Copyright Society: first sale

Digital First Sale

Panelists: Karyn T. Claggett, U.S. Copyright Office

History repeating: we studied digital first sale before studying digital first sale was cool.  PTO is now having a series of roundtables on these issues—next will be in Boston, June 25.

What has changed since 2001?  Is there tech that could now overcome issues we identified about expanding digital first sale?  Is there “transporter” technology to move a material object and not leave a copy at the point of departure?  ReDigi argued that it had created that kind of tech.  Court disagreed.  It didn’t matter whether there was an additional copy, because the issue was that the reproduction right—a new copy—was implicated.  There was no move, and that was the most significant aspect. But as a policy matter, transporter tech would be very important. (Suppose we do get a transporter.  Could you carry a book through a portal without infringing?)

Courts are confused about sale/ownership.  Cases like Vernor v. Autodesk; international law (CJEU UsedSoft case) doesn’t comport with how we’ve considered the issue of ownership v. licensing.  If courts are confused, consumers are even more confused.

John Villasenor, Brookings Fellow and Professor of Electrical Engineering and Public Policy at UCLA

What would happen if we had digital first sale?  Consider 1 million music fans who each like to listen to a particular song that last 3 minutes once a week.  How many copies of that song would there need to be for there to be a good chance that you could borrow it and listen it at a time of your choosing, then give it back?  Answer: a few hundred.  Could we restrict transfers so you can’t loan a work for a few seconds?  If you try to draft language to allow reasonable transfers but restrict others, it turns out to be impossible, or at least he hasn’t seen it.   (I guess we’re about to conduct a natural experiment on this w/software in the EU.)

We’ve already moved to a licensing based system of digital distribution. When there’s no sale, then there’s no issue of §109 expansion, because there’s no ownership.  In today’s license-based ecosystem, permissible downstream uses are addressed by a mix of contract and IP.  (And consumer protection law, if you live in the EU.)  Much of the argument for digital first sale stems from consumer frustration about limits.  There is a legit concern consumers face with their ability to dispose of works; we should acknowledge those concerns.  You can get more content now than 5-10 years ago, but that doesn’t erase the concerns about the messaging we give consumers. 

Consumers who buy copies of digital works are often subject to complex agreements saying they’re licenses.  If you asked a person on the street what it means if you hit the “buy” button for a movie, there’s an enormous amount of confusion (RT: where “confusion” means “beliefs I don’t like”), and that’s not on them. There is at least an ethical and possibly a legal obligation under deceptive marketing laws to properly inform consumers about what they’re getting. This should not be addressed through copyright law, or through law altering the kinds of agreements providers are allowed to offer consumers, but it is an important issue. Solving the problem of better disclosure will go a long way to put to rest the clamor for a digital first sale doctrine. Market forces would lead to license-based content that would give consumers more options than they have. Letting market develop on its own, not one size fits all statutory approach.  Licenses could explicitly be designed to permit reasonable downstream dispositions.  Could emulate the rules about tangible copies.

(This is an interesting approach, and has a fair amount of logical force.  The problem it will face is that disclosures, as a class, struggle to work even when the consumer is motivated and attentive to the relevant information.  Consumers will not be motivated to attend to this information, and that means it will be all but impossible to “educate” them about the terms the copyright owner would like to impose.  I do not believe that it is possible to have your cake and license it too here.  There’s a reason the button doesn’t say “license.”  It says “buy” or “rent.”  And I expect it will continue to do so.)

Janet Cullum, Cooley LLP

ReDigi did not avoid the cloud model. Suppose you have a legit copy, stored in the cloud.  You transfer your account to me, losing access while I have it, and I listen to the copy. ReDigi decision itself doesn’t put that off limits.  Contracts?

Apple and Amazon are thinking about how to facilitate this consumer desire to be able to resell their works. Patent applications: Apple has contemplated a Bitcoin-like encryption model, where your digital copy needs a key to access.  To transfer for sharing/sale you’d give that key to the next user.  Why would Apple contemplate this?  These reflect desires to create a userbase/keep it happy. Walled garden! You’ll buy more music/more devices. 

Qwest TV commercial, circa 1999: In the middle of nowhere, small motel with few amenities: but all rooms have every movie ever made, in any language, anytime, day or night.  The future is here! Does ownership make sense? Or should I just rent/stream?  Like renting a Zipcar.  In the future, will consumers tell themselves they don’t need to possess anything, either digitally or physically, because they can get anything, any time, in a seamless fashion?

(RT: Note the implicit assumption that there is a fixed pool of creators who create stuff, and users who consume stuff, and never the twain shall meet, certainly not at the point of fair use. How one crosses over to become a creator is an exercise left for the reader—or should I say the renter.  And that’s even setting sociology/psychology aside, though a renter class behaves differently than an owner class for those reasons as well.  Another observation: if the people in this room buy this story, then they really should go for net neutrality. A country in which our internet service is much slower and more expensive than that in other developed countries is not one where it makes sense for consumers to switch and be at the mercy of Verizon.)

Streaming probably won’t eclipse all models—slower in books than in music. Cons: Economic model isn’t sufficient for compensation right now, even though pros are that consumers discover more works, lower manufacturing costs, and possible replacement for piracy.  Consumers have to give up resale right; depend on provider keeping things current/having internet access. But the pros for them are greater: vast quantities of content, convenient.

Studios and labels are winners: broader audiences, cut deals with streaming platforms that give them increased control—generate royalties, ownership interest in some platforms.  To the extent platforms are successful, they too reap economic rewards.  (I wonder how the Hachette/Amazon fight works as evidence here.  The platforms love not studios/publishers/etc.)

Smaller players have it harder: they don’t have the bargaining power that large legacy rightsholders do.  NYT: trade groups representing 1000s of independent labels and musicians appealed to the EC, accusing YouTube of unfair contracts for a planned music service, and threat to block labels’ content if they don’t sign.  Am Ass’n of Indep Music has asked FTC to stop YouTube from blocking its members’ content, abuse of a dominant position.

For streaming services: profitability challenges are huge.  50,000 foot view—if they can continue this kind of growth, they can monetize audiences.  People wondered whether Facebook could make money not long ago. ASCAP consent decree must be revamped.  Tensions in marketplace: Pandora won case on whether rightsholders could pull out from streaming.  Ultimately, regulation + market will get us past the profitability challenge.

Q: if there’s digital first sale, what prevents HathiTrust or other mass digitizers from selling their copies?

A: Nothing.

Claggett: concern would be if they’re lawfully made copies, they could be sold/resold.

Interplay between contracts and exceptions—should contract be able to override fair use?  Many communities argue we should have presumption in favor of exceptions so contract couldn’t trump them.

Q: what about the legal force of the “buy” statement on the button?  If they present it as a sale, don’t consumers have the right to treat it that way?

Villasenor: good case law on the difference between license and sale from 9^th Circuit, though Vernor is specific to software. UMG v. Augusto: UMG distributed promo CDs, the court rightly found that recipients weren’t bound by the statement on the CDs that they were promotional.  Contract providers have a need for simplicity. But they pay a downstream price. Content providers would do us/themselves a service by providing more clarity that they’re not selling ownership; lack of clarity annoys consumers, gives consumers reasons to complain to legislatures, and gives plaintiffs ammunition to say they’re owners.  Judge/jury’s reaction to a big “buy” button is of concern.

Claggett: EU case: found that there was actually a sale.  It may be claimed to be a license, but court found it to be a sale for, among other things, policy reasons—otherwise there’d never be a first sale if even unlimited licenses were found to be just licenses and not sales.

Cullum: typical consumer—but that case will never be brought.  Harder to bring a class action.  Intermediary drives the case law: who’s making the market? Harder to predict in that context.

Q: how do you counsel clients in the EU?

Cullum: you have to look at local law.

Villasenor: extraterritorial recharacterization of what the content owner thought were licenses as sales is of concern, especially given Kirtsaeng.

Claggett: look at key aspects of UsedSoft case that made it sale.  Are there additional restrictions on licensors you could put?

Villasenor: without agreeing w/the decision, the court cited the perpetual nature of the license as dispositive in terms of recharacterizing. 

We aren’t ever confused when we rent cars about whether we actually own things, even though the car contract is long. Companies could do a lot to provide more clarity/flexibility if they wanted to, but they shouldn’t be forced to, but they could distill messages to consumers.  (RT: Yeah, because when you rent a car you return it, and it’s gone. That’s not like the Placebo track I bought eight years ago, which lives to this day in my iTunes.  Try to distill the message that supposedly needs to be conveyed here, which is quite distinct from the message Hertz sends you: it’s yours forever, but you don’t own it.  I don’t think that works as well.  I’m also perplexed by the idea that it’s shocking that a perpetual license should be characterized as a sale, since the very reason you want to end the transaction having purchased access to the content forever is so that you won’t lose it for failing to pay the rent: conventionally speaking, you wish to own it.)

Villasenor continues by noting that content owners would love to stop selling things across the board, but that would create various provisions.

Claggett: there are arguments for giving various rights/privileges to the lawful possessor of content, whether or not it’s owned.

Cullum: evolution to expectation of “free.”  Production cost is cheaper, but that expectation must be changed—and she thinks it will, because business models can provide value-adds like prepopulated playlists and additional services that are worth paying for and can be ratcheted up in price over time until they start making lots of money.