This is an archive page. What you are looking at was posted sometime between 2000 and 2014. For more recent material, see the main blog at http://laboratorium.net

How profitable is Google? Imagine an ATM, spitting out $20 bills, ten per second, running twenty-four hours a day, year-round. That gets you to the right order of magnitude.

A small bit of housekeeping on the docket this morning. When the parties filed their proposed schedule for the amended settlement, they suggested February 4 as a deadline for the U.S. government to submit its views. Judge Chin’s scheduling order, however, didn’t actually set such a deadline. The Department of Justice wrote a letter to Judge Chin, saying they were also fine with February 4, and ask him to formally approve that deadline. He did.

A correspondent writes:

So here I am at home writing up a promotion narrative and I don’t have the faculty member’s book here — I forgot to schlep it home from school — but I need to look at it. Google Books to the rescue! There it is for me to page through! And I can even search for the passages I want, something I couldn’t do with the hard copy. Also, I like the book so much I think I may get myself a copy.

Robert Darnton has another essay in the New York Review of Books, this one on the amended settlement. Not much new if you’re familiar with Darnton’s general views and with the settlement, but, as always, he writes beautifully.

I’ve never met him, but I was on his mailing list. I was surprised, but not that surprised, to learn that he’s bipolar.

As the Google Books settlement continues to evolve, Jonathan Band is right there with it, summarizing its provisions for the public. He’s penned a third entry in his Guide for the Perplexed series (Part I dealt with the original settlement and Part II with the revised library agreement with Michigan), this one detailing the settlement amendments. It’s a quick, very helpful overview of what’s changed.

James Grimmelmann, James Grimmelmann on The Google Settlement: What’s Right, What’s Wrong, What’s Left to Do, Publishers Weekly, Nov. 23, 2009:

On Google’s home turf—search—what it does is almost wholly good. When search engines work right, they empower users to seek out whatever they want to learn. That’s the exact opposite of broadcasting, in which a few big speakers choose what everyone else hears. In a world where everyone can self-publish, search engines turn what would otherwise be deafening cacophony into the best party ever, where every guest can instantly join the conversation that most interests them. That’s good for freedom, and good for democracy. Building better search engines is a moral imperative. …

A comprehensive digital books database would be a very good thing, but expediency is no substitute for the rule of law. Even if this settlement is “fair, reasonable, and adequate” to copyright owners, as the law requires, it would still be important to make sure that it doesn’t create a bad precedent for future class actions. What if Union Carbide had been able to settle the Bhopal lawsuits on terms that allowed it to keep on pumping out toxic methyl isocyanate in the future, provided it split the revenue with people living nearby? …

I had hoped that Settlement 2.0 would fix these problems, but it’s turned out to be more like Settlement 1.1: one big feature cut and a bunch of small bug fixes. The feature cut is that most foreign books—those from everywhere besides the U.S., the U.K., Canada, and Australia—are out of the settlement entirely. The new class excludes continental Europe, Japan, and New Zealand, home to the squeakiest wheels the first time around. They wanted out of the settlement, and they got it.

My blogging over the last few months or so may make me seem like Johnny One-Note, and it is true that Google Books issues have been absorbing a lot of my attention. But I do get out of the house now and then. I’ve been to some fun conferences in the last few months, and I’ve finally gotten my slides from them online.

• This weekend, I went to Denver for two terrific events. The first was a Silicon Flatirons roundtable organized by Paul Ohm on intermediaries. I was asked to say a few words to set the stage, so I talked about the kinds of distinctions we might make among different online intermediaries. Due to the room setup, my slides went unused, but here are the slides I would have shown: Anatomizing Intermediaries.
• The second event was a Cyber Civil Rights symposium organized by the Denver University Law Review. I tried to sidestep some of the customary debates over intermediary liability by thinking about unmasking harassers as an alternative to suing them. My comments will be published in the Law Review’s online component, but in the meantime, here are my slides: The Unmasking Option.
• Last week, I went to Trebor Scholz’s The Internet as Playground and Factory conference, where I gave a slightly updated version of my Ethical Visions of Copyright Law talk.
• I gave an updated version of my stump speech on Facebook to a October workshop at NYU on privacy regulation. The organizers—Ira Rubinstein, Helen Nissenbaum, and Kathy Strandburg—did a really crackerjack job in filling the room with interesting people. NYU is a hotbed of good work on privacy and privacy policy (see, e.g., Helen’s forthcoming book), although I don’t get the sense that the law school really understands what a good thing it has going. Anyway, my stump speech has evolved sufficiently since the days a year and a half ago when I was giving it as “Peer Produced Privacy Violations” that I thought I might usefully put online a fresh version of it. In the conclusion, I play with an idea that’s increasingly intriguing me: what if we treated Internet privacy the way we dealt with product safety? Along the way, I rant about how the kids these days really do care about privacy, if not in the ways we codgers always expect. Please enjoy The Myths of Privacy on Facebook.
• I’ve also put online the slides from a longer talk on social network site privacy I gave this summer at a conference at the Open University of Catalonia. This one largely recapitulates the argument of Saving Facebook, but with pictures.
• Back in August, I presented at the IP Scholars conference (on behalf of Paul Ohm and myself) a work-in-progress version of our book review of Jonathan Zittrain’s The Future of the Internet—And How to Stop It. Shortly after the conference (and partly as a result of feedback I received there), we decided to split the project into two parts. The first half, a positive review of the book itself, will be forthcoming in the Maryland Law Review in mid-2010. The second half, a larger assessment of the intellectual currents of Internet Law, is still on the assembly line; it has a motor and frame, but no wheels yet.

As always, my slides are available from my presentations page.

New indoor record for least time elapsed between delivery of food and “How’s everything?”: 37 seconds

I had quite a pair of fortunes (from the same cookie) a few weeks back:

• You are the master of every situation.
• Keep your plans secret for now.

The parties aren’t the only ones burying filings late on Friday night. Earlier (but not much earlier) this evening, Amazon filed a motion asking Judge Chin to reconsider yesterday’s ruling preliminarily approving the settlement.

It’s a full-on attack on the settlement; Amazon’s theory is that the future-claims issue is such a fundamental flaw in the settlement that there is no way Judge Chin could ultimately approve it. The briefing in support of this argument makes a familiar set of points from Amazon’s earlier brief about future claims, citing some familiar cases whose names I expect will become regular fixtures here at the Lab:

• Uhl v. Thoroughbred Technology and Telecommunications, Inc., 309 F.3d 978 (7th Cir. 2002)
• Wal-Mart Stores, Inc. v. Visa USA, Inc., 396 F.3d 96 (2d Cir. 2005)
• National Super Spuds, Inc. v. New York Mercantile Exchange, 660 F.2d 9 (2d Cir. 1981)
• Schwartz v. Dallas Cowboys Football Club, Ltd., 157 F. Supp. 2d 561 (E.D. Pa. 2001)
• UniSuper Ltd. v. News Corp., 898 A.2d 344 (Del. Ch. 2006)

Thus, Amazon argues, Judge Chin should save time and resources, reject this settlement, and give the parties another 30-45 days to negotiate a settlement that includes only releases relating to past claims. In arguing that this change wouldn’t seriously hurt the parties, Amazon uses their own past words against them:

The premise of both the initial proposed settlement and the Proposed Amended Settlement is that, in order to create a sufficiently valuable corpus of works for Google’s future projects, any settlement must provide Google with a license for the future use of class members’ works unless they come forward and ask not to be included. Yet, in defending the de factoexclusivity that such an arrangement would grant Google over large numbers of copyrighted works, the parties have repeatedly asserted that this feature of the proposed settlement is insignificant, because there are in fact very few literary orphan works. These two positions are logically inconsistent. If the corpus of orphan works is small, then modifying the settlement to require rightsholder consent for future uses—thus making orphan works ineligible for these projects—should be inconsequential.

Along the way, the memorandum raises the point that almost no one has actually yet seen the Third Amended Complaint; Amazon “obtained a non-conformed copy by contacting one of Plaintiffs’ attorneys shortly after the Court order permitting the plaintiffs to file it.” The motion doesn’t actually advert to anything interesting in the Third Amended Complaint, but I am looking forward to seeing it. (Black’s defines a “conformed copy” as “An exact copy of a document bearing written explanations of things that were not or could not be copied, such as a note on the document indicating that it was signed by a person whose signature appears on the original.”)

Additionally, Amazon asks that objectors be allowed to address not just changed provisions but anything in the settlement, to “avoid prejudicing objectors by unduly handicapping their ability to present arguments accurately and forcefully.” The Amazon proposal here is actually quite clever; it asks the court to allow objectors to state that their new objection supersedes their old one, so the court doesn’t have to read duplicative filings or stitch them together. (Of course, if Judge Chin and his clerks have already plowed through the mountain of paper, this proposal doesn’t actually save them much work.)

I expect to see a reply memorandum from the plaintiff class lawyers in fairly short order, arguing that the future claims issue is not problematic, and that in any event objections of this sort are improper at the preliminary approval stage and can be fully aired at the final approval stage in February. I suspect that Amazon had substantial pieces of this motion written and ready to go even before Judge Chin issued his preliminary approval order—and I suspect that the plaintiff class lawyers have substantial pieces of their reply written and ready to go, just in case anyone pulled a stunt like this. I also expect Judge Chin to move quickly in ruling, given that the currently scheduled start of the renotice period is only a few weeks away.

This summer, back-to-back, 7-11 featured the Transformers-themed Bumblebee Blast and the G.I. Joe-themed Liquid Artillery. Neither sounds like something one would ever want to drink voluntarily.

Very helpful cleanup of what “personal information” means on social network sites:

1. Service data. Service data is the data you need to give to a social networking site in order to use it. It might include your legal name, your age, and your credit card number.
2. Disclosed data. This is what you post on your own pages: blog entries, photographs, messages, comments, and so on.
3. Entrusted data. This is what you post on other people’s pages. It’s basically the same stuff as disclosed data, but the difference is that you don’t have control over the data — someone else does.
4. Incidental data. Incidental data is data the other people post about you. Again, it’s basically same same stuff as disclosed data, but the difference is that 1) you don’t have control over it, and 2) you didn’t create it in the first place.
5. Behavioral data. This is data that the site collects about your habits by recording what you do and who you do it with.

Panelists include Jonathan Band, Matthew Schruers, Ralph Oman (former Register of Copyrights!), and moderator Lisa Dunner.

Fred von Lohmann’s series on the amended settlement continues with a post on competition issues. On the unclaimed-works issue, he reaches a pragmatic conclusion—or as pragmatic a conclusion as one can reach “without a crystal ball”:

This raises a worthy question: if legislation is necessary to fix the competition problem posed by the settlement, then why do we need a class action settlement in the first place? Why not solve what seems like a quintessentially legislative problem with legislation, instead? (As Amazon points out, that’s exactly what was done when music publishers brought a class action against the first digital audio tape (DAT) recorders).

Here’s where realpolitik enters the equation. Google correctly points out that Congress has been working on orphan works legislation for years, to no avail. And none of the legislative proposals came close to the comprehensive solution embodied in the proposed settlement. So the question boils down to a political one: do you believe that approval of Settlement 2.0 will make orphan works legislation more likely, or less likely? Without a crystal ball, it’s hard to know.

On the pricing of the institutional subscription, he finds irony:

Of course, the chances of [monopolistic price gouging] coming to pass are hard to know in advance. As we have pointed out, if many large publishers pull their books out of the ISD database, then perhaps the ISD service won’t become indispensable to universities after all. So, ironically, the more successful the ISD proves to be, the more of a danger its pricing mechanism might prove to be for higher education.

I’m not sure I fully agree with Fred’s claim that these were the “two principal dangers” with the original settlement from a competition point of view. The orphan/unclaimed works issue was and is definitely #1, and the pricing of the institutional subscription is probably #2, but plenty of objectors also raised the coordinated pricing of individual books (call that a strong #3), along with a whole host of smaller issues like the MFN clause and the fixed 63/37 split. Leaving all of these issues out of the picture downplays the progress that the amended settlement makes in dealing with them. Perhaps, as Einer Elhauge argued, we shouldn’t have been worried about them at all before, but if you disagreed with him, then the (such as dropping the MFN and clarifying the language on algorithmic pricing) are a genuine improvement in the settlement.

Judge Chin has granted the parties’ request for preliminary approval. This isn’t a substantive ruling. just a procedural one. It puts us roughly back where we were a few months ago, before the parties withdrew the initial proposed settlement, and sets a schedule for re-notice, opt-outs, objections, and the fairness hearing. That schedule is the one the parties requested:

• Notice begins: Monday, December 14, 2009
• Opt-out/objection/amicus deadline: Thursday, January 28, 2010
• Plaintiffs move for final approval: Thursday, February 11, 2010
• Final fairness hearing: Thursday, February 18, 2010

Those who wish to appear at the fairness hearing must file a Notice of Intent to Appear by February 4, 2010.

Brian Lavoie and Lorcan Dempsey, Beyond 1923: Characteristics of Potentially In-copyright Print Books in Library Collections, D-Lib Magazine, November/December 2009, tries to give some tentative answers about the shape of the elephant. From the introduction:

The analysis that follows examines the characteristics of US-published print books, with an emphasis on books that are likely in copyright according to US copyright law. As with our earlier article, the analysis is based on data from the WorldCat database, which represents the aggregated collections of more than 70,000 libraries worldwide. The analysis focuses on three areas: the WorldCat aggregate collection of US-published print books; the subset of this collection published during or after 1923 - i.e., those potentially associated with copyright and/or orphan works issues; and the combined print book collection of three academic research library participants in Google Books - again, with an emphasis on materials that are potentially in copyright.

Lots of detailed tables on dates, authors, genres, audience level follow. From the conclusion:

This article characterizes the aggregate collection of US-published print books in WorldCat, with a special emphasis on materials published during or after 1923, and therefore either potentially or definitely in copyright. Findings from the analysis indicate that the collection of US-published print books in WorldCat is quite large, encompassing about 15.5 million print books. Nearly two-thirds of these - those published after 1963 - have a high likelihood of being in copyright; less than 15 percent - those published prior to 1923 - are almost certainly in the public domain, with the rest - those published between 1923 and 1963 - potentially in copyright if copyright was renewed. The post-1923 materials collectively account for more than 80 percent, or about 12.6 million, of the US-published print books in WorldCat. It is difficult to predict how many of these print books might be orphan works, but even a small fraction would, in terms of absolute numbers, be considerable, and require a substantial effort to investigate and clear copyright. One study, based on an examination of a random sample of books, estimates a cost of approximately $200 for each title for which digitization and access permissions were obtained.

(Via ResouceShelf.

Pamela Samuelson doesn’t like the amended settlement:

What stands out after my initial review of GBS 2.0 is that changes were overwhelmingly made to placate the governments of France and Germany, as well as the U.S. Department of Justice (DOJ).

Google is apparently hoping that if it can get these governments off its back, GBS 2.0 will be approved. Hundreds of authors, publishers and other interested parties raised dozens of objections to GBS 1.0, but their concerns were almost completely ignored. GBS 2.0, for example, does not address issues raised by academic authors about the risks of price gouging, lack of user privacy protections and restrictions on various uses that can be made of GBS books, even though most of the books in the GBS corpus are academic-authored books.

Kenneth Crews has some thoughts on the implications for libraries:

GBS 2.0 is a double whammy for libraries. First, the ISD’s scope is slashed. No longer “worldwide,” the settlement is now only about books registered with the U.S. Copyright Office (which will be dominantly U.S. books), and books originating from the United Kingdom, Canada, and Australia. Gone are all other books from Europe, Asia, Africa, South America, and other regions. Because the settlement is now tightly limited, so will be the ISD. The big and (probably) expensive database is no longer so exciting. Many of the books under GSB 2.0 are likely already available to many libraries.

The second whammy is legal. Because the settlement does not cover all books, liabilities surrounding some large portion of the books already shipped by libraries and scanned by Google are not released. Copyright owners from France, Argentina, New Zealand, and China retain the right to commence yet another lawsuit against Google, conceivably drawing libraries into the melee. Why the libraries? Rightsholders could claim that libraries are “contributory infringers” by making the books available. Moreover, many libraries and Hathi Trust, continue to hold book scans received from Google that are now outside the settlement.

A little birdie told me that Dan Clancy and Gary Reback would be on the McNeill-less NewsHour on Monday. The little birdie was wrong:

Reback was set to go on NewsHour to debate Google Books with the engineer that does most of the talking for Google - Dan Clancy. Both are quite able to defend their positions intelligently. Or at least, they would have. Clancy never showed up, leaving Reback at the studio, alone.

Why did Google back out? According to Reback, Google told the show that they didn’t want to put an engineer against a lawyer on TV: “They said I was a lawyer but Clancy was not, so the debate would be unfair.” Reback says that’s ridiculous. “Clancy goes around peddling his story and appearing on panels with lawyers all the time,” he said (which is true). Reback also notes that “Google has hundreds of lawyers, dozens of whom have worked on this. Surely, they could find a lawyer to debate if they were afraid to put Clancy up” (which is also true).

For their part, Google says they aren’t interested in debating the legal niceties of Google Books on broadcast television. Gabriel Stricker, head of search communications at Google, says that they were told Harvard profession Robert Darnton would be their opponent on the debate (his thoughts on Google Books are here, and wow he desperately needs an editor), and that Reback was added at the last minute. When they found out about the change, Stricker says, it was too late to find an appropriate Google attorney as a replacement.

Stricker says Google wants to have a “philosophical and ideological conversation about Google Books” in front of a mainstream TV audience, not a legal debate. And he adds “the fact that Gary finds it necessary to try to create conflict surrounding the existence of the conversation is exactly why we would prefer not to have a conversation with him.”

Fred von Lohmann of the EFF has been blogging about the amended settlement. After short posts on the lack of privacy changes and the contours of the debate, he has a longer discussion of the settlement’s access implications:

Taken together, these features mean that the Google Books project could potentially provide Americans (and only Americans, as the settlement only authorizes Google to offer Display Uses of in-copyright books to U.S. Internet users) with unprecedented instant access to a large collection of books that previously were available only in research university libraries. In particular, like the Internet before it, Google Books could make specialized resources available to people who otherwise might never be able to access them (see, e.g., Google’s agreements to digitize U. of Wisconsin’s Native American collection and U. of Texas’ Benson Latin American collection). …

First, under the settlement copyright owners can pull their books (see Section 3.5, “Right to Remove or Exclude”) out of all the products and services envisioned by the settlement, including full-text search and limited “snippet view” access. This is essentially the “take the money and run” option—the copyright owner collects a per-book payment from Google for books already scanned, but then the public gets no online access to these books unless and until the copyright owners negotiate new deals with Google or other online providers. This effectively gives copyright owners a unilateral right to trump fair use, essentially “unpublishing” their books online. Some observers expect that most major publishers will opt to “take the money and run” for both their in-print and out-of-print titles, leaving gaping holes on the virtual shelves of Google Books. If this takes place, then the settlement would only foster access to orphan and unclaimed works. Still good, but far short of full access to every book in the University of Michigan library.

At times, I’ve wondered why they hadn’t already done this. I guess they were waiting to get it right. My initial impressions are very positive—this is a great resource, and I look forward to its expansion to cover other kinds of legal materials.

It looks like the Chinese are still taking potshots at the old version of the settlement:

The company has offered a compensation settlement of $60 per book to authors, as well as 63 percent of the revenue from online reading. To reject the package and Google’s right to scan their works, however, writers must appeal before Jan 5 next year.

Zhang, who met with executives from Google on Nov 2, blasted the offer as unacceptable.

“Google violated Chinese writers’ copyright. It is ridiculous for a business in the US to set a deadline for Chinese writers to protect their interests. Also, the company should clearly admit to its infringements and negotiate with Chinese authors sincerely,” he said.

There is much more at the link, but the article and some of the authors quoted in it seem to be unaware of the existence of the amended settlement, which would take most Chinese authors out of the class entirely. Does news of such thing still travel by clipper ship?

You may recall the Open Book Alliance’s widely quoted initial reaction to the amended settlement:

Today, Google, the Authors Guild, and the Association of American Publishers released their revised book settlement proposal in an attempt to fix the deeply flawed legal agreement. Open Book Alliance co-chair Peter Brantley said, “Our initial review of the new proposal tells us that Google and its partners are performing a sleight of hand; fundamentally, this settlement remains a set-piece designed to serve the private commercial interests of Google and its partners. None of the proposed changes appear to address the fundamental flaws illuminated by the Department of Justice and other critics that impact public interest. By performing surgical nip and tuck, Google, the AAP, and the AG are attempting to distract people from their continued efforts to establish a monopoly over digital content access and distribution; usurp Congress’s role in setting copyright policy; lock writers into their unsought registry, stripping them of their individual contract rights; put library budgets and patron privacy at risk; and establish a dangerous precedent by abusing the class action process.”

The digitization of books has the potential to unlock huge volumes of our shared cultural knowledge, and the Open Book Alliance supports efforts to make books searchable, readable, and downloadable. But there is a right way and a wrong way to accomplish this goal. The right path embraces openness, competition, and the public good. Last week, the Open Book Alliance issued a set of requirements that the new settlement proposal must adhere to in order be true to these principles. Most critically, the settlement proposal must not grant Google an exclusive set of rights (de facto or otherwise) or result in any one entity gaining control over access to and distribution of the world’s largest digital database of books. It is clear that Google has failed to meet these requirements.

Is it just me, or was this statement drafted without the benefit of actually seeing the amended settlement? It sports a telling lack of detail. I’m reminded of Dave Barry’s book report on “The Old Man and the Sea”:

‘The Old Man and the Sea’ is a short novel weighing less than two pounds written by the author Ernest Hemingway. It concerns an old man who becomes involved with the sea (or, as it is sometimes called, ‘the ocean’). As the book (‘The Old Man and the Sea’) unfolds, the author, Ernest Hemingway, writes about these two major themes - (1) the old man, and (2) the sea - and the things that happen to both the main character, which is the old man, and a major body of water, played by the sea, as viewed by the author, Ernest Hemingway, and as we reach the 106-word mark in this book report we can see that …

Oh, snap. Steven Pinker just got served by Malcolm Gladwell.

The Wall Street Journal has a story today with some very interesting language on the DOJ’s reaction to the amended settlement:

People familiar with the matter say the Justice Department remains concerned that the fact the settlement gives Google immunity from lawsuits related to orphan works may be anticompetitive. The department is expected to file its reaction to the modified agreement by early next year.

A spokeswoman for the Justice Department said the department is reviewing the revised agreement and its investigation into the settlement is “ongoing.”

(Dear Mr. Murdoch: If you aren’t going to put your stories online in a way that lets me link to them for my readers, I’m not going to link them at all.)

By the way, I’m now using “GBS” to “Google Books settlement” rather than to “Google Book Search.” The settlement is really about the books, rather than about search. Fortunately, the acronym can remain the same.

Randy Picker moves fast. It’s only Monday, but he already has up a twelve-page paper on the amended settlement, Assessing Competition Issues in the Amended Google Book Search Settlement. His focus is on antitrust, and there’s a lot of good stuff in there.

For one thing, sports a very elegant analysis of the new waiver of antitrust immunity on the timing of various parties’ decisions. If the DOJ can wait until pricing problems develop before needing to challenge them in court, some of the concerns can be kicked down the road, and addressed if and when they do develop. Since waiting creates no immunity, we might as well wait to see whether the Consumer Purchase program does a faceplant.

I also found this passage on the algorithm persuasive (i.e. he agrees with me):

There still is this characterization question: should we think of this arrangement as a horizontal arrangement or as a vertical arrangement much more like iTunes? The fact that the authors are acting in unison with Google is what pushes this towards horizontal characterization and nominal antitrust doctrine—think Socony-Vacuum—is quite unfriendly to any horizontal agreement on prices even, perhaps, one that says we are going to agree on competitive prices. That is almost certainly to allow form to trump reality if the ASA actually operates in the manner that it sets out.

And on the key question—the new Unclaimed Works Fiduciary—I loved this paragraph:

Now I feel like Moses: I can see the promised land but apparently I can’t get there. Creating the UWF is a nice way to solve the conflicts problem identified by DOJ. DOJ had expressed a concern that holders of unclaimed works didn’t necessarily have the same interests as those of active rightsholders. The UWF mechanism enables separate representation of those interests. But the settling parties have limited the role of the UWF to merely stepping into the shoes of the registry in some circumstances. They could have broadened the role for the UWF to have the UWF step into the shoes of the rightsholders of unclaimed books instead. Had that been the focus, the UWF would then be an elegant solution to the going forward problem of how to license the orphan works.

This matches my understanding of the UWF’s limited role (about which I will blog in more detail when I get a chance). The new, limited-purpose UWF doesn’t really address the fundamental antitrust issue with the settlement: Google’s use of the class action to obtain exclusive access to unclaimed works. There are also serious issues with the “class action” part of that phrase, so I’m not entirely surprised that the parties didn’t give the UWF those broader powers. But Picker’s analysis makes me wonder whether the amended settlement was drafted with that possibility in mind, only to have one or another of the parties get cold feet and insist on restricting the UWF’s powers.

I believe that UWF is short for “Unpronounceable Word Fiduciary.”

I managed to get my hands on the remaining articles from the special GCP issue on the settlement.

First, we have Timothy Brennan’s The Proposed Google Book Settlement: Assessing Exclusionary Effects. Much of it has already been mooted by the rapid pace of events. For example, Brennan finds the DOJ’s position on Google’s ability to discount puzzling in light of the Supreme Court’s Leegin decision, which held that prohibitions on retailer discounting aren’t per se illegal. Interesting, but now irrelevant—the amended settlement gives Google power to offer any discounts it wants to (on its own dime, of course). The same goes for the MFN, which Brennan has some issues with—the MFN is gone, too.

After some mildly hand-wavy discussion of market foreclosure, Brennan closes with the following passage on orphan works:

The central concern of the commentaries is the settlement’s provisions regarding orphan works. One can fairly wonder how a settlement can cover a class whose members by definition cannot be identified. Leaving that aside, the case overall presents a puzzle about how one should treat property that is probably owned but where the owner cannot be found. The default view on this is not clear; one can think of this as equivalent to estray, e.g., defining obligations to find the owner of a lost wallet before the finder can legally keep the money.

The record in the Google Books case establishes that this is a significant issue. Even Hausman and Sidak, who are skeptical of arguments based on suppression of competition from orphan books, estimate that 9 percent of the books Google has copied are orphan books. One wonders if some doctrine in copyright akin to adverse possession in property law should apply. A copyrighted work could enter the public domain after it has been openly distributed for some amount of time. The distributor’s liability could be limited if it has undertaken due effort to find the copyright holder. Even if such a legal regime is a good idea, a separate issue is whether it should be the result of common law evolution, statutory intervention, or, as in this case, a settlement between Google and holders of non‐orphaned books.

This passage might have benefitted from citations to the state briefs on the application of unclaimed-property law, and to the literature on orphan works legislation, which would have imposed diligent-search requirement in exchange for a limitation of liability.

Next, we have Isabel Davies et al., Online Distribution of Copyright Works: Google Books in a Broader European Policy Context. This one is a readable summary of the European debates over the last few years. It closes with the sentiment:

The opposition to the Google Books issue and the time that it is taking to reach a settlement acceptable across the board in the United States creates the opportunity for Europe to take the lead and forge its own path with Europeana, perhaps producing a solution that will act as a blue print for digitization legislation on a global basis. The question remains whether Europe is prepared to step up and take the lead.

Now that Europe is basically out of the settlement, the issue is in theory easier to address, as there won’t be the American lawsuit complicating matters. But I’m skeptical that Euroope’s political apparatus will do much on the issue without the brash Americans pushing on it with their infernal lawsuit.

Finally, there’s Ian Forrester’s Google: The Benign Monopolist?. This one is mostly an overview of the settlement from the perspective of copyright policy. I imagine that the antitrust community will find it an interesting discussion of those issues, but for those already stepped in settlement lore, nothing in it will be new or surprising.

The EFF and ACLU of Northern California are unimpressed.

In a provocative blog post, Jane Litte analogizes the new Unclaimed Works Fiduciary to Blackwater, the mercenary-slash-goverment contractor. (Technically, I guess the UWF would be Xe, post-name change.) Her point is that the Registry will be performing a fundamentally governmental function but in a privatized way. If so, is the settlement akin to an act of privatized eminent domain?

I think this is a fair question about the settlement, although primarily in a metaphorical sense. The nonexclusive nature of the licenses, I think, precludes a real takings or public use argument. But it’s an interesting metaphor; I’ve played around with “adverse possession” (or perhaps “prescriptive easement”) and “eminent domain” as ways to characterize the settlement. Both of these ideas get at the way in which the settlement represents a deployment of the state’s legal authority to reallocate underused property, with all the troubling implications that raises. One way of putting the ultimate question is whether this deployment has been hemmed in with sufficient safeguards, substantive and procedural, to make this of state power legitimate.

While we’re trading in metaphors, I’d also like to suggest that the move from Registry to UWF for dealing with the orphans is fertile ground. In a sense, the UWF is a true court-appointed guardian. It’s going to be (assuming there’s no funny business in the Registry’s charter) responsible for safeguarding their interests. “Fiduciary” means that if they later show up, they can sue it if it disregarded its duties to them. This is a real improvement over Settlement 1.0, which I criticized for not containing “internal” procedures to make sure the orphans are properly protected if claiming copyright owners opt out of the various Google programs. The UWF will be responsible for pulling the unclaimed works out, too, if the deal is no good. That’s a substantial improvement.

At the same time, the need for a UWF underscores just how remarkable this class action is. What other class action is so dangerous to class members that they need an ongoing guardian, able to act on their behalf without instructions, just to safeguard their interests? It’s precisely because the settlement trades in future claims based in future conduct that these possibilities for abuse arise and must be guarded against. But stare closely enough at that fact, and it leads, I think, to asking difficulty questions about the propriety of the class action itself.

Side thought: I initially read the title of Jane Little’s post as asking whether the settlement would create a “backwater” for books. I hope not, but what a metaphor!

UPDATE 11:00 PM: I reread the amended settlement more closely, and I don’t believe that it empowers the UWF to pull unclaimed books out of Display Uses.

Gillian Spraggs has the scoop. The parties seem to have used the opportunity to expand the diversity of the named plaintiffs in ways that go beyond just nationality. For example, Daniel Baum was a law professor for twenty-six years.

• 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).

I note that the DOJ asked for “a week to ten days” after the objection deadline to formulate its response. The plaintiffs propose to give it a week.

My students and I are at work on a summary document that will taxonomize the various objections to Settlement 1.0 and then describe, in brief, which ones of them Settlement 2.0 tries to address, and how. We’re also trying to figure out how best to put Settlement 2.0 on the Public Index to foster discussion. Plus we have a few more surprises up our sleeves that we’re pretty jazzed about. We should have some exciting stuff to announce in over the next couple of weeks.

The amended settlement dropped in the eleventh hour. Here’s a redline version and here’s the related memorandum with the procedural details. The best news coverage is the New York Times story; the best blog coverage is Danny Sullivan’s. I’ve just completed a first pass through the amended settlement, tweeting all the way.

The biggest change is that the class of affected books has been dramatically reduced, although not at all in the way I was expecting. Books registered in the U.S. are still in, as are books published in Canada, the U.K., and Australia. That change should largely take foreign non-Anglophone books out of the settlement entirely. There are a number of other tweaks to the definitions to deal with specific classes of works (e.g. comic books).

On the antitrust front, there are changes to address two of the specific concerns raised by the DOJ. The 63/37 split (and almost everything else) will be negotiable, and Google will have more extensive rights to discount books provided it’s willing to eat the lower margins. These were unnecessary provisions in Settlement 1.0; Google and the parties aren’t giving much up by agreeing to these changes.

A number of other concrete issues have been fixed here and there. Unclaimed funds are no longer diverted to the Registry or other copyright owners. Google won’t share personal information with the Registry except via legal process. Creative Commons licensing options are explicitly provided for. There will be an Unclaimed Works Fiduciary to act on behalf of the unclaimed (including orphan) copyright owners at the Registry. (It’s called a “Fiduciary,” but the scope of its duties isn’t actually specified. I guess we need to see the Registry charter for that.) The secret termination clause is now “Intentionally Omitted.” Many other disputed issues from round 1, such as other privacy risk and less favorable treatment of Inserts, remain unchanged.

The heart of the settlement’s promise, peril, and problems has always been its treatment of unclaimed works—a category that contains the orphan works. Settlement 1.0 allowed Google to use and sell them on an opt-out basis, and Settlement 2.0 does the same. That gave Google exclusive access to a market segment that no one else can enter, and thus raised antitrust concerns; the DOJ hit this barriers-to-entry point hard in its Statement of Interest.

Settlement 2.0 has a very interesting response. The Unclaimed Works Fiduciary (or “UWF”) is given extensive powers to set terms on their behalf. Most of this, while good for protecting the interests of orphan owners, is irrelevant to the antitrust issue. It doesn’t matter whether the UWF can object to the pricing bins if no one besides Google is able to sell these books at any price. But there’s one, very important, very enigmatic exception, spelled out in section 6.2(b)(i):

General. The Registry will be organized on a basis that allows the Registry, among other things, to (i) represent the interests of Rightsholders in connection with this Amended Settlement Agreement, (ii) respond in a timely manner to requests by Google, Fully Participating Libraries and Cooperating Libraries, and (iii) to the extent permitted by law, license Rightsholders’ U.S. copyrights to third parties (in the case of unclaimed Books and Inserts, the Unclaimed Works Fiduciary may license to third parties the Copyright Interests of Rightsholders of unclaimed Books and Inserts to the extent permitted by law).

When I worked through Settlement 1.0, I concluded that “to the extent permitted by law” didn’t by its own terms authorize the Registry to license third parties (such as competitors to Google). If such licensing would otherwise be legal, it empowers the Registry to act, as a matter of its charter. But the settlement is quite careful to indicate that all the relevant licenses flow directly to Google, and the Registry holds no rights itself, so it has nothing to license in a way that would be “permitted by law” except where the copyright owners have voluntarily given it such powers, which can’t happen for unclaimed works.

By carrying over the same language for the UWF, Settlement 2.0 confirms that Google will have the only game in town for the unclaimed works. Nothing in the amended version gives the UWF any new powers for such purposes; the other clauses that refer to it give it power to negotiate specific terms with the Registry and Google under the various programs the settlement specifically allows Google to set up. The DOJ all but invited Google and the plaintiffs to empower the Registry to license Google’s competitors; they declined that all-but-invitation. They’re going to try to tough this one out; the DOJ will have to decide whether to back down or to fight, as this amended settlement doesn’t give it one of the central changes it asked for.

Now, it’s true that “to the extent permitted by law” is an obscure phrase in this context. My best working hypothesis is that the parties used to to take account of possible future changes to the law. Thus, the New York Times story says, “The trustee, with Congressional approval, can grant licenses to other companies who also want to sell these books… .” And Danny Sullivan, listening to the conference call, wrote:

Richart [JG: Sarnoff, presumably]: we are certianly hopeing that this settlement is the key that unlocks a positive outcome on the legistlative process on orphanworks as nw there’s a way to actually implement any legistlation that congress decides on orhpahn work

As I understand this, their idea is that the Registry, via the UWF, will be in a position for Congress to rely on in any orphan works legislation. All it would take is a wave of Congress’s magic wand, and the Registry will be handing out licenses to Google competitors. Thus, apparently goes the argument, we’re helping facilitate a Congressional orphan works solution. “To the extent permitted by law” leaves the door open for future action of this sort.

It’s a very clever hack. I have my doubts whether it’s legal.

Suppose that Congress doesn’t act. In that case, Google will be the only distributor able to sell to these orphan works. The settlement gives it preferential access to this market, while leaving an absolute legal barrier to entry in the way of all competitors. The speculative possibility that Congress might someday act to open up the market to others doesn’t create actual competition now. “To the extent permitted by law” changes nothing. So if the settlement would fail on antitrust grounds, this clause can’t save it.

Suppose, on the other hand, that Congress does act, blesses the Registry, and creates a statutory licensing system operated by the UWF. That would be a reasonable outcome. It would create legitimate, competitive, compensated access to orphan books. But if the whole thing depends on Congress, why do we need the class action?

I’m sure there will be much more to say about the amended settlement in the days, weeks, and months to come. My instant reaction is that it makes a number of meaningful, if modest, improvements, but leaves unaddressed the central issue that led me to worry about the settlement in the first place.

Via Metafilter, your daily dose of pure joy.

Looks like someone took me up on my invitation.

See also: bumper stickers.

The parties have asked the judge to delay their filing deadline until Friday. Yes, until Friday the 13th. The New York Times reports:

But on Monday, the parties submitted a letter to the court requesting an extension to Nov. 13. In the letter, the group indicated that it had met with the Justice Department before and after the October status hearing and had met as recently as Friday, Nov. 6.

I am eager to see this letter.

UPDATE 3:40 PM: Ask and ye shall receive: Judge Chin’s order granting the request.

Regular French Google-blogger and -tweeter has an important blog post (Bing translation) on a very important decision by the Tribunal de Grande Instance de Paris on choice of law in a copyright lawsuit against Google. The case involved a Perfect 10-style scenario: unauthorized parties posted a copyrighted image that Google then automatically indexed. In a decision that appears to turn in part on the role of Google France and in part on the Frenchness of the author and the work, the court held that the appropriate copyright law to apply would be French, not American. As Callimaq notes, this case has substantial implications for the Google Book Search case.

Rituparna Bhuyan, Centre Protests ‘Copyright Violation’ by Google Books, Financial Express (New Delhi), Nov. 9, 2009:

Web portal Google Books’ initiative to create a digital library by scanning printed publications has triggered alarm bells in India, forcing the Centre to take up the matter with the US government. In a meeting held in the last week of October here, senior Indian officials told their US counterparts that the portal would encroach upon the copyrights of Indian authors and publishers. …

India’s point of view on Google’s initiative was conveyed by commerce and industry ministry to a team of officials lead by United States Trade Representative, Ron Kirk. “The US side took on board India’s concerns,” a official privy to the development told FE on condition of anonymity.

I believe this is some of the “diplomatic stress” that Marybeth Peters was referring to.

Kenneth Crews tweets:

Google books: The parties have asked the court to extend the date to Nov 13 for the amended settlement. #GBS #copyright

The revised settlement is due to drop today. I’m all a-twitter, and I’ll be all over Twitter as soon as it hits.

Peter Hirtle has been busy. He, along with Emily Hudson and Andrew Kenyon, also has out a new book, Copyright and Cultural Institutions: Guidelines for Digitization for U.S. Libraries, Archives, and Museums. It’s not directly about the Google project or lawsuit, but it does provide a lot of context for thinking about the digitization opportunities and challenges faced by libraries. The bulk of the book is a copyright primer, including a very clear chapter on the remarkably intricate section 108 privileges. The most interesting sections are towards the end, when the book walks through a risk management approach to digitization, and then provides two case studies: oral histories and dissertation.

In a particularly nice touch, the book is available for purchase in physical form (either through CreateSpace or Amazon) or for free download. The goal, Peter informs me, is to recover the production costs. The layout and design of the book are quite elegant, so buy one and give it to your favorite archivist.

Douglas Baird’s chapter about INS v. AP in Intellectual Property Stories contains the following remarkable example of the compression techniques people resorted to when every letter was costly:

t scetus tdy dodd 5 pw f potus dz n xtd to t pips, ogt all pst cgsl xgn q sj is uxl.

That’s a “typical AP cable” transmitting the results of a Supreme Court case. Look at all familiar?

The full translation back into English is inside.

(Continued . . .)

His point is that there is so much room separating the Marybeth Peterses and Brewster Kahles of the world that no “legislative solution” could actually satisfy all the people calling for one:

In summary, if we can read between the lines of her testimony, Peters’s preferred legislative solution, would bend over backwards to protect the interest of rights holders. It would do nothing to lower the transaction costs. Google could, if it wanted, try to convince Congress (and the Copyright Office) that “a solution that is more like a compulsory license may make sense for those engaged in mass scanning,” but the Office’s traditional opposition to compulsory licenses stands. I don’t see anything in Peters’s testimony that suggests a likely legislative solution that would solve the four problems facing mass digitization. …

Kahle’s vision is one where “out-of-print” equals “orphan,” and anyone is able to make non-commercial use of the work unless an owner comes forward. It is an inspiring vision, but to suggest that this is “roughly how Orphan Works legislation” works is disingenuousness worthy of Sergey Brin himself. Under the proposed legislation, orphan and out-of-print books could be used by anyone for any purpose - but only after a reasonable search for a copyright owner had been conducted. Out-of-print but non-orphan books could not be used, even if the purpose was non-commercial. No one other than Kahle has dreamed of a system where failure to maintain a work in print would limit one’s copyright monopoly.

He concludes by arguing that the settlement, while flawed, is good enough that we should embrace it, rather than waiting for Congress to square the circle of satisfying everyone calling for Congress to act.

He writes:

The most desirable change from an antitrust perspective would be to allow Google’s competitors to exploit orphan works on the same terms as Google. The problem with this solution is that it further strains the boundaries of class action law and looks more and more like private legislation. This should not, in my view, be enough to derail the deal if the parties can show that all of the relevant sub-class interests were adequately represented.

The Author-Publisher Procedures enhance the coordinating function of the Settlement by streamlining the incorporation of existing author-publisher contractual terms into the framework of the Google Book universe. However, where an existing author-publisher contract gives both parties some control over electronic exploitation, or simply fails to make any provision for electronic rights, the Author-Publisher Procedures effectively overwrite those contracts. These new terms do not appear to systematically disadvantage either authors or publishers, but they strike me as a one size fits all solution that could be substantially improved upon.

Finally, I expect the revenue sharing aspects of the deal to become more complicated.

It’s hard to argue with that last one.

Just a reminder to all the academics in the audience that I’m interested in what you’re doing in relation to the settlement. If you publish something or post it online, let me know so I can blog about it. If you’d like feedback at any stage of the writing process—from initial idea to near-final draft—I’m happy to give comments on anything and everything you send my way.

In advance of Monday’s forthcoming settlement, the Open Book Alliance has issued a list of what it calls “baseline requirements” for Settlement 2.0:

In advance of Monday’s forthcoming settlement, the Open Book Alliance has issued a list of what it calls “baseline requirements” for Settlement 2.0:

• The settlement must not grant Google an exclusive set of rights (de facto or otherwise) or result in any one entity gaining control over access to and distribution of the world’s largest digital database of books.
• Authors and other rights holders must retain meaningful rights and the ability to determine the use of their works that have been scanned by Google.
• The settlement must result in the creation of a true digital library that grants all researchers and users, commercial and non-commercial, full access that guarantees the ability to innovate on the knowledge it contains.
• All class members must be treated equitably.
• The settlement cannot provide for competition by making others engage in future litigation.
• Congress must retain the exclusive authority granted by the U.S. Constitution to set copyright policy.
• All rights holders impacted by the settlement must have a meaningful ability to receive notice, understand its terms and opt-out.
• The parties that negotiated the settlement must live under the terms to which they seek to bind others, rather than their own separately negotiated arrangements.

The OBA’s rhetoric is always fun to watch. It has a wonderfully authoritative tone: this is how it’s going to be, and if the parties don’t think so, then, well, they’ve got another think coming. Having now seen Gary Reback in action at D is for Digitize, I can see where they get the style.

In an unsurprising turn of events this week, Judge Chin denied the visual artists’ motion to intervene. I say “unsurprising” because this is the second time Judge Chin has denied their request. Their first request was denied summarily, so the motion to intervene was actually before Judge Chin as a request for reconsideration. And no, he hasn’t changed his mind.

This time around, he explained his reasoning in somewhat more detail, though. First up is his analysis of why intervention would be untimely:

The parties to this litigation have been negotiating and renegotiating a complicated settlement agreement for several years. In the last two months, the parties have begun working with the Department of Justice to revise the agreement to attempt to avoid antitrust problems, and they are due to re-file their new settlement proposal on November 9, 2009. The proposed settlement, if approved, would allocate rights to millions of textual materials. It represents thousands of hours discussion, compromise, and legal draftmanship [JG: sic]. Yet, the movants propose to intrude at the very end of this long process, and to add the question of millions of pictoral [JG: sic] materials to the equation. Intervention could “destroy [the parties’] settlement and send[] them back to the drawing board.” (citation omitted).

On one level, this has to be right. As I discussed in my analysis of agenda-setting in this class action, giving someone the power to propose a specific agreement is all but inherent in the way we handle class actions. If the ASMP feels that the settlement will cause them harm, the objection and amicus window is designed to give them an opportunity to be heard. Many of their arguments are properly characterized as objections that the settlement violates the public interest and should thus be rejected. Moreover, since the settlement wouldn’t bind the visual artists, the court correctly and repeatedly notes that they’re free to file their own lawsuit against Google.

In terms of timing, the court does talk about the years of negotiation and the parties’ interest in not being prejudiced by having this monkey wrench thrown into it. But when the chips are down, I read the opinion as saying that the visual artists should have moved to intervene when the initial settlement was announced:

As long as movants retain their right to bring their own lawsuit, they have not been harmed by this litigation, and they do not retain an automatic right to intervene a year after the proposed settlement agreement was filed. From the perspectives of fairness and efficiency, it makes more sense for the movants to file their own lawsuit than to be permitted to delay this lawsuit.

The other issue the court needs to deal with is the visual artists’ argument that they were entitled to be part of the final class, given that they were included in the class in amended (2006) complaint. As the court wrote:

The movants are incorrect: the broad class that was defined in the first amended complaint does not entitle them to intervene at this late date. “The braod language of a complaint does not vest in putative class members a right to be part of the class ultimately certified by the District Court.” Holocaust Litigation, 225 F.3d. at 202. The center of this litigation has always been copyrighted textual materials. I reject the argument that class counsel acted in bad faith in the way they defined the class because there are legitimate reasons for limiting the class to holders of textual copyrights. …

The holders of pictoral [JG: sic] copyrights are differently situated from the holders of textual copyrights. Frankly, in the context of an online database that is searchable using keywords, it makes sense to prioritize the rights to word-based material.

The Holocaust Litigation case cited by the court is reasonably on point. This was the lawsuit against Swiss banks for their complicity in the Holocaust. The original complaints were broadly drafted, but the $1.25 billion settlement covered only a class of those “persecuted or targeted for persecution by the Nazi Regime because they were or were believed to be Jewish, Romani, Jehovah’s Witness, homosexual, or physically or mentally disabled or handicapped.” A group of ethic Poles moved to intervene, but the Second Circuit denied their request. As it explained, “the evaluation of such a motion [to intervene] does not require a court to take into account the language of the original complaint.” And in a footnote, the court addressed the Poles’ argument that there was no coherent reason to include, e.g. Romani, and not also them:

Appellants draw attention to the Settlement’s inclusion of groups whose losses apparently do not form part of this estimate-that is, those who fall under the categories of Romani, Jehovah’s Witnesses, homosexuals, and the disabled-and suggest that it implies suspect motivations. See Appellants’ Brief at 25 (“[C]lass plaintiffs took the more expedient route of dropping some of the more numerous classes from the consolidated complaints in defining the settlement classes.” (footnote omitted)). The inclusion of these groups arguably calls into question appellees’ argument that ethnic Poles have been excluded because the Settlement amount does not reflect their losses. Nevertheless, we are not persuaded that appellants have been inadequately represented. Counsel for appellees explained at oral argument that the inclusion of these smaller groups represents a decision on the part of plaintiffs to include those groups treated “essentially indistinguishably from Jews” under the Nuremberg race laws. Putting aside the question of the parties’ motivations, we note that the groups in question presumably encompass far fewer claimants than the category of ethnic Poles (and certainly fewer than the broader category of persons persecuted on the basis of national origin, who would undoubtedly seek to be included if ethnic Poles succeeded in intervening). Under Rule 24 the inclusion of much smaller groups could well be warranted on the grounds that their ability to pursue their interest in the subject matter of this litigation would otherwise be impaired. Moreover, it would be up to class members-not potential intervenors-to challenge the Settlement’s allocation of funds. In any event, we discern no basis for questioning the motives of the class representatives or their counsel.

While the ASMP et al. are considering their options, with that precedent on the books, I don’t think they have much chance of success on appeal. I’ve seen suggestions that they should do so and then move for consolidation; I would think that, after this ruling, the last thing they would want would be to back in front of Judge Chin.

Those inclined to pore through Judge Chin’s language for tells as to his thinking on the settlement itself are encouraged to do so in the comments.

As I suspected, the crux of the case is whether French or U.S. law applies to the scans being made in the U.S. Also worth reading for the following quote:

“It is an infernal machine, it never stops. It is a disgrace. It is cultural rape,” huffed Serge Eyrolles, to the media, in the best tradition of French aesthetes, at a press conference in September.

Two online magazines have posted symposia on the Google Book Search settlement. There’s original writing from some of the top scholars working in the area. Unfortunately, both of these magazines feature gated access! Big frowny face.

First, the Economists’ Voice has:

• Paul Courant’s The Stakes in the Google Book Search Settlement
• Pamela Samuelson’s reply, The Google Book Settlement: Real Magic or a Trick

It’s just like reliving the K is for Keynote panel from D is for Digitize, only in highly articulate written form. Courant approaches the problem as the economist he is, moving quickly through background and summary, then focusing on the monopoly issues around the orphan works. He ends with an endorsement of making the settlement nonexclusive, but not at the cost of scuttling the whole thing:

Thus a revised settlement (as suggested by the U.S. Department of Justice, p. 25) that provided competitors with the ability to use the orphan works on the same terms as Google, or legislation with similar consequence, would be an unambiguous improvement over the original settlement. The great benefit of a new market in electronic versions of the literature of most of the twentieth century could be realized with reduced risk of monopoly and without enriching unrelated parties with the fruits of orphans’ labor. In contrast, scuttling of the settlement or greatly limiting the volume of works to be covered would put us back to where we started—the only people with good access to the scholarly and cultural record of the twentieth century would be people with physical access to research libraries, and even for them, that literature would be more difficult to access than most works published before or since.

Samuelson’s reply is more far-ranging, touching on pricing, privacy, reader rights, class-action abuses, and orphan works. She’s less optimistic that the settlement can be patched in a way that addresses all of these concerns—particularly given that many of these “fixes” would stretch the class action procedure even further:

Even if the amended settlement accommodated core objections, I question whether the class action settlement process can be used to achieve such a massive restructuring of the market for digital books as the GBS deal would bring about. Typically such settlements resolve only the specific dispute between the parties after the judge has assessed the merits of the lawsuit and determined that the class representatives and their lawyers adequately rep- resented the interests of the class as a whole. The broader the settlement’s scope, the greater the size of the class, the more forward-looking are its terms, and the more the agreement releases the defendant from liability for future conduct, especially conduct different in kind from the issue in litigation, the less likely it is that a judge will or should approve it. The GBS deal is troublesome on all of these grounds, so the parties should be seeking a legislative, not a judicial, blessing for it.

Meanwhile, GCP: The Online Magazine for Global Competition Policy, has a five-article symposium issue:

• Timothy Brennan, The Proposed Google Book Settlement: Assessing Exclusionary Effects
• Einer Elhauge, Framing the Antitrust Issues in the Google Books Settlement
• Randy Picker, Antitrust and Innovation: Framing Baselines in the Google Book Search Settlement
• Isabel Davies et al, Online Distribution of Copyright Works: Google Books in a Broader European Policy Context
• Ian Forrester, Google: The Benign Monopolist?

So far, I’ve only been able to get my hands on the Elhauge (via our librarians’ bag of tricks) and the Picker (via SSRN). The Elhauge is a nice 8-page version of his (pro-settlement) analysis, in case you found the 60-page version or the 32-page version too much to take in. He repeats his point that the output of orphan works will expand:

Thus, the settlement would clearly increase the output of orphan books from zero to something, and lower the price from infinity to something lower, even under the worst case assumptions that no rivals will ever enter, no rightsholders of orphan books will be found, and the algorithm price will be supracompetitive. The effects on orphan books are thus entirely beneficial to consumer welfare even under worst case assumptions. In fact, the benefits to consumer welfare are even greater because there is some chance rivals could enter this market, it is implausible no rightsholders would be found, and the algorithm price is (we shall see) designed to mimic competitive prices. Further, with the settlement, rightsholders in orphan books will get some royalties, and thus will also be better off than without the settlement.

Of course, the United States’s brief largely rejected this view, so Elhauge closes with a reply to it:

To support the assertion that the settlement forecloses Google rivals, the DOJ brief cites almost exclusively to a series of barriers to rival entry that were not created by the settlement and would apply equally without any settlement. The only exception is that the DOJ mentions the settlement’s most favored nation clause, but its brief never explains how having the same costs as Google for unclaimed books could discourage a rival from entering a market that (by hypothesis) would have supracompetitive prices. Nor does the brief explain how equal costs for unclaimed books could matter if the brief were right that rivals could not offer unclaimed books at all given other entry barriers that the settlement didn’t create.

He finishes with the hope that the Department of Justice’s “tentative” views will be rethought as the revised settlement is presented. He cites the fact that the brief was signed by Deputy Assistant Attorney General William Cavanaugh, rather than his boss, Assistant Attorney General Christine Varney. I somehow doubt that a filing of this significance got out the door without her acquiescence—and besides, it’s featured prominently on the Antitrust Division’s home page, which presumably she checks now and then.

The Picker piece, on the other hand, is entirely new work, and it’s very good. His central argument is that the output-expanding aspects of the settlement can’t be allowed to provide a blanket immunity from antitrust review of the anti-competitive aspects. Once again, he’s managed to articulate something I was thinking but had never been able to phrase clearly:

Start with a hypothetical to frame the analysis. It is 1974 and we are approaching the dawn of the VCR. At a point before any VCRs have been sold, the potential manufacturers come together and put in place a price fixing regime for the new VCRs. Cartelization at birth. When faced with an antitrust challenge to the price fixing, the manufacturers defend their action on the ground that output has risen relative to the pre VCR baseline. Given the output expansion, their behavior is clearly pro competitive and therefore insulated from antitrust liability and remedies. How should we assess that claim?

We should have in mind social welfare in four states of the world: (1) the pre VCR world; (2) a world with VCRs and price fixing; (3) a fully competitive VCR world; and (4) finally, a world with an antitrust remedy that blocks VCR price fixing. Welfare clearly increases as we move from world 1 (no VCRs) to world 2 (price fixed VCRs) to world 3 (competitively priced VCRs). The question for an antitrust enforcer is whether blocking price fixing pushes us to competitively priced VCRs or relegates us to a world without VCRs. If price fixing was actually marginal for the sustainability of VCRs, we would clearly be better off to permit the price fixing.

Note how misleading it is to just focus on output. On the first day of a new antitrust class, you tell your students the central message of antitrust: Monopolies reduce output and thereby harm consumers. When you look at an antitrust problem, look for the output reduction. Now flip the proposition: If we see an increase in output, should we think that there are no antitrust concerns? That proposition almost certainly turns out to be wrong and not a meaningful guide to how we do or should implement antitrust policy. Prior to the creation of the cartel, VCRs were not offered at all. In some fundamental sense, the cartel will clearly expand output of VCRs as there were none before. Is that output expansion a good defense to the price fixing case? I think the black letter response to that is no, as price fixing is per se illegal.

Why might that make sense? Because of what we think will happen when we block the price fixing. We don’t compare the before and after worlds; that is, the world before the VCR cartel with the world with the cartel. Instead we focus on what we believe will happen if the cartel is blocked. Antitrust assumes that the price fixing isn’t essential to the creation of the new good. Blocking the proposed deal doesn’t mean that VCRs will go away but rather, instead, we will direct the innovative instinct in a different direction, namely, competitively priced VCRs. That is the world that we think will emerge when we use antitrust to intervene to block the VCR cartel. The relevant comparison isn’t the before and after worlds as to the cartel but rather the worlds with and without the antitrust remedy.

I also particularly liked this passage:

Try a second hypothetical. Suppose two firms got together to form a pro competitive joint venture in market X but at the same time planned to fix prices in an unrelated market, market Y. The firms don’t try to hide this collusion, indeed quite the opposite. They argue that their new arrangements are net pro competitive, that is that the benefits to consumer in market X far exceed any losses that consumers will suffer in market Y. Should the government ignore the price fixing in market Y?

No. We do not let firms assemble bundles of projects and justify anticompetitive projects based on the notion that, on balance, the sum of the benefits of the new projects remain pro competitive. The opposite rule would mean that firms would routinely seek to bundle anticompetitive projects with pro competitive projects.

Government regulators should and do separate projects when firms present a bundle to the government. If the joint venture in market X actually makes sense, we should expect the firms to move forward independent of whether their anticompetitive effort in market Y is blocked. We will not let pro competitive benefits in one market operate as a protective shield for anticompetitive behavior in other markets or in that market. We maximize competitive benefits by not allowing private firms to spend their pro competitive benefits on anticompetitive behavior.

I’ll comment on the other papers as I get my hands on them.