"Proverbs for Paranoids, 1: You may never get to touch the Master, but you can tickle his creatures."

- Thomas Pynchon, Gravity’s Rainbow, 1973


A colleague in Europe recently forwarded to me the Google agreement with the CIC libraries. Even though I had been told this new agreement had some very different language from that in prior contracts, it was still eye-opening reading.

Simply put, the CIC libraries are contributing in-copyright material to Google for scanning, but for the first time (known to me), they will not get a copy back. (Michigan and Wisconsin have previously negotiated arrangements with Google, and the terms of the CIC agreement hold only to the extent that it permits the two universities to participate in a CIC-wide project; the contract terms of Michigan and Wisconsin are not superseded.)

Let's go to the contract. Clause 4.7(b) states:

Escrow Deposit. As Google "successfully processes" the works contained in the Selected Content, Google will place the University Digital Copy of such Selected Content in escrow on a secure server maintained by Google at Google’s cost and expense.

(The University Digital Copy being a copy of the material digitized by Google.)

Clause 4.9(a) then stipulates that "Works in dispute" may be withheld from Escrow deposit, and 4.9(b) permits a CIC institution to assert that a work thought by Google to not be public domain, is actually in such a condition, and as long as the CIC University partner indemnifies Google and provides assurances that any claims can be addressed, the institution can receive a copy of it. (This is a high bar, so don't anyone wait in line for it to be invoked unless there is fair certainty; that said, I know the University of Michigan has encountered situations [podcast] where they feel Google treads extremely conservatively, and where a reading of the law is clear enough to grant the University assurance to act.)

It is in "4.10 University Digital Copy of In-Copyright Works" where the story starts getting interesting:

In General. As noted in Section 4.1 above, Google may, in ways consistent with applicable Copyright Law, select and Digitize In-Copyright Works contained in the Selected Content. Such works will be part of the University Digital Copy and, as such, the Digitized files will be maintained in escrow as set forth in Section 4.7 above and released to the Source CIC University as set forth in Section 4.11. Until such time as these In-Copyright Works are released, Google agrees to provide CIC Universities with searchable access to such In-Copyright Works as described in Section 4.6 above.

(4.6 provides for a hosted solution by Google for the CIC libraries.)

Then come the constraints:

4.11 Release of In-Copyright Works Held in Escrow. Subject to the terms of this Section 4 Google agrees to enable download capability from the escrow to the CIC Administrative Offices or the applicable Source CIC University for one copy of the digital file for any In-Copyright Work(s) held in escrow in the event that any of the following release conditions (each, a "Release Condition") occurs:

(a) the In-Copyright Work becomes in the public domain;

(b) a Party has obtained permission through contractual agreements with copyright holders that includes the right to make a copy of the In-Copyright Work and to provide it to the CIC or Source CIC University;

(c) well established case law exists that In-Copyright Works can be copied and held by the CIC Administrative Offices and/or the Source CIC University without infringing on the rights of a copyright holder;

(d) if at any time Google is in material breach of its obligations under Section 4.3(b) or 4.6(a) and Google does not remedy any such failure within ninety (90) days after its occurrence (or, in the event such failure is caused by technical problems or causes similar to those described in Section 12.5, within such longer period as Google, working diligently, reasonably requires to remedy such problems); or

(e) the CIC Administrative Offices or the Source CIC University and Google agree in writing that the release of a particular In-Copyright Work or Works is legally supported and appropriate under the terms of this Agreement.

In other words - pretty much, unless Google ceases business operations, or there is a legal ruling or agreement with publishers that expressly permits these institutions (excepting Michigan and Wisconsin which have contracts of precedence) to receive digitized copies of In-Copyright material, it will be held in escrow until such time as it becomes public domain.

That could be a long wait.

Why was this done this way? What, after all, are the chances that residents of these U.S. Midwestern States might actually receive a digitized copy of the works that their tax dollars made a contribution towards purchasing? (N.B.: Not all CIC members are public institutions, but the overwhelming majority of them are.)

The answer to this question may well depend on the outcome of the larger contest.

In an article early this year in The New Yorker, "Google's Moon Shot," Jeffrey Toobin discusses possible outcomes of the antagonism this project has generated between Google and publishers.

Paramount among them, in his mind, is a settlement:

Google's endeavor is encountering opposition. A federal court in New York is considering two challenges to the project, one brought by several writers and the Authors Guild, the other by a group of publishers, who are also, curiously, partners in Google Book Search. Both sets of plaintiffs claim that the library component of the project violates copyright law. Like most federal lawsuits, these cases appear likely to be settled before they go to trial, and the terms of any such deal will shape the future of digital books.

Toobin then goes on to delineate what such a deal might actually look like:

The terms of such a deal aren't hard to imagine. The Authors Guild is concerned that pirated copies of the books on Google's site could leak to the public, and so the organization would insist on security measures. [...] As for distribution of the proceeds from the site, Google might agree to share revenue with publishers, in the way that radio stations pay for the music they play; publishers could receive a fee based on a statistical analysis of how often their books are viewed. Google could pay in cash or in kind, with advertising.

That's an eye-opening observation, and from my perspective as the Director of the Digital Library Federation, one that has not received enough speculation.

Obviously, any settlement would not cover the In-Copyright materials that are already part of Google's Partners' Program, in which publishers submit their material directly to Google for digitization (if necessary) and inclusion in Google Book Search. And, I think by any likely definition, a settlement would not include public domain material, which is differentially encumbered in these agreements. Therefore, a settlement must quite manifestly concern itself with works that are believed to be In-Copyright but where no publisher has stepped forward with an explicit opt-out, or where there is suspicion that they might be In-Copyright, but it is not known for certain (i.e., they are "orphan" works).

If this is what a settlement might cover, publishers will race to establish their historical rights portfolios with a zeal that will be astounding to watch.

More importantly, as Toobin intimates, remuneration must be involved, and it is at least open to suspicion that libraries will have to license access to the material covered in a settlement. (Perhaps paranoid on my part, but I have been long concerned about the new liabilities associated with moving into a realm of digital monograph licensing). The publishers and authors are not suing over tiddlywinks; there is real money at stake for them, far into the future, and the legal bills alone are already breathtaking. I find it hard to wrap my head around how a voluntary collective licensing arrangement such as the one Toobin describes might work in practice, but if the opportunity arose, I suspect that Google is clever enough to come up with something.

(I should note that a settlement would have several beneficial secondary consequences: someone, for example, will have to create a registry to record the known rights status of orphan works, a goal that everybody who is sentient, reasonably sane, and not drugged out of their minds desires. A derivative concern is whether or not such a registry would be public, or more specifically, what parties would have access, through what means, and under what conditions.)

Let us briefly return to the CIC deal. If a settlement along the lines that Toobin suggests occurs, then it is seriously in question whether the CIC institutions (again, excepting Wisconsin and Michigan, which have prior digitization agreements) would be able to obtain access to their In-Copyright works until such time as they fall into the public domain; I find it hard (not impossible, but hard) to imagine why publishers, as a community, would permit the CIC to obtain such copies; the "library copy" is something that has deeply irritated them since the Google Book Search program started.

I think the CIC agreement is a significant enough departure from the prior public contracts that we must take notice of its suggestions that the relationship between Google and publishers is maturing, and that Google is more cautious of the distribution of In-Copyright material than they ever have been before.

That may well be the harbinger of something broader; something that libraries and the larger public might not find much cause to celebrate. Observing, "[A] settlement that serves the parties' interests does not necessarily benefit the public," Toobin invites Larry Lessig to comment:

"It's clearly in both sides' interest to settle," Lawrence Lessig, a professor at Stanford Law School, said. "Businesses in Internet time can't wait around for years for lawsuits to be resolved. [...] For the publishers, if Google gives them anything at all, it creates a practical precedent, if not a legal precedent, that no one has the right to scan this material without their consent. That's a win for them. The problem is that even though a settlement would be good for Google and good for the publishers, it would be bad for everyone else."

Being neither Google nor a publisher, I'm in the part of the global community encompassed within Lessig's "everyone else."

Lessig continues:

"If Google says to the publishers, 'We'll pay,' that means that everyone else who wants to get into this business will have to say, 'We'll pay,' " Lessig said. "The publishers will get more than the law entitles them to, because Google needs to get this case behind it. And the settlement will create a huge barrier for any new entrants in this field."

A settlement between Google and publishers would create a barrier to entry in part because the current litigation would not be resolved through court decision; any new entrant would be faced with the unresolved legal issues and required to re-enter the settlement process on their own terms. That, beyond the costs of mass digitization itself, is likely to deter almost any other actor in the market.

And that to me is potentially the saddest loss, should such an arrangement come to be realized. Because in real terms, across this vitally important collection of humanity’s literature and thought, of all the ways of thinking about books and working with ideas on the Web, we might be left with only one way.

- Peter Brantley, Digital Library Federation, 2007:06:13