On Friday, 13 March, I had the privilege of being able to attend a Conference at Columbia Law School on “The Google Books Settlement: What will it mean for the long-term?” While there were a number of librarians in the audience, the great majority of attendees seemed to have been from New York’s publishing world. There was a Twitter feed going on during the meeting, but I generally prefer Rebecca Tushnet’s long, reflective summaries of the meetings she attends. I didn’t see her in the audience, so instead I will share my notes and reactions. I’ll start in this posting with long summaries of the two morning keynote presentations, and in a second posting discuss the afternoon panel discussions – as well as share my own assessment of the whole day.
After an introduction from June Besek of Columbia that briefly summarized the settlement, the meeting kicked off with what was for me the highlight of the day: Mary Beth Peters, the Register of Copyrights, providing her view of the settlement. She started by noting that she had recommended against the Library of Congress participating in Google’s initial Library Partners program because she was not convinced that Google’s indexing of copyrighted books was a fair use. I believe she opened this way in order to establish that a settlement to the case might have been required (if Google's actions were indeed infringing). She is not opposed to the settlement in principle, but she is worried about the scope and the forward impact that the settlement may have. She noted that some critics (including Brewster Kahle and Robert Clarida) have suggested that the settlement may create new rules for the use of copyrighted works, and in effect institute new copyright legislation without having to go through the legislative process that in theory at least is supposed to balance private rewards with the greater public interest. Doing this is really hard – she cited the example of orphan works legislation, which still hasn’t passed. But it does ensure a level of public engagement that is different than in the settlement.
She then posed a discussed a number of issues for which she does not have the answer:
- How does this settlement affect orphan works? Will it make it harder or easier for orphan works legislation to pass?
- The settlement, she suggested, is a compulsory license for the benefit of one company. Could Congress have enacted such a license for everyone?
- Are there treaty obligations that are implicated? Probably yes, if this was legislation – but maybe not if a settlement.
- The Copyright Office sponsored a three-year long study group tasked to examine the exemptions available to libraries, archives, and museums in order to make sure that they are compatible with the digital age. Does the Google settlement obviate its work?
- What about foreign works?
- What about journals? Will journals go to Congress for a legislative solution or will they sue Google, too?
- What about libraries that are not part of the settlement?
- Does the settlement represent the interests of all authors? One would assume that legislation would have acknowledged and addressed all types of copyright owners, but the parties in the litigation were not so constricted.
In sum, Peters suggested that there are huge public policy issues at play in the settlement, issues that might have been better addressed through legislation and the creation of a compulsory license. (A real surprise, since in the past the Office has been opposed to new licenses. I guess this is a mark of how serious the big issues are).
Peters’s talk for me was an illustration of the important issues at stake in the settlement. We have a regulatory mechanism in place to ensure that works can be used in new and creative ways while at the same time respecting the rights of rights holders. It is called copyright law. The Google settlement may make the careful balances found in copyright law (as well as the public procedures to change it) moot, replacing them with private contractual arrangements instead.
A cynic might wonder how much attention has been paid recently to the public interest when changing copyright law. But before he reintroduced the "Fair Copyright in Research Works Act" without changes from last year's bill, Rep. Conyers at least had a public hearing on the proposed legislation and IP law school professors could issue a statement stating that the NIH public deposit mandate had nothing to do with copyright. Conyers chose to ignore the public comments and re-introduced the bill as written by publishers – but at least he had the hearings. That isn’t the case with the Google Book Settlement – there is no public input into its terms.
Most disturbing of all was Peters’s admission that not one member of Congress has asked the Copyright Office to comment on the settlement – even though it may fundamentally change how Americans can access and use copyrighted information. This might be because the enormous sea-change that the Settlement represents has not sunk in with the public. For example, the Copyright Office was worried that there might be a huge rush to register works just prior to the Jan. 2009 deadline in order that they would be eligible to be included in the settlement. Only two publishers, however, sought to register their past publications. (To me, this is also more evidence of how we have to live with copyright rules that were designed to serve and protect a miniscule portion of the country’s creative output, but in the process throttle other types of creativity.) But while the public might not understand how important the Settlement is, I would have assumed that the Copyright Office does. I was therefore disappointed to learn that it will not be commenting on the proposed settlement (even though one assumes that the US government holds copyright on at least some works created by non-government employees and could then comment on the settlement as a member of the class).
Peters was followed by Randal C. Picker from the University of Chicago Law School who spoke on possible competition and anti-trust issues in the proposed settlement.
He opened with some hypothetical examples of fair and infringing uses, and the moved into what he says are the three key components of the Settlement: the creation of a collection of digital files; the scope of rights included in the license granted to Google through the settlement; and the creation of the Book Rights Registry and its relationship with rightsholders. We could if we wanted address competition in each of these three areas. Picker then looked at each in more detail.
With the digital files, Google considers then to be its property, and places restrictions on the files: see 3.12 in the settlement. It also places restrictions on the use of the digital files, such as limiting certain users to nonconsumptive research – you can use the books so long as you don’t read them. Use is also limited to qualified users, and for-profit entities can only use research corpus with G & R’s consent. Google is allowed to use the digital files in a number of ways that may help its business model (search and advertising). Picker thinks Google’s restrictions on the use of the digital files are legitimate.
Picker is troubled by the consumer purchase model found in the settlement. Google is going to create a pricing algorithm to maximize revenue from sales, which sounds a lot like antitrust to him. But after a fascinating review of recent cases involving essential facilities doctrine (Aspen Skiing, Trinko), he couldn’t argue that third party access to Google’s digital file collections should be mandated.
Picker raced over the 3rd key element – the Registry – and postulated that it would be very tough to require a second, competing rightsholder registry. The ASCAP/BMI model wouldn’t seem to work here, but primarily for practical reasons.
So if there is going to be more competition, it will have to be in the licenses that the rightsholders provide to Google and others. And there are elements in the license terms that work against competition. Foremost among these is the “Most Favored Nations” clause, 3.8(a), which benefits Google only. In the settlement agreement, only Google gets to make use of orphan works – those works whose rights holders to not announce themselves to the Registry. The Registry would be able to offer license terms to others for those works with “active rights holders” who have made themselves known to the Registry, but they cannot license works from “non-active rights holders.” (Picker said “active authors,” but I suspect he really meant rights holders.) Even more troubling, the royalties from the use of the orphan works go not to the holders of the rights in those works, but to all the other active authors and Google. In Picker’s view, the settlement allows Google and the Registry to turn orphan works into a private public domain.
Picker sees to possible solutions to this antitrust problem. First, he thinks other for-profit firms should be able to use orphan works under the same term as Google (even though the Registry itself does not have the authorization of the rights holders to negotiate on their behalf). Second, he thinks that non-profits should be able to use orphan works without liability until such time as a rights holder steps forward.
The morning speakers (and the Q&A that followed) identified the major themes that would flow throughout the rest of the day. All three speakers said that they found the settlement difficult to read and understand. While people seemed comfortable with the idea of rights holders getting paid for the use of their work, there was less comfortable with the idea of rights holders getting paid when other people’s works were use. And while both Peters and Picker found elements in the settlement that they found troubling, neither suggested that the settlement should not be approved.