Revised Google Book Search Settlement from a Library Perspective

Posted on     10 minute read

× This article was imported from this blog's previous content management system (WordPress), and may have errors in formatting and functionality. If you find these errors are a significant barrier to understanding the article, please let me know.

Late, late in the day last Friday, the principle parties in the Google Book Search case submitted a revised settlement agreement agreement to the court. This post takes a look at the changes to the settlement from a library perspective. To keep this manageable, I'm not including discussion of library-oriented elements that haven't changed; to read more about that I recommend the ALA/ACRL/ARL paper and/or previous posts on DLTJ. I'm also not including discussion on some aspects of the legal impact of the settlement (the appropriateness of setting policy via class action, the antitrust considerations of Google's sole license to unclaimed works, etc.); for that I encourage browsing the writings of James Grimmelmann (any posting of his prefaced with "GBS" in the title). I will link off to some of the library-oriented discussion pieces of Grimmelmann and others in this post. If you really want the in-depth view of the settlement and the surrounding discussion, visit The Public Index, a website devoted to chronicling and commenting on aspects of the settlement.

How We Got Here

Even with the previous caveats, though, it is probably useful to review how we got to this point. Back in 2005, the Authors Guild (AG) and the Association of American Publishers (AAP) sued Google over their scanning, indexing, and display activity of books scanned from libraries. All was pretty quiet for years until October 2008 when the parties filed a request with the court to form a class action (expanding the original parties to now include all copyright holders of books) and a corresponding settlement of that class action. Many agree that this is a pretty wonky use of class action law as a way to solve the problem of works for which copyright holders couldn't be easily identified (the so-called "orphan works" problem). In November 2008, the judge agreed to allow the class action to move forward, gave preliminary approval to the settlement, ordered that the settlement notice be published, and opened up the opt-in/opt-out/objection process. It was supposed to end in May 2009, but there were many requests for an extension due to the complexity of the settlement and the judge granted an extension until September. The fairness hearing for the settlement was to be on October 7th, but the parties to the lawsuit asked for a postponement because they wanted to submit an amended settlement to deal with the objections. Notably, the U.S. Department of Justice stepped in at nearly the last moment with issues regarding the settlement. The judge agreed to the request and set a new date in early November to receive the revised agreement.

The Amended Settlement

On November 13th, the parties submitted a revised/amended settlement agreement for consideration by the court. (If you want to do your own comparison of past-versus-new, take a look at the marked-up "readline" version of the settlement showing the changes.) We're now waiting for the judge to act on the request for preliminary approval of the amended settlement. As Grimmelmann notes, the calendar of events from here on out will probably look something like this:

  • Notice begins: Monday, December 14, 2009.
  • Opt-out/objection/amicus deadline: Thursday, January 28, 2010 (45 days later).
  • DOJ files its response: Thursday, February 4, 2010 (7 days later).
  • Plaintiffs move for final approval: Thursday, February 11, 2010 (7 days later).
  • Final fairness hearing: Thursday, February 18, 2010 (7 days later).

Described below are the changes that are likely of interest or have some effect on libraries or the library world in general (plus some really odd stuff at the end). In the case where section numbers are listed, they refer to the numbered sections of the amended settlement. Capitalized words/phrases have defined meanings in the settlement agreement or associated documents; if you are really curious, you can look them up there.

It is also important to note what hasn't changed. The terms of institutional subscriptions in the amended settlement agreement are essentially identical ((with the exception of adding the word "amended" in front of "settlement agreement" and other such editorial modifications.)) to the terms specified in the original agreement. ((It should be noted, though, in the amended agreement between Google and the University of Michigan there are guidelines for handling disputes in the pricing of institutional subscriptions.)) Nor is there a change to the number of terminals granted to a higher education institution. (The number of terminals to public libraries is now more flexible; see below.)

Definition of Book

The definition of Book (§1.19) in the amended settlement has now been limited to items with a place of publication in the United States, Canada, the United Kingdom and Australia. This would seem to address many of the objections that were raised by European countries that they should not be bound by this agreement. An article in Library Journal quotes a Wall Street Journal article as saying the change "would cut the number of works covered by the settlement by at least half..." In the press conference call that announced the amended settlement, Google stated its desire to work with the governments/courts/publishers of other countries to create similar deals with them. ((Noted by Danny Sullivan of Search Engine Land in his summary of the conference call.))

Also, in §2.2 ("Authorization of Google, Fully Participating Libraries and Cooperating Libraries") we see the explicit exclusion of microform-format materials from inclusion in what can be scanned from libraries. Previously, microform materials were explicitly included. I wonder if the change might be because the act of microforming a book creates copyright rights for that microformed derivative. The area of rights is already so complicated, the lawyers might have been looking to trim down their troubles.

Definition of Periodical

Periodicals are a special class of content as defined in the settlement (§1.104) because they are specifically excluded from the definition of a Book: "The term 'Book' does not include: (i) Periodicals..." The definition of Periodical is pretty consistent with what you'd expect -- a serial publication intended to be produced indefinitely with continuity from issue to issue, etc. In the amended settlement, the definition of Periodical now explicitly includes "any book form compilation of the foregoing." I've got to wonder if they mean "bound" periodicals. Bound periodicals, of course, would be included in the library stacks sucked up by the Google Book Search scanning teams -- either as explicit locations within the library or interfiled with the monographs based on classification numbers. We know that copyrighted journals are in the Google Book Search database from the participating libraries. But this definition of Periodical excludes those scanned versions from the settlement agreement. Does that mean there is still a liability hanging out there for these scanned journals from library collections? It might, especially taken in combination with this addition to §7.2.a.iv on liability limitations: "This Amended Settlement Agreement neither authorizes nor prohibits, nor releases any Claims with respect to, any volumes that are Digitized by Google and provided to any Fully Participating Library except and solely to the extent that such volumes are Books or contain Inserts." (Note: Google does have a program for indexing/displaying current Periodicals, similar to the program for Google Book Search for Publishers.)

Definition of "Institutional Consortium" Changed

§1.76 removes the exception that "Online Computer Library Center (OCLC) affiliated networks" were not included in the definition of "Institutional Consortium." The definition of institutional consortium remains as those consortia in the United States that are members of ICOLC. That itself is still sort of an odd definition because the "Coalition [is] an informal, self-organized group" -- if you say you are a member of ICOLC, you are a member of ICOLC.

Number of Public Access Terminals

A key part of the agreement, from the perspective of libraries, was the the inclusion of a free Public Access Service for library patrons. The original agreement specified that Google would provide one terminal per 4,000 FTE students at Associate's Colleges, one terminal per 10,000 FTE students for other not-for-profit higher education institutions, and one terminal per public Library Building. The amended settlement in §4.8.a.3 provides for the possibility that the Registry may authorize additional terminals per public Library Building at the discretion of the Registry. (The Registry, of course, doesn't actually exist yet, so we can only guess if they would offer additional stations to public libraries, or under what circumstances.)


A big part of objections from libraries is the disparity of privacy expectations between how libraries handle patron records and the more permissive way that Google logs and tracks users' activities. The amended agreement does include a new section (§6.6.f) on privacy: "in no event will Google provide personally identifiable information about end users to the Registry other than as required by law or valid legal process." The settlement is silent on the disposition of usage records within Google. This does not satisfy the concerns of the Electronic Frontier Foundation, among others.

Changes to "Additional Revenue Models"

The original settlement agreement included several other ways that money might be earned from the scanned books: Print on Demand, Custom Publishing, PDF download, Consumer Subscriptions, and Summaries/Abstracts/Compilations. (The other ways, as specified in the agreement, being Google Ad links, institutional subscriptions, and title-by-title consumer purchases.) Of these, only Print on Demand, File Download, and Consumer Subscriptions exist in the amended settlement. (§4.7) Gone is Custom Publishing, which would have allowed for the per-page pricing of derivatives for "course materials" or "other forms of custom publishing for the educational and professional market." Gone, too, is the sale of derivatives of a Book; presumably this would have come in the form of some computer-generated "Cliff Notes" form. It is interesting to note, though, that "PDF Download" was changed to "File Download" and now includes EPUB as well as "other formats for use on electronic book reading devices, mobile phones, portable media players, and other electronic devices." A return, perhaps, of the text-to-speech function that was so controversial in the latest edition of the Kindle device?

Disposition of Unclaimed Funds

In the amended settlement, the destination of unclaimed funds was split into two pieces: funds received by the Registry for books that haven't been claimed by a Rightsholder and funds received for books that have been claimed by a Rightsholder but the Rightsholder has now disappeared. In the latter case, the funds are transfered to the "appropriate governmental authority" for such abandoned funds (§6.3.a.ii). This was a large part of the objections made by states attorneys general.

In the case of the former, after six years and in every year afterwards up to 25% of the unclaimed funds from unclaimed books can be used by the Registry to attempt to locate Rightsholders (§6.3.a.i). After 10 years, the remaining funds from unclaimed books will be given to not-for-profit "entities that advance literacy, freedom of expression, and/or education" in the U.S., Canada, the U.K. and Australia.

Inclusion of Creative Commons Licenses

A new section (§4.2.a.i -- "Alternative License Terms") was added that would enable Rightsholders to specify the use of a Creative Commons license for a work that would, in effect, open up the use of the item for no cost (§4.2.b.i.1) from within the Google Book Search platform.

Tightening Up Timings

There are several places we see where expectations of how long activities guided by the settlement agreement should take are tightened up. For example, in §3.5.a.i we see this marked up text: "A Fully Participating Library will implement a Rightshoder's Removal direction withinfor a Book as soon as reasonably practicable, but in any event no later than ninety days after notice from the Registry." Also in §3.5.b.i we see this marked up text: "Google will implement a Rightsholder's exclusion direction withinpromptly, but in any event no later than thirty days after notice from the Registry...".


  • Rightsholders can now specify minimum and maximum pricing for the title-by-title consumer purchase option. (§4.2.c.i)
  • The Pricing Algorithm is now unilaterally set by Google; it used to be based on common agreement between Google and the Registry. (§4.2.c.ii.2) There is still a way for the Registry to check up on the results of the algorithm. (§4.2.c.ii.3)
  • Rightsholders can now renegotiate the 70%/30% revenue split specified in the settlement agreement. (§4.5.iii)
  • The secret Right-to-Terminate Agreement (Article 16 of the old agreement) is omitted entirely in the amended agreement.

Small and Odd Stuff

Sometimes I wonder what actually goes on in some of the back-room negotiations for these agreements. For instance, according to §1.19, the definition of "Book" no longer includes calendars. Someone thought it might? Also, in the definition of "Principle Work" the example was changed from "The Old Man and the Sea" to "To Kill a Mockingbird". A lawyer wasn't a fan of Verlag's work?

The text was modified to update a link from to on November 16th, 2012.

The text was modified to update a link from to on November 21st, 2012.