Yes, there might be a hypothetical market for thumbnail licensing, but that’s not the point. The point is that there is a very real market for licensing books to use them in the way Google Print will use them.

Yes, The NYT might like to make bloggers license quotes, but it’s not allowed not because there isn’t a potential market for such licensing but because quoting for criticism and journalism is fair use that is not only explicitly allowed by the statute, but would also easily pass the fair use test. What Google is doing is not even close to be a clear case like that.

I’m aware of the Simpsons example from Lessig’s book. But the point of Lessig’s argument in that passage wasn’t that the 4 second clip should be covered by fair use–because it should be obvious to anyone that it is an instance of fair use–but that users won’t make use of a work even if they know they’ll qualify for fair use because they can’t afford a lawsuit and content companies get away with extracting rents they don’t deserve. So, I’m not sure how this is relevant here.

You say, “Your premise seems to me that a copyright holder has a right to exploit every conceivable derivative market for his work, no matter how tenuous, unless some kind of utilitarian calculus shows that doing so harms society.” That’s not exactly what I’m saying. But I am saying that if a utilitarian calculus shows a great benefit to society that only marginally harms copyright holders, then that should be considered fair use. And by making that showing is how I think Google can win.

What I’m saying fits squarely with precedent. The section you quote from Sony buttresses my argument. You quoted: “In the Sony Betamax decision, the court held that ‘there is no likelihood that time-shifting would cause nonminimal harm to the potential market for, or the value of, respondents’ copyrighted works.’” Notice that the court was careful to include in its opinion the “potential market”. That’s because the statute includes potential markets in the equation. Courts have followed the statute and looked at a potential market. You go on to say, “It didn’t consider whether the VCR harmed a hypothetical market for time shifting licenses.” It did. It sounds absurd to think of “time shifting licenses,” as if one would call up MGM and buy permission from them to record a show so you can watch it later. But that is what DVDs and VHS tapes are. What the film industry wanted was to make VCRs play-only with no record button. In that world, if channel 10 was playing Jaws Tuesday at 8 pm, but you wanted to watch it on Saturday, you would have to go to the store and buy the VHS tape to do so–in effect licensing time-shifting. Was this potential market harmed? Yes. Did the court consider this? You bet. Was this harm “minimal” and outweighed by an incredible gain in social welfare? Yes, and that’s why the copyright holder’s copyright should yield.

Yes, there might be a hypothetical market for thumbnail licensing, but that’s not the point. The point is that there is a very real market for licensing books to use them in the way Google Print will use them.

Yes, The NYT might like to make bloggers license quotes, but it’s not allowed not because there isn’t a potential market for such licensing but because quoting for criticism and journalism is fair use that is not only explicitly allowed by the statute, but would also easily pass the fair use test. What Google is doing is not even close to be a clear case like that.

I’m aware of the Simpsons example from Lessig’s book. But the point of Lessig’s argument in that passage wasn’t that the 4 second clip should be covered by fair use–because it should be obvious to anyone that it is an instance of fair use–but that users won’t make use of a work even if they know they’ll qualify for fair use because they can’t afford a lawsuit and content companies get away with extracting rents they don’t deserve. So, I’m not sure how this is relevant here.

You say, “Your premise seems to me that a copyright holder has a right to exploit every conceivable derivative market for his work, no matter how tenuous, unless some kind of utilitarian calculus shows that doing so harms society.” That’s not exactly what I’m saying. But I am saying that if a utilitarian calculus shows a great benefit to society that only marginally harms copyright holders, then that should be considered fair use. And by making that showing is how I think Google can win.

What I’m saying fits squarely with precedent. The section you quote from Sony buttresses my argument. You quoted: “In the Sony Betamax decision, the court held that ‘there is no likelihood that time-shifting would cause nonminimal harm to the potential market for, or the value of, respondents’ copyrighted works.’” Notice that the court was careful to include in its opinion the “potential market”. That’s because the statute includes potential markets in the equation. Courts have followed the statute and looked at a potential market. You go on to say, “It didn’t consider whether the VCR harmed a hypothetical market for time shifting licenses.” It did. It sounds absurd to think of “time shifting licenses,” as if one would call up MGM and buy permission from them to record a show so you can watch it later. But that is what DVDs and VHS tapes are. What the film industry wanted was to make VCRs play-only with no record button. In that world, if channel 10 was playing Jaws Tuesday at 8 pm, but you wanted to watch it on Saturday, you would have to go to the store and buy the VHS tape to do so–in effect licensing time-shifting. Was this potential market harmed? Yes. Did the court consider this? You bet. Was this harm “minimal” and outweighed by an incredible gain in social welfare? Yes, and that’s why the copyright holder’s copyright should yield.

Second, there is no market for thumbnails. Is there a market in this case? That depends on how you define the market, but I think the answer is yes. If you define it broadly–is there a market for licenses to permit copyrighted work to be included in searchable databases?–the answer is likely yes. Look as Amazon Pages and Search Inside the Book, Lexis, Westlaw, etc. But even if the market is defined more narrowly–is there a market for licensing to allow snippets–the answer may well still be yes. You have to look no further than Google’s own Print Publisher program.

I still don’t buy it. You could just as easily argue that there’s a market to license thumbnails. Now, obviously, such a market would be a big administrative headache and generate very little revenue from copyright holders, but that doesn’t mean it couldn’t exist.

Indeed, any kind of derivative work now recognized as fair use could instead be subject to a licensing regime. I have no doubt that New York Times would love to charge bloggers a nickel every time the quoted Paul Krugman in a blog post. Or check out this chapter of Lessig’s free culture, where he talks about Fox News attempt to charge him $10,000 to leave a 4-second clip of the Simpsons in the corner of one of his shots. I don’t just think that’s stupid because of a utilitarian calculus that society has been harmed. I think that Fox’s right to profit from the Simpsons doesn’t include the right to place such absurd limits on reasonable re-use of its products.

Your premise seems to me that a copyright holder has a right to exploit every conceivable derivative market for his work, no matter how tenuous, unless some kind of utilitarian calculus shows that doing so harms society. I don’t think that’s consistent with precedent. In Arriba Soft, the court focused on the market for the original thumbnails, not on the viability of a hypothetical thumbnail market. In the Sony Betamax decision, the court held that “there is no likelihood that time-shifting would cause nonminimal harm to the potential market for, or the value of, respondents’ copyrighted works.” It didn’t consider whether the VCR harmed a hypothetical market for time shifting licenses. Likewise, in the Diamond decision, I don’t recall them even discussing whether there was a market for licensing space-shifting rights.

Copyright law strikes a balance between the interests of authors and the interests of the consuming public. The standard you propose–that any conceivable derivative market is presumptively the property of the copyright holder absent compelling argument to the contrary, strikes me as a poor way of striking that balance.

Hmmm… I feel like we’ve had this argument before.

Second, there is no market for thumbnails. Is there a market in this case? That depends on how you define the market, but I think the answer is yes. If you define it broadly–is there a market for licenses to permit copyrighted work to be included in searchable databases?–the answer is likely yes. Look as Amazon Pages and Search Inside the Book, Lexis, Westlaw, etc. But even if the market is defined more narrowly–is there a market for licensing to allow snippets–the answer may well still be yes. You have to look no further than Google’s own Print Publisher program.

I still don’t buy it. You could just as easily argue that there’s a market to license thumbnails. Now, obviously, such a market would be a big administrative headache and generate very little revenue from copyright holders, but that doesn’t mean it couldn’t exist.

Indeed, any kind of derivative work now recognized as fair use could instead be subject to a licensing regime. I have no doubt that New York Times would love to charge bloggers a nickel every time the quoted Paul Krugman in a blog post. Or check out this chapter of Lessig’s free culture, where he talks about Fox News attempt to charge him $10,000 to leave a 4-second clip of the Simpsons in the corner of one of his shots. I don’t just think that’s stupid because of a utilitarian calculus that society has been harmed. I think that Fox’s right to profit from the Simpsons doesn’t include the right to place such absurd limits on reasonable re-use of its products.

Your premise seems to me that a copyright holder has a right to exploit every conceivable derivative market for his work, no matter how tenuous, unless some kind of utilitarian calculus shows that doing so harms society. I don’t think that’s consistent with precedent. In Arriba Soft, the court focused on the market for the original thumbnails, not on the viability of a hypothetical thumbnail market. In the Sony Betamax decision, the court held that “there is no likelihood that time-shifting would cause nonminimal harm to the potential market for, or the value of, respondents’ copyrighted works.” It didn’t consider whether the VCR harmed a hypothetical market for time shifting licenses. Likewise, in the Diamond decision, I don’t recall them even discussing whether there was a market for licensing space-shifting rights.

Copyright law strikes a balance between the interests of authors and the interests of the consuming public. The standard you propose–that any conceivable derivative market is presumptively the property of the copyright holder absent compelling argument to the contrary, strikes me as a poor way of striking that balance.

Hmmm… I feel like we’ve had this argument before.

Does Google profit when they offer this service? Probably yes. (I, for one, don’t think corporations do anything–especially of this magnitude–altruistically.) Although they do not display ads in the snippets pages, ads are displayed in the search results page. Also, providing this service draws users to Google’s Print service generally where there are pages with ads. It therefore wouldn’t be unthinkable for a court to find that Google is profiting by using the copied works. What we need to argue is not that Google is not profiting, but that even if it is it should still be found to be a fair use because, as in Arriba, “the public benefit of the search engine” is great. That is, we need to make courts focus not on the commercial middleman, but on the fair use that the individuals using the service are making. We can’t do this by denying that the middleman is commercial.

How is a thumbnail different? First, folks who put up pictures and text on the web are tacitly accepting that their content will be indexed (and copied whenever it is accessed); book publishers, on the other hand, don’t put their content on the web. Second, there is no market for thumbnails. Is there a market in this case? That depends on how you define the market, but I think the answer is yes. If you define it broadly–is there a market for licenses to permit copyrighted work to be included in searchable databases?–the answer is likely yes. Look as Amazon Pages and Search Inside the Book, Lexis, Westlaw, etc. But even if the market is defined more narrowly–is there a market for licensing to allow snippets–the answer may well still be yes. You have to look no further than Google’s own Print Publisher program.

That last prong is going to be the hardest to get around. We’re going to have to convince a court–if it is the case–that without a fair use designation a market will never provide the kind of “universal service” (I can’t believe I’m using that term, and it gives me pause) that a Google Print can supply–even if it harms a potential market for licensing. We have to convince a court that the same way we don’t countenance a market for licenses to quote excerpts in book reviews (even though such a market is conceivable), the type of copying that Google Print plans to do should be allowed despite the possible existence of a licensing market because the social benefit that would otherwise be lost outweighs any interest in preserving publisher rights. You don’t do this by denying that there is a market for licensing snippets.

Does Google profit when they offer this service? Probably yes. (I, for one, don’t think corporations do anything–especially of this magnitude–altruistically.) Although they do not display ads in the snippets pages, ads are displayed in the search results page. Also, providing this service draws users to Google’s Print service generally where there are pages with ads. It therefore wouldn’t be unthinkable for a court to find that Google is profiting by using the copied works. What we need to argue is not that Google is not profiting, but that even if it is it should still be found to be a fair use because, as in Arriba, “the public benefit of the search engine” is great. That is, we need to make courts focus not on the commercial middleman, but on the fair use that the individuals using the service are making. We can’t do this by denying that the middleman is commercial.

How is a thumbnail different? First, folks who put up pictures and text on the web are tacitly accepting that their content will be indexed (and copied whenever it is accessed); book publishers, on the other hand, don’t put their content on the web. Second, there is no market for thumbnails. Is there a market in this case? That depends on how you define the market, but I think the answer is yes. If you define it broadly–is there a market for licenses to permit copyrighted work to be included in searchable databases?–the answer is likely yes. Look as Amazon Pages and Search Inside the Book, Lexis, Westlaw, etc. But even if the market is defined more narrowly–is there a market for licensing to allow snippets–the answer may well still be yes. You have to look no further than Google’s own Print Publisher program.

That last prong is going to be the hardest to get around. We’re going to have to convince a court–if it is the case–that without a fair use designation a market will never provide the kind of “universal service” (I can’t believe I’m using that term, and it gives me pause) that a Google Print can supply–even if it harms a potential market for licensing. We have to convince a court that the same way we don’t countenance a market for licenses to quote excerpts in book reviews (even though such a market is conceivable), the type of copying that Google Print plans to do should be allowed despite the possible existence of a licensing market because the social benefit that would otherwise be lost outweighs any interest in preserving publisher rights. You don’t do this by denying that there is a market for licensing snippets.

How are my robots different than the Arriba Soft case? Arriba Soft was a commercial entity, and I think a thumbnail is pretty strongly analogous to an excerpt.

On the first prong, keep in mind that Google displays no ads on in-copyright books that haven’t been authorized by the publisher. So while Google is certainly a commercial company, it’s not clear that the use Google makes of the books is especially commercial.

Google Print isn’t a substitute for or competitor with Amazon’s pay-for-pages program. Amazon offers pages, chapters, or (most often) entire books. Google Print offers short excerpts. So, as I’ve argued before, I think the fourth prong will weigh in favor of Google, as it did in in Arriba Soft. If Google Print harms the market for “excerpt rights” couldn’t you just as easily argue that Arriba Soft harmed the market for “thumbnail rights?”

How are my robots different than the Arriba Soft case? Arriba Soft was a commercial entity, and I think a thumbnail is pretty strongly analogous to an excerpt.

On the first prong, keep in mind that Google displays no ads on in-copyright books that haven’t been authorized by the publisher. So while Google is certainly a commercial company, it’s not clear that the use Google makes of the books is especially commercial.

Google Print isn’t a substitute for or competitor with Amazon’s pay-for-pages program. Amazon offers pages, chapters, or (most often) entire books. Google Print offers short excerpts. So, as I’ve argued before, I think the fourth prong will weigh in favor of Google, as it did in in Arriba Soft. If Google Print harms the market for “excerpt rights” couldn’t you just as easily argue that Arriba Soft harmed the market for “thumbnail rights?”

You’re right that Nimmer might be missing the point of Google Print and might be making it sound nefarious. But the bottom line is that Google is “distributing” “massive numbers of copies of other persons’ property” “with impunity.” You can’t only look at the quantity of what is being distributed. (Sure, it’s just a snippet, and the third prong of fair use will weigh in favor of Google.) You have to also look at the other prongs of the test. Those include the first prong, whether the use is commercial (which it is), and the fourth prong, “the effect of the use upon the potential market for or value of the copyrighted work.” Especially now that Amazon has announced a pay-for-pages program I don’t see how one can argue there isn’t a market for this kind of licensing. You say, “Book reviewers, journalists, bloggers, and others have been ‘distributing’ other peoples’ works for decades.” Yes, but for purposes that don’t affect the market for the original and that are explicitly mentioned in the statute as allowed.

You’re right that Nimmer might be missing the point of Google Print and might be making it sound nefarious. But the bottom line is that Google is “distributing” “massive numbers of copies of other persons’ property” “with impunity.” You can’t only look at the quantity of what is being distributed. (Sure, it’s just a snippet, and the third prong of fair use will weigh in favor of Google.) You have to also look at the other prongs of the test. Those include the first prong, whether the use is commercial (which it is), and the fourth prong, “the effect of the use upon the potential market for or value of the copyrighted work.” Especially now that Amazon has announced a pay-for-pages program I don’t see how one can argue there isn’t a market for this kind of licensing. You say, “Book reviewers, journalists, bloggers, and others have been ‘distributing’ other peoples’ works for decades.” Yes, but for purposes that don’t affect the market for the original and that are explicitly mentioned in the statute as allowed.

The defendant in Arriba Soft did not keep the original copies and continue to use them, but used only the highly modified, transformed thumbnail images, while Google plans to keep and make available copies of the original works.

How is a thumbnail a “highly modified, transformed” copy, while an excerpt–which provides a much smaller and less useful fraction of the original work–is not? It’s true that Google is planning to “keep” copies of the original works, but it is only planning to “make available” excerpts which are every bit as “modified” and “transformed” as a thumbnail.

Don’t you think the above bullet point is misleading?

The defendant in Arriba Soft did not keep the original copies and continue to use them, but used only the highly modified, transformed thumbnail images, while Google plans to keep and make available copies of the original works.

How is a thumbnail a “highly modified, transformed” copy, while an excerpt–which provides a much smaller and less useful fraction of the original work–is not? It’s true that Google is planning to “keep” copies of the original works, but it is only planning to “make available” excerpts which are every bit as “modified” and “transformed” as a thumbnail.

Don’t you think the above bullet point is misleading?

Imagine if, instead of scanning the books and storing them on its servers, Google deployed an army of robots at the University of Michigan library. Each time a user did a search, a robot would go to the appropriate book, turn to the appropriate page, and scan it, and provide he user with an excerpt. No book would ever be scanned in its entirety, and no copies of books would be stored on Google’s servers. Obviously, that would be extremely slow, expensive, and impractical, but I think everyone would agree that that’s fair use, would they not?

So I think there’s a pretty strong case that the excerpts, as such, are fair use under established law. Maybe there are a few exceptions, but I don’t know of any, and I can’t think of any reason to think that Google Print is less likely to pass muster under the four-part test than other similar-length excerpts that have withstood fair use analysis.

The harder question is whether the means employed–storing electronic copies of the works on Google’s servers–is fair use. I agree that that’s not an open-and-shut case under copyright law.

But I think to sensibly analyze it, you have to keep clearly in mind that the “copies” being distributed are short excerpts, not entire books. While Nimmer’s article doesn’t explicitly deny this, I think the way he talks about it is extremely misleading. He talks about Goole claiming the right to “copy and distribute the property of another person,” to do so “with impunity,” to “keep and make available copies” of copyrighted works, and to make a “massive number of copies of other persons’ property.” While those phrases might be technically true, I don’t think that conveys an accurate idea of what Google is doing. Google isn’t “distributing” “massive numbers of copies” “with impunity.” It’s displaying a tiny portion of copyrighted works, for a clearly defined purpose, with numerous safeguards in place to ensure that no one is able to get the whole book. Nimmer didn’t seem to find any of those details worth mentioning.

I think his policy argument at the end is especially misleading. He talks about how authors have the exclusive right to re-distribute their works. Fair enough. But what doe she mean by distribution? If he includes the reproduction of brief excerpts under the term, then authors have never had the right to “control how the work is distributed.” Book reviewers, journalists, bloggers, and others have been “distributing” other peoples’ works for decades. On the other hand, if by “distribution” he only means distributing entire works–or substantial portions thereof–then Google isn’t claiming the right to do so. Either way, that paragraph doesn’t make sense.

Imagine if, instead of scanning the books and storing them on its servers, Google deployed an army of robots at the University of Michigan library. Each time a user did a search, a robot would go to the appropriate book, turn to the appropriate page, and scan it, and provide he user with an excerpt. No book would ever be scanned in its entirety, and no copies of books would be stored on Google’s servers. Obviously, that would be extremely slow, expensive, and impractical, but I think everyone would agree that that’s fair use, would they not?

So I think there’s a pretty strong case that the excerpts, as such, are fair use under established law. Maybe there are a few exceptions, but I don’t know of any, and I can’t think of any reason to think that Google Print is less likely to pass muster under the four-part test than other similar-length excerpts that have withstood fair use analysis.

The harder question is whether the means employed–storing electronic copies of the works on Google’s servers–is fair use. I agree that that’s not an open-and-shut case under copyright law.

But I think to sensibly analyze it, you have to keep clearly in mind that the “copies” being distributed are short excerpts, not entire books. While Nimmer’s article doesn’t explicitly deny this, I think the way he talks about it is extremely misleading. He talks about Goole claiming the right to “copy and distribute the property of another person,” to do so “with impunity,” to “keep and make available copies” of copyrighted works, and to make a “massive number of copies of other persons’ property.” While those phrases might be technically true, I don’t think that conveys an accurate idea of what Google is doing. Google isn’t “distributing” “massive numbers of copies” “with impunity.” It’s displaying a tiny portion of copyrighted works, for a clearly defined purpose, with numerous safeguards in place to ensure that no one is able to get the whole book. Nimmer didn’t seem to find any of those details worth mentioning.

I think his policy argument at the end is especially misleading. He talks about how authors have the exclusive right to re-distribute their works. Fair enough. But what doe she mean by distribution? If he includes the reproduction of brief excerpts under the term, then authors have never had the right to “control how the work is distributed.” Book reviewers, journalists, bloggers, and others have been “distributing” other peoples’ works for decades. On the other hand, if by “distribution” he only means distributing entire works–or substantial portions thereof–then Google isn’t claiming the right to do so. Either way, that paragraph doesn’t make sense.

Distribution to the public is also an exclusive right. As a general matter, you can violate it by distributing without authorization someone else’s copyrighted work — whether it’s a whole book or just a little bit of it. If you can successfully defend what you’re doing under fair use, then you won’t be liable for infringement. But, there’s no denying that you are still distributing someone else’s copyrighted work. So, Nimmer is right to make the statement that Google will be distributing someone else’s copyrighted works. There’s no denying that that is what Google will be doing. The question is whether it constitutes an infringement or not; whether it can be defended successfully as fair use or not.

You say, “Google Print doesn’t distribute copyrighted books, aside from displaying small snippets that are generally agreed to be fair use.” However, I’m not sure that there is general agreement that short snippets always qualify as fair use (especially when Nimmer, one of the foremost experts in the field doesn’t agree). For one thing, what determines fair use is a multi-part, contextual test. So I’m not sure anyone can give an opinion about whether something is fair use based only on the quantity of the copy distributed.

Lots of folks seem to be arguing that this is an open and shut win for Google under existing fair use precedent. As someone who would like to see Google Print qualify as fair use, I think we have to be realistic about the fact that fair use as it is right now gives Google Print at best a 50-50 shot. We have to make the case that fair use can logically be extended–given its traditional purpose–to encompass Google Print. GP will–hopefully and rightly–push out the bounds of fair use to take into account new technology (i.e. the network), but fighting this fight within the existing parameters might be a losing proposition.

Distribution to the public is also an exclusive right. As a general matter, you can violate it by distributing without authorization someone else’s copyrighted work — whether it’s a whole book or just a little bit of it. If you can successfully defend what you’re doing under fair use, then you won’t be liable for infringement. But, there’s no denying that you are still distributing someone else’s copyrighted work. So, Nimmer is right to make the statement that Google will be distributing someone else’s copyrighted works. There’s no denying that that is what Google will be doing. The question is whether it constitutes an infringement or not; whether it can be defended successfully as fair use or not.

You say, “Google Print doesn’t distribute copyrighted books, aside from displaying small snippets that are generally agreed to be fair use.” However, I’m not sure that there is general agreement that short snippets always qualify as fair use (especially when Nimmer, one of the foremost experts in the field doesn’t agree). For one thing, what determines fair use is a multi-part, contextual test. So I’m not sure anyone can give an opinion about whether something is fair use based only on the quantity of the copy distributed.

Lots of folks seem to be arguing that this is an open and shut win for Google under existing fair use precedent. As someone who would like to see Google Print qualify as fair use, I think we have to be realistic about the fact that fair use as it is right now gives Google Print at best a 50-50 shot. We have to make the case that fair use can logically be extended–given its traditional purpose–to encompass Google Print. GP will–hopefully and rightly–push out the bounds of fair use to take into account new technology (i.e. the network), but fighting this fight within the existing parameters might be a losing proposition.