Framing Openness

So I want to pick up on Phil’s Openness Uber Alles post from yesterday. I am not sure that I understand exactly what openness means. Phil, how do you see that as different from a consent system?

So the 1996 Telecommunications Act turns local telephone networks into a partially open system by imposing a mandatory access regime. If at the end of this mob, this group decides it wants to open a local telephone company, the 1996 Act helps. I don’t need to ask the phone company’s permission to use various features of their network, as the law requires that they give me access (for a fee of course, though the current phone companies think that fee is way too low).

Now you might say that this isn’t an open system, for even though we wouldn’t need the phone company’s consent, I do need them to participate in making the legally-required access work, and if I don’t like how they are living up to their legal obligations regarding access, I will have to get the government to jump in to enforce (as occurred in the Trinko case). So is the absence of consent a necessary part of openness but not sufficient?

Now switch to Grokster and do that in pieces. So, from the perspective of the 15-year old teenager—I should confess that I write as the father of one—a laptop computer is an open system, as he can add whatever programs that run on his system. He doesn’t need to get Dell’s consent to add any application program, and, to flip that around, Dell hasn’t tried to control what programs are added (parental consent is a more complex matter!).

From the perspective of the creators of Grokster, the Internet protocol is an open system. (Tim Wu has a nice description of all of this in his Application-Centered Internet Analysis paper.) The system is designed to facilitate the introduction of new applications—the end-to-end principle—and again, it is not a consent-based system: Grokster doesn’t need to ask Vin Cerf’s permission to take advantage of the protocol.

But now we have to turn to copyright. One of the key things that US copyright law does is to define when I need to get consent to use someone else’s copyrighted work—precisely or not so precisely given the uncertainties of the fair use doctrine. Unlike mandatory access of the phone system, mandatory access of a copyrighted work is typically self-enforcing: I already have access to the work and don’t need the copyright owner’s permission to get access (obviously, the Digital Millennium Copyright Act puts interesting pressure on that regime, but that is an issue for another post (maybe even another mob?)).

But if access isn’t the issue in copyright, consent is. So Grokster emerges. Open from the perspective of the 15-year old and his laptop; open from the perspective of the designers of the Internet protocol. How open is it from copyright’s perspective?

This is precisely the issue at stake in the Grokster case and in Sony’s famous substantial noninfringing use test. That test creates what I have called elsewhere an innovation safe harbor: meet the test as an innovator and you are assured—sometimes after litigation—that you don’t face secondary copyright liability. I think that there are problems with the Sony test (here) and that we should modify the test to do better (here), but at its core, the standard we set in copyright for nonconsensual use of copyrighted works is about one type of openness.

I am comfortably in the camp of those who believe that an author should have some rights associated with her work and indeed don’t believe that even most of the copyright left favor eliminating all of the attributes associated with copyright.

In that sense, at least as to copyright, defending some level of closedness—is that the opposite of openness?—should be straightforward. Phil: Yes?