Intellectual Property Law

Google settlement: Changing Defaults

By Susan Crawford

Back in 2004, Google felt quite strongly that its “Book Search” project was a fair use of copyrighted material.  It kept pointing out that it wasn’t displaying complete copies of books covered by copyright.  Instead, it scanned the material and then displayed snippets from these books in response to queries.

I was persuaded by this argument.  It seemed to me that the making of the first copy - the copy needed to generate the snippets - was a necessary step to providing a card catalog, in essence.  An incidental copy.  The snippets themselves, I thought, were fair because they weren’t supplanting the marketplace for the book.  There was no existing marketplace for “licensing-the-copying-of-books-for-use-by-search-engines” at the time that Google started its project.  It seemed to me as if the publishers who sued (and perhaps not the authors being published) wanted to be sure to participate in Google’s profits in a way that went far beyond what copyright law would require.

Yesterday’s settlement agreement is remarkable in many ways.  It’s a proposed settlement of a civil, private lawsuit, but the agreement feels public.  It affects an entire industry, not just the parties concerned.  It sets up a new kind of special-purpose collective rights association (h.t. James Grimmelmann), like ASCAP or BMI.  Instead of Google acting to create access to a great library of books, it seems to point to the creation of a tremendous bookstore.  Perhaps that’s the same thing, but it’s worth thinking about the changed default settings that this arrangement creates.

Here’s a simple default-change:  In the original project, Google was scanning in-copyright and in-print books.  Although it wasn’t displaying them in full, it was displaying snippets - a few sentences surrounding the sought word.

Now, if I understand the agreement correctly, the default will be that a search triggering a response from an in-copyright book will lead to bibliographic material.  Nothing else will be shown *unless* the publisher affirmatively requests that the book be part of the displayed material.  If that happens it will be easy for us to buy the book, we’ll get to see much more of it online than only snippets, and 2/3 of whatever money Google makes from the entire experience will go to the new collective rights organization (the Registry).

That may be all to the good, but it’s a substantial change, and it may affect arguments that other online players want to make about the legality of their content-related activities.

On the other hand, a very positive additional outcome from the settlement is that participating libraries will be able to use complete digital copies of the scanned books. And Google points out that the deal will bring millions of books to light online.

The bottom line:  In 2004, Google thought it was shaping its conduct according to the law.  But Google’s lawyers also have to shape the working situation in which Google operates.  That working situation, from 2004 until yesterday, has involved great bitterness coming from publishers who could otherwise be Google’s allies.  The pure legal principle of the fair use argument, weighed against that bitterness and the possibility of a win-win deal, had to defer.

This settlement, and whatever will happen in the Viacom case, signals yet again (as if we needed reminding) that Google isn’t necessarily in the business of serving long-term policy arguments that would be “good” for everyone else.  The company certainly has the long-term best interest of the internet as part of its DNA, and its financial future is (luckily for us) inextricably intertwined with the internet’s success.  But there will be moments when other interests trump.

Full post as published by Susan Crawford Blog on October 29, 2008 (boomark / email).