Singularity Law

Last month I was honored to speak at the mid-Winter Copyright Society U.S.A. meeting in San Francisco. My presentation was focused primarily on the Google book search settlement, and its implications for copyright owners in general. I had previously written on how I felt that Google had gotten the better end of the deal, which I reiterated at the meeting. That generated a lively debate, including comments from some of the attorneys present who had apparently been involved with the settlement discussions.

Even after hearing their side of the story, I am still convinced, that the book industry got a raw deal.

Factual Background

The facts of the case are straightforward. In 2002, Google began quietly digitizing books. Their goal apparently was to digitize every book ever published, whether or not it was still under copyright. The “Google Book Search” project was officially launched in 2004 with agreements with numerous large, research libraries that had agreed to give access to their collections to Google. The Google Book Search service permitted Internet users to search the complete text of the books that had been scanned. However, depending on whether or not the book was still under copyright, the user could either display the entire text (public domain books) or short snippets of text before and after the keywords searched for (in-copyright books).

In September 2005, Google was sued for copyright infringement by the Authors Guild, the American Association of Publishers and a number of individual authors in a class action. The plaintiffs claimed that the project was “massive copyright infringement” and should be shut down. The litigation dragged on for over 3 years, during which time Google scanned, digitized and indexed over seven million books from major university libraries.

Summary of Settlement

On October 28, 2008, the parties announced a settlement of the litigation. The agreement, stretching to 141 pages, provides that Google will pay $125 million plus establish a new licensing system with publishers. The system will allow any copyright owner to opt-out of the settlement, will require Google to pay 63% of all revenues generated by the use of copyrighted books to a “collection society” (the “Book Rights Registry”) for distribution to copyright owners, and require the implementation of a DRM/subscription model for full-text access, but not for searching –- which will be limited to short snippets for in-copyright works.

Who Won?

While the copyright owners will receive compensation for the use of their in-copyright, but out-of-print books, and will require Google to negotiate separate deals with copyright owners of in-print, in-copyright books, the settlement, in my opinion (as well as an increasing number of critics) definitely favors Google for a number of reasons.

1. First mover advantage. Google has been feverishly digitizing books for over 4 years now. It has processed at least seven million books. While perhaps not a “critical mass” of books yet, it is an enormous head start over any potential competitor.

2. All publishers bound by the settlement. Because the case was filed as class action, all publishers will be bound by the terms of the settlement unless they choose to opt out. This includes not just U.S.-based publishers, but publishers in every Berne Convention country – basically every country with a publishing industry. For publishers that do not opt out, they will be bound by the terms of the settlement. Google does not have to negotiate separately with every publisher (except those that might opt out). Any future competitor, however, will need to negotiate with every publisher – a very expensive proposition – and there is no guarantee that a competitor would be able to cut an identical deal with every publisher, or even a deal that is as good as what Google is getting. Indeed, if the settlement plays out as I envision, publishers will be so angry that any competitor that attempts to duplicate Google’s service will be in for a rough time cutting a reasonable deal.

3. Barriers to entry. Google already has deals with a number of the major research libraries to have access to their collections for free, in exchange for which they will get a copy of the digital database (and some other benefits). Any competitor probably will not be able to make the same deal with these major libraries. It is unlikely that the same libraries would want to put up with the disruption to their organization of having multiple scanning projects going on at the same time. Competitors would need to find other collections to digitize, or pay a large amount of money to these libraries – a cost that Google did not have to incur.

4. High settlement amount. The large amount will discourage others from entering the field – giving Google a virtual monopoly on every book coming within the settlement. Publishers will not be willing to give subsequent competitors a better deal.

5. Others cannot claim fair use. It will be more difficult for others to claim “fair use” if they decided to compete with Google. While there was no determination of whether Google’s conduct was or is a “fair use,” a judge in a subsequent case may look at this settlement as “evidence” that massive digitizing of copyrighted works is not a fair use – otherwise, why would Google, which claimed fair use all along, have settled for such a large amount? A decision in this case that Google’s use of copyrighted material was a fair use would have opened the door for many competitors to piggyback on the court’s decision and built competitive databases protected by fair use. So a lack of a ruling on this pivotal issue definitely plays into Google’s hands.

6. The settlement grants a compulsory license to Google for all books printed prior to February 5, 2009. The settlement establishes a legal business model for Google to monetize this huge and growing database of copyrighted materials with the copyright owners’ blessing. Google get 37% of all revenues generated, where previously it was getting nothing. (This is a much larger percentage than iTunes gets for their music downloading service.)

The settlement agreement is, in essence, a compulsory license for Google to copy virtually any copyrighted work without having to negotiate individually with copyright owners. And it sets the “market rate” for any negotiations Google will have in the future with copyright owners who are not part of the settlement.

7. Google gets to set the rates for the various revenue streams. Perhaps one of the most amazing aspects of the settlement is that Google gets to decide how it will monetize the sale of the plaintiffs’ copyrighted works going forward. I cannot think of any previous copyright infringement suit where the alleged infringer not only got an expansive license to continue its allegedly infringing activities, but the infringer is given greater rights to the works than when it was allegedly infringing, and given complete discretion over how it will monetize the exploitation of the plaintiff’s works in the future. (It would be as of the record companies had settled their copyright infringement suit against Napster by giving Napster complete access to their entire catalogs, and letting Napster set the price for selling those songs to its file-sharers, including the right to continue to allow file-sharers to download the songs for free!)

Under the settlement, Google has the right to make money from (i) selling downloads of in-copyright but out-of-print books, (ii) selling advertising on pages in which the copyrighted works are accessible by users, and (iii) selling subscriptions to libraries for access to the book database. But it is Google, and not the copyright holders, who gets to set the price for these revenue streams. Theoretically, Google could charge nothing for these uses of plaintiffs’ books, and the book publishers would get nothing more than the settlement amount. While that is not likely to happen (since publishers still have opt-out rights that could be exercised if they are dissatisfied with the way the scheme is working), there is nothing in the settlement agreement that requires Google to charge a minimum amount, or anything at all.

8. Few copyright owners are likely to opt out. While copyright owners are free to opt-out of the system, it would be foolish for most copyright owners to do so. They essentially have three choices – go along with the system, opt-out and forego the revenues they would get under the system, or try to set up their own system to monetized digitized version of their works (which hasn’t worked in the music industry, and is unlikely to work here). A fourth choice — to do a deal with another database developer — is unlikely for the reasons stated above.

9. The Book Rights Registry will operate at the publishers’ expensive, not Google’s. While Google has earmarked $37 million to establish the “Book Rights Registry,” the actually operating expenses will be borne by the publishers out of their cut. This will save Google untold millions in administration costs, and having to track down the copyright owners of orphan works, works owned by publishers who have either gone out of business or merged into another publisher, and individual copyright owners who may have died and the current ownership of the copyright is splintered amongst his or her heirs.

Conclusion

The cost to Google to settle ($125 million) sounds like a lot on the surface, but will average out to far less than $1 per work that Google will be digitizing over the next several decades. Certainly a cheap price to pay for a compulsory license to a treasure trove of the world’s greatest (and not so great) books.