United States, it is difficult to determine which works are in the public domain, especially those from between the 1920s and 1960s, he said. Google’s approach is to scan books without authorisation and remove them if told otherwise, and publishers have agreed not to sue if mistakes are made. But this “safe harbour” against lawsuits only applies to Google, not to libraries or individuals, he noted.

On orphan works, the millions of books who rights holders are not known or cannot be found, Google’s payment to the registry takes care of its obligation, but libraries are not helped by that.

Limits to Library Exceptions?

Yet another question is what will happen to limitations and exceptions to copyright typically granted to libraries. These exceptions depend on the works not being commercially available, but what if increasingly all works are available for commercial use, as in the Google case, von Lohmann asked. Additionally, he asked what will happen when the original books are discarded, but errors are made in the scanning, as has happened with Google. Google should not be depended upon to keep the copies, he said, adding that preservationists should be concerned.

Other concerns are that licences will not provide the books but rather will just be an agreement with terms allowing access to the books kept by someone else (who could change the terms). The issue of exhaustion also comes up, for instance if a licence-holder seeks to transfer the licence (such as in the way donors used to donate their books to libraries). And von Lohmann raised privacy issues as Google will know the behaviour of each user and has not specified how privacy will be protected.

Finally, he pointed to the traditional importance of free public access to information, and that libraries have been “pivotal” for the poor to access information, which might be affected if one must pay a private entity for access.

Alternatives for Libraries

There may be a few alternatives for libraries, von Lohmann said, such as obtaining direct permission from copyright holders themselves to scan books, or through special exceptions that allow libraries to engage in digitisation.

But other options are emerging, such as creating a “crowd-sourcing” library (a term coined by author Jeff Howe that refers to outsourcing a task to a large group of people) allowing users to contribute their “enthusiasm and expertise,” he said. Rather than acting as “retailers” of information, libraries could be seen as bulk providers, such as public.resource.org, which gathers government public domain material and makes it available for free access on the internet.

But building resources based on the public domain relies on content being available, and more and more publishers are making this difficult, he said. For instance, they may propose to governments to publish material for them, but then they require payment to access it.

Another example is the Encyclopedia of Life, which is attempting to create a website with one page for every species on earth that is freely accessible. They are obtaining permission from scientific journals but are running into some difficulties such as in finding the rights holders, and so are considering inviting users to scan and upload text, which would be allowable under US copyright law. In general, von Lohmann envisioned a copyright system that would harness the knowledge that users already possess. He questioned any copyright law that would punish someone for reproducing content that was from an infringing source. “Why not instead think ahead,” he said. “These filesharing services are never going to go away.”

Another option is to provide hosting for the collections of others, he said. Some countries provide limitations on copyright liability for online service providers that store content for users, which has contributed to the growth of internet services such as Google’s YouTube service. Libraries could use the same copyright protections to provide an online repository for their users to deposit digital materials, he said.

In still another model, users can be seen as the library, he said. The original Napster music downloading service, before it was sued for copyright infringement, amassed the most comprehensive library of recorded music in history, all from unpaid volunteers. Libraries should embrace the notion that users could copy and distribute music files, by pushing for a solution to compensating artists while preserving efforts of volunteers, he said.

The Google project was settled out of court, which may prevent the outcome from being a precedent, noted von Lohmann, who added, “I think it [the Google project] raises many questions that are going to be with libraries for many years.”

William New may be reached at wnew@ip-watch.ch.