A Trade-Off?

In light of the commenter referenced below, perhaps the Madster ruling should be seen in terms of this trade-off, starting from where we were with Grokster and (how I view) Napster:


Knowledge with an ability to act is no longer enough, because you still have to balance infringing and non-infringing uses. The analysis requires a cost-benefit analysis that holds non-infringing uses in high regard. You also have to consider what affirmative conduct was taken to advance infringement.


At the same time, if the infringing uses are substantial, and you can limit those uses without much burden, you will have to reengineer your technology. Even if you have no ability to act today, we can make you create an ability to act tomorrow.

Contributory and 512

(UPDATED 3:00, 7:25 PM) A comment to my second Madster post below makes some very valid points. Particularly, he brings up the importance of Madster’s affirmative conduct in the case and in Posner’s standard.


I’m not sure this is grossly different from Napster, where that affirmative conduct was also key. The difference is that, here, even having the site and facilities might not be enough if you didn’t affirmatively assist others in infringement. That’s going even further than Grokster.


But, what about DMCA 512?

Update: Color me wrong and sloppy. The commenter gently reminds me that DMCA 512 is, duh, a safe harbor, a limitation on liability, not a liability standard in itself. My bad. Who is this commenter? Somebody get him a weblog.


Random thought: when would a court interpret forfeiting one’s safe harbor as accepting liability? (Here, I’m reminded of Rehnquist’s opinion in Bush v. Gore, where he said that Congress wrote a safe harbor to mean a drop dead deadline, not a safe harbor.  So, in DMCA 512, saying what avails one of liability would actually define liability itself. Dubious reasoning, I think, in both cases.)

Refining Thoughts on Madster

Slept on it, and my fears have moderated a little. Some of this may be redundant – I’m just trying to talk out my thoughts as I analyze the case.

James Grimmelmann has a good post on the topic.  He points out how Posner does provide a blueprint for people who want to argue substantial non-infringing uses. Someone also added a helpful comment at Lawmeme about how this may be a step up from the Napster standard.


I think it’s important to recognize, as those two do, that Posner appears to chip away at one of the most harmful parts of the Napster opinion.  If knowledge alone is enough to trigger contributory liability, even people like Morpheus are in danger. Knowledge of one infringing use isn’t enough. From what I can tell, Posner means that you have to know about a substantial number of infringing uses, and you have to know about them when getting rid of them would not be too burdensome.


So here’s where my fears might be unfounded. If you read Napster to mean that contributory liability rested only on the knowledge standard, Posner is helping out, pushing us closer to the Grokster standard. The “cabined within the architecture of the system” language only exists in the vicarious liability section of Napster, and you could read the passing references to material contribution and ability to purge material in the contributory section as irrelevant. (Posner almost collapses the distinction between the two doctrines, which seems rather reasonable.)

Moreover, even if you don’t read Napster that way (and I don’t think you should, particularly in light of Grokster), you could read Posner’s decision as ultimately equalling the Grokster and Napster standards. That isn’t such a bad thing in that it aligns it with Grokster. It simply emphasizes the point that knowledge alone is not enough.

So my only real worry, then, is that you can read Posner’s decision as meaning, if it’s not costly, you should have to completely reengineer your system if doing so will prevent infringing uses. Posner didn’t do this in the language of Grokster, which focused on whether Grokster and Morpheus could act on their knowledge of infringing uses at that particular moment given their current systems’ designs.


Here’s what counters that fear: the fact that Posner doesn’t think AIM would be at risk, even if it used encryption for privacy purposes. If AIM, which has access to its servers, wouldn’t have to do anything, Morpheus probably wouldn’t have to do anything. Probably.


I still worry about these vague standards. The law is all about line drawing, and some lines I guess are better than no lines. But the fact that so much hinges on a case-by-case cost-benefit analysis means that this will continue to be a problematic legal issue. Again, think about what Professor Lessig said before the argument: “If innovators have to guess how a federal court will balance the effects of its technology upon copyright interests, then there will be less innovation. Thus the genius of its rule was that it said essentially this: if your technology is “capable” of a “substantial noninfringing use,” then the question of whether your technology should be permitted is no longer a judicial question.”

The extreme version of Sony is long dead, but that doesn’t mean we should forget the “genius” of it, that it created a framework where innovators could innovate without being worried about what they could be sued for.

More on Madster

I have a feeling this case might occupy my thoughts for a little while.  The more I think about it, the more I read it, the more confused I am.  This has been the theme of the case, so I’m not surprised.  Maybe I’ll change my mind about the following, but, right now, here’s how I’m feeling about this result.


I knew the facts went against Madster, so I was hoping that Posner would at least get the theory right.


In some ways, he adds to Sony.  For instance, he notes that, counter Napster, specific knowledge (and, presumably, material contribution) does not trigger liability.  Substantial non-infringing uses are still important.  But he clarifies this by asserting that, when there are substantial infringing uses and you can control those uses without gross burden, you’re a contributory infringer.  So is this a meaningful distinction?


And what in the hell is he talking about in that sentence I put in bold in the post below.  Really – what does it mean for a feature to add value to the system? What would be a disproportionate cost?  Think about the VCR and fast forwarding. Let’s say Posner’s right and fast-forwarding or using pause to skip commercials does create a derivative, infringing work.  Couldn’t you take that feature out without substantial burden and without harming the substantial non-infringing uses? Well, you can get out of that by saying that fast-forwarding and pausing add value to the equipment in non-infringing ways. 


You can play this out with P2P too.  When I think of Napster, I think Posner’s statement is consistent with that case.  But when I broaden it to Morpheus, I’m not sure.  It depends how you balance the all these burdens and benefits.


And more importantly, it’s how you balance the infringing and non-infringing uses. That is what stands out to me in this opinion.  Posner’s focus is on a cost-benefit analysis that compares infringing and non-infringing uses.  Look at how he talks on page 13.  His assessment gives weight to substantial non-infringing uses but under no circumstances stops there.

The sentiment that all inquiry ended with a finding of substantial non-infringing uses faded in Napster.  There, it turned on knowledge.  Here, it turns on an amorphous cost-benefit analysis.  It sounds a lot like Napster, the result is a lot like Napster – I’m not sure if anything’s really that different.  And, in that way, it doesn’t feel like this decision is terribly bad.

But here’s the thing: all he’s done is added more vague language that’s going to get someone in serious trouble.  He’s taken one vague standard and replaced it with another, perhaps vaguer standard.  Someone’s going to take this and apply it to Morpheus.

And, once again, I wish that these facts had not been litigated.  Posner knew that the encryption scheme was, in this case, an attempt to put one over on the court.  I must commend him for trying to distinguish Madster’s use from AIM’s or another service’s use.  I know he was trying to draw some firm distinctions.  But, in the end, I think the facts of the case led to some really messy principles.