SIVACRACY.NET

Many people have written me asking me to elaborate on why I thought Google was out of line when it planned to copy millions of library books without permission from the copyright holder.

I am writing a paper about Google that comes out of my ALA talk. In it, I will walk through the precedents and arguments. In the mean time, check out Video Pipeline, Inc. v. Buena Vista Home Entertainment, Inc. This is a case about a company that did something close to what Google hoped to do -- with tragic results. It's not a sophisticated preliminary injunction decision. But I fear it's close to what Google had waiting for it.

In other news, The Register covers the story here. Sadly, this story confuses readers about the position of libraries in the Google Library plan. Many librarians were excited about it. Many were wary. There was not unified opinion. Libraries themselves did not take positions, save the four that were contracted with Google to allow access to their collections. The American Library Association, to my knowledge, did not take an official position, although its president did speak and write critically about the Google deal.

Copyfight has an interesting debate about my post. Many smart people weigh in here, including Derek Slater, who takes issue with my argument (while acknowledging that it's very brief).

That Was Sarcasm, Right?

I'm having trouble following Siva's reaction to Google's pausing its Google Print program.

Siva is quick to point to fair use cases counting against Google - that's odd, considering that in many contexts he would readily criticize bad strands of copyright caselaw. Google Print is making intermediate copies of the entire works in order to display a few pages from those books. With respect to books under copyright, the entire book is not displayed. Is Siva saying that any intermediate copying of an entire work by a corporation is unfair? What would that mean for reverse engineering? It'd be an admittedly tough case for Google to win, but I'm not so sure that Google actually is dead in the water under caselaw. If it is, I don't think it ought to be.

Regardless, the caselaw doesn't amount to what Siva implies it does. Though it's only a brief citation, it seems Siva seriously misreads American Geophysical Union v. Texaco. The court didn't rule against Texaco because it was a corporation. In fact, the appeals court specifically disagreed with the district court's "undue emphasis" on the for-profit nature of Texaco.

In the end, Texaco's copying of research papers for achival was found unfair under the first factor, for reasons closely connected to the fourth factor (market impact). And, on that point, I'm not sure what Siva's applauding. This case followed a broad reading of the works' potential market so that it includes any "traditional, reasonable, or likely developed markets." That does not comport with Sony's more constrained reading of this factor - it certainly seems "reasonable" that the television program copyright holders could and would develop a market that time-shifting undermines. The court also suggested that if the copyright holder is offering licenses for a use, then that's a developed market. This is one case in the dangerous, gradual expansion of this factor - see Mp3.com and Napster.

In American Geophysical Union, the court did note that the first factor (and thus the fourth, at least under Sony) was more likely to be considered fair under the first factor where it "produces a value that benefits the broader public interest." Reverse engineering would fit that category; Texaco's didn't. It seems that it is here that Siva seriously takes issue with Google.

We can put aside caselaw and go to straight-up normative analysis - Siva thinks that this Google Print is bad, bad, bad. What I see is gross hyperbole. What Google's doing is nothing like widespread infringing file-sharing on P2P. Sure, they're copying the entire book, but they're only providing small selections. I don't see how that amounts to a "copyright meltdown." (I know that you can try to do different searches to over time accumulate the whole book, but Google does enough to frustrate that, I think.)

Libraries good, corporations bad doesn't ring true for me. Without a doubt, I'm glad that people are becoming more skeptical of Google, despite their "we're not evil" mantra. However, in this case, Google was providing an important public service, one that happened to benefit the company commercially, but one that also did not pose a serious threat to copyright holders (in fact, it probably would help them), and for those reasons I think Google Print should be lawful.

As Jason put it: "This is a clear example of copyright failing the public in the digital age. Google isn't selling the books; they just need to scan them to help Internet users find what they're looking for. The fact that publishers are able to hold up this process works against consumers and the marketplace, not in their favor."

I trust that Derek understands that Texaco is not the only thing I am bringing to this argument. Nor is it simply a matter of "corporations bad, libraries good." Well, libraries ARE good. And copyright law recognizes the distinction quite appropriately.

When examining potential precedents like Texaco, Sony, and most importantly Kelly v. Ariba Soft, it's important to follow who is doing the copying and where the copies are being used and distributed. And it's VERY important to remember who bought the books. It ain't Google. The case would not be a slam dunk either way. But I was pessimistic about Google's chances to prevail and worried more if it did. The blowback would have been serious.

Anyway, there is much more to this question, so you will just have to wait for the paper to get my full argument. But basically, I feared that if Google were to go whole hog toward this project then courts would have had to react to the massive scale of the copying (which Derek calls intermediary but the publishers consider substantial nonetheless). If Google lost, the statuatory damages would have been just as massive as the project itself. That lovely IPO? Gone. Generations of cool Google innovations? Never happen.

I am afraid my "woulds" and "shoulds" are confusing. My concern about the Google project has more to do with the potential repercussions of the project in both courts and in Congress than with the project itself. My issues with the project itself stem from Google's rude behavior and the bad position its contracts put libraries in. I also have general concerns about the position of libraries in our information ecosytem (good) and their relationships with commercial services like Amazon.com and Google (bad).

That said, Google Print is a potentially great product and I cheer it on. I trust that many publishers will see that having their back catalogs searchable by Google is a really good thing for business (corporations, good).

As more publishers follow Oxford University Press in the effort to put out-of-print works through a print-on-demand service, they will see a flurry of new sales for old products. Meanwhile, readers and researchers could have the benefits of searching the text of these books. But it's important for the long term -- and for cultural democracy in general -- to convince copyright holders through argument and experience that they need not (in fact, should not) be so possessive. This will take time, patience, transparancy, and experimentation. Google's plan to just go ahead of scan the books demonstrated none of these features.

We can and will have access to millions of copyrighted books through Google. But this process should occur with the proper licenses.

Once again, I think we should recognize that unless we think copyright should not exist, copyright holders should be able to decide when to license their works to other companies. This is far from absolute. But it's common sense and generally true. Only in unusual circumstances, such as when markets fail to provide an essential public good, should we consider radical moves. This is not one of those cases. The service is not an essential public good -- just a cool idea. And the market was not failing. Publishers were at the table.

Google seemed to agree with me for a while. For months they negotiated with publishers over the terms. Then, out of blue, Google turned on the publishers and announced they were just going to go ahead and digitize these four library collections. Snap.

Google messed up by going all unilateral on the publishers. There was no market failure here. Transaction costs were not prohibitive. They were working out the deal. This was not the recording industry shunning Napster. This was how copyright is supposed to work.

So I am very pleased that Google has decided to wait until November to give itself a chance to work things out with publishers. This latest move is not an example of corporate cowardice. It's wisdom. Google has some brilliant copyright lawyers working there. And I am sure they are way ahead of me on this stuff. And Google's new "opt-out" plan deftly dodges the "orphan works" problem: if no one comes forward to say a book is theirs and is valuable, it goes up on line.

I am confident that an imperfect Google Print project will do more for the Copyfight than a perfect one. Readers will notice the holes in the service and ask "why is there no Steinbeck in here?" or "where is Lolita?" Then they will realize that copyright lasts too long for anyone's good. I am also confident that what Google will produce with permission from publishers will be pretty great.