June 23, 2004
And much more of the Hatch floor statement grates on my nerves. Rather than starting from the broad problems of this bill, I want to examine an interesting move in the statement’s argument:
“A rule that punishes only control also produces absurd results. Secondary liability should focus on intent to use indirect means to achieve illegal ends. A rule that punishes only control degenerates into inane debate about which indirect means was used. Thus Napster and Grokster are regulated differently – though they function similarly from the perspective of the user, the distributor, or the copyright holder.”
“A rule that punishes only control also acts as a ‘tech-mandate’ law: It mandates the use of technologies that avoid ‘control’ – regardless of whether they are suited for a particular task. Napster was punished for processing search requests efficiently on a centralized search index that it controlled. Grokster escaped by processing search requests less efficiently on a decentralized search index that it did not control. Rewarding inefficiency makes little sense.”
The point about decentralized P2P being definitively less efficient is overstated. Beyond efficiency, there are also advantages to not having a single point of control (e.g., failure at a single point does not disrupt the entire network).
But I think the point makes some sense: it is silly that we let Morpheus get a free ride while Napster had to be shut down. Unlike Hatch, however, I certainly don’t think that this inconsistency demands a rule such that that Morpheus is also liable. Indeed, I think we can still leave Napster liable while avoiding the “tech mandate”/tech ban problem. To create a more consistent and beneficial doctrine, we should tweak – but not eviscerate – the Napster decision.
Interpreted broadly, Sony stands for the principle that we should let new technology develop completely unconstrained from secondary liability. As I have discussed elsewhere, I think Sony rule should actually be interpreted a little narrower. Sony stands for not using secondary liability to force tool-makers to modify how their tools operate, but secondary liabilty could apply to the tool-maker’s conduct in relation to the tool-user.
Think of this in terms of Napster. The Napster decision seemed to follow this reading of Sony, noting that Napster’s conduct had to be distinguished from its architecture. Furthermore, the court said that Napster simply had a responsibility to police cabined within the system’s architecture. However, during the remedy phase, the court forced Napster to do something that it could not do within the architecture of its system -eliminate all infringement and implement new filtering tools. Instead, the court could have simply asked Napster to block certain users (a power it retained through its login server) and remove particular files (not all instances of a song, which its file index did not have any ability to identify) when notified by a copyright holder. They could do all that without changing a single line of code in the system.
But isn’t this what the Hatch aide was calling “inane”? You might find this to be simply trying to find a loophole to protect piracy. Those who support the negligence rule would suggest that, if one is willing to put that much burden on Napster, then why not also force them to make a small adjustment to their code that might not even affect non-infringing uses – why not go the extra step and weigh the costs and benefits?
But what change would not affect non-infringing uses, definitively? We don’t know, and no weighing can properly take into account these undefined and thus indeterminately impacted uses. Sony understood this point – that by constraining the way a tool functioned, we risked impairing legitimate uses. The VCR ushered in “time-shifting,” a use that most people probably did not conceive of prior; it, and other technologies, could also usher in other untold uses, which would have been impacted by any change to the tools themselves. Telling Napster to respond to notices of infringement is consistent with this rule as well as with a rule that forces Morpheus to do nothing. One could sever Napster’s infringing uses from the non-infringing uses, known or not; by addressing their conduct only, new uses of the technology could continue to evolve. In so doing, we could get rid of the de facto “tech mandate” that effectively bans centralized P2P and thus no longer “reward inefficiency.”
Certainly, even this burden on centralized P2P operators would have some impact on technology innovation, as would the benefit of infringing uses reward certain technologies. However, I think it places a reasonable burden while largely retaining the clarity, certainty, intent, and benefits of the Sony rule. Also, one can imagine similar problems with this rule as with DMCA 512 notifications. All the more reason to revisit exactly how we deal with counter-notification, but that I’ll have to leave for another day.
One could also argue that, though this rule might make Sony/Napster/Grokster consistent, it’s still an “inane debate” because it does nothing to stop piracy. People who want to infringe would simply move to Morpheus. True enough. I’m not trying to come up with a rule with the sole objective of ending piracy. Indeed, I like this rule in part because I think the alternatives, like a negligence rule or the INDUCE Act, would be so dreadful for innovation and would do little to stop piracy. I’ve addressed the negligence rule here. As for the INDUCE, if it is as benign as Hatch makes it out to be, then it will have little impact on piracy; in particular, non-commercial, open-source developers of Gnutella, Fasttrack, et. al. clients could quite plausibly argue that they had no intent to aid infringement. So I concede the point about this change not fixing the piracy problem, but I don’t think that that makes this adjustment to current doctrine an inane undertaking.