Last week, I attended MSU’s Fifth Annual Conference on Innovation and Communications Law, where I saw a wonderful presentation by Joshua Pearce, an engineering and material sciences professor from Michigan Tech, on “distributed open-source digital manufacturing” (a.k.a. open-source 3D printing). The hardware Joshua presented is called RepRap:
RepRap takes the form of a free desktop 3D printer capable of printing plastic objects. Since many parts of RepRap are made from plastic and RepRap prints those parts, RepRap self-replicates by making a kit of itself – a kit that anyone can assemble given time and materials. It also means that – if you’ve got a RepRap – you can print lots of useful stuff, and you can print another RepRap for a friend…
I love conferences that bring lawyers together with technologists, because they really help the lawyers among us understand what’s at stake for developers of new technologies that intersect—maybe “collide” is the better word—with intellectual property law. Joshua’s presentation ended with a plea to the lawyers in the room to prevent IP law from inhibiting the development and proliferation of open 3D printing technologies, which promise to revolutionize—maybe “disrupt” is the better word—our entrenched, centralized, and outsourced manufacturing model.
I don’t know what the patent landscape looks like for the machines and methods underlying 3D printing, but from the copyright perspective, distributors of 3D printers are almost certainly insulated from secondary liability for infringement by Sony v. Universal, a seminal Supreme Court case from 1983 holding that the manufacturers and distributors of reprographic technologies that can be used to infringe copyrights are exempt from liability for the infringements of end users if the technologies they distribute are “capable of substantial non-infringing uses.” The Supreme Court later clarified in MGM v. Grokster that Sony’s safe harbor will not protect distributors who operate with the proven intent to induce end users to infringe, even if the technologies at issue have substantial non-infringing uses. What Sony means for distributors of 3D printers is that they will not be liable for the copyright infringements of end users as long as they don’t encourage users to print copies of copyrighted works, including, for example, sculptures, toys, or useful articles of industrial design that incorporate copyrightable expressive elements. The technology itself is therefore safe from copyright law and from copyright injunctions prohibiting distribution.
The trickier question with 3D printers involves liability for direct infringement by end users. No one doubts that 3D printing will proliferate unauthorized copying of copyrighted works. The controversial question is what the law should do about that proliferation. The migration of 3D printing technology from the lab to the home is underway right now, and it will likely accelerate quickly given that RepRap is designed to be both low-cost and self-replicating. By happenstance, that migration coincides with the initiation of a comprehensive legislative review of U.S. copyright law, making this an opportune moment for policymakers to revisit the legal status of copying for personal use. Under our current system, copying for personal use arguably falls under the rubric of fair use, but that’s far from a foregone conclusion, even with respect to established technologies. The issue has been further complicated over the years by an increasingly expansive interpretation of what counts as a commercial (and therefore market-harming) use of a copyrighted work. Under that expansive interpretation, every unauthorized copy—even a copy made only for personal use—is viewed as commercial in nature because it theoretically represents a lost sale for the copyright owner. Restoring the eroded definitional boundary between copying for personal use and copying for commercial gain would be a step in the right direction for copyright reformers. For a lucid and still-very-timely consideration of how the law of copyrights should treat personal copying, see Jessica Litman’s article “Lawful Personal Use” from the Texas Law Review (2007).
The IP maximalists have to be stopped. It remains to be seen what it will take to stop them. Their absurdist claims of power to control people’s minds are discrediting the entire legal system, which is a very bad idea.
One thing that will probably cause the writing of some very bad law is the fact that much of the stuff that will be at issue won’t be copies in the sense that we usually think of them. Unless you have access to the original CAD (and or manufacturing files from which something was made) a 3D “copy” is going to result from a 3D scan or (more likely) a 3D reconstruction from one or more photographs, which then get turned into a point cloud, which then gets turned into a mesh or some other solid-like representation. We don’t usually think of photos of objects as “copies” of those objects.
But to prevent unauthorized reproductions of copyrighted/trademarked/otherwise-protected objects, you’re either going to have to prevent public display (e.g. Flickr) of multiple images containing that object somewhere in the frame, or else prevent the widespread use of tools for creating point clouds from multiple images.
Or go after the unauthorized reproductions the same ways that companies currently go after unauthorized reproductions of their physical products, which seems unlikely to satisfy anyone.
(There’s also the possibility of insisting that printers compare the object they’re about to print to some library of originals, but I don’t see that as terribly workable except for shutting down an entire industrial sector.)
People who understand copyright law are used to thinking of photos as copies (or derivative works) of the 3D objects they portray. That’s not to disagree with you about the making of bad law, however, because members of Congress are not necessarily people who understand either the nuances of copyright law or the new technologies that implicate it. Your point about how the technology will work in practice highlights the fact that the process of making a 3D copy of an object will often involve the making of several 2D copies along the line. I agree that building IP enforcement mechanisms into hardware is misguided. I think the past dozen years have taught us that DRM for digital goods hasn’t been a great success in preventing infringement and has caused headaches for lawful purchasers of locked copyrighted content.
Ah, I really hadn’t thought about “derivative work”. But I think that these issues are going to get a lot worse. We’ve already seen people getting dinged for video “pirating” when copyrighted moving images appear in the background of their personal video clips. If the IP maximalists get their way, that kind of protection is going to crop up to control any set of images from which a point cloud could be derived. Now think about most people’s immediate environments and how far you can look in any direction without seeing a copyrighted/trademarked object.
“The technology itself is therefore safe from copyright law and from copyright injunctions prohibiting distribution.”
_Existing_ copyright law. I’m no lawyer but Sony-vs-Universal reads as saying Congress _hadn’t_ outlawed tech which has substantially noninfringing uses, not that Congress _couldn’t_ do so. (And DMCA’s circumvention prohibitions certainly haven’t been struck down.) It seems like there isn’t a year that goes by anymore, where Congress doesn’t at least _threaten_ to do something shockingly radical, so I cringe when I see the word “safe” used like that.
What an awful time it must be, to try to plan a tech business. Who _knows_ what’ll be illegal 4 quarters from now?
Very interesting – I wonder how 3D printing and IP will play out in court.
What do you think of the idea of using algorithms to protect the public domain by going after obvious IP see: http://www.thingiverse.com/thing:73427
Any road to higher quality patents is a good road to follow. It’s hard for examiners to know everything that’s out there in terms of prior art when they examine a patent application, yet it’s critical to the integrity of the system that patents not be granted for inventions that aren’t truly novel and nonobvious. If the wisdom of the technical crowd will help get us there (see Peer to Patent), then we should crowdsource the search for prior art and obvious combinations of prior art. If algorithms will help gut us there, then we should automate that search.