Fabbers: Mega-Damage by Micro-Patents

As my occasional postings on the subject indicate, one area that fascinates me is that of fabbers, aka 3D printers. One reason is that they effectively make our analogue world digital, in the sense that they allow 3D objects to be captured as digital representations, and then printed out. The other reason - a consequence of the first - is that by producing these representations, they transport many of the trickiest digital issues into the analogue domain. But more of that anon.

Of course, I'm not the only person to have noticed that the world of fabbers has the potential to bring the same kind of disruption to the analogue world that computers and the Internet have brought to the digital sphere. One benefit of this growing appreciation is that there are more resources available about 3D printers and related areas, and that they are becoming richer as the field grows.

A good example is a new report commissioned by the White House Office of Science and Technology Policy [freely available as pdf]. Indeed, I'd say that it probably represents the best single introduction to this whole field.

One virtue is comprehensiveness. As well as covering areas like fabbers (my own personal interest), it also discusses desktop milling machines; laser cutters and engravers; sewing and embroidering machines; and desktop circuit makers. It has a handy run-down of the main companies active in this area, including makers, aggregators and designers; and a detailed examination of the advantages and uses of desktop personal manufacturing devices. It is particular strong on the importance of educating the younger generation in these technologies - rightly so, since these will be the tools with which they will shape and build their world.

But for me, the most interesting part - because most problematic - is that which touches on the legal issues surrounding this rapprochement of the analogue and digital ways:

In our long tail world of media and information, files containing digital music, content, video, artwork, and data are easily copied. As a result, chasing down copyright violators of digital works has become as unproductive as chasing after a cloud of gnats. Preventing copyright and patent violations in the world of electronic blueprints and small-scale manufacturing machines will be equally challenging. The personal fabrication process spans both the digital and physical worlds and involves two components that involve intellectual property issues: the electronic blueprints and the resulting physical object. While one can protect a digital blueprint using digital rights management, this approach offers only a partial solution, since once the electronic blueprint is put to work fabricating physical objects, it can be used to produce as many objects as the maker wants to make. The resulting physical objects, unlike a music file, cannot be digital signed.

The report boldly suggests a way of handling some of these problems, encapsulated as one of its recommendations:

13. Explore micropatents as a smaller, simpler, and more agile unit of intellectual property

Here's its explanation of the concept:

An inventor would submit, for a few hundred dollars, a document describing their invention to a centralized government micro-patent repository. The document would be time-stamped and immediately publicly released, without having to be subject to the traditional tests of novelty, utility and non-obviousness. The inventor’s micro-patent application would claim very few fields of use, perhaps there could even be a mandatory limit on scope to qualify as a micro-patent. By filing this document, the inventor would immediately be granted an implicit, short-term (say 5 year) exclusive right to her new disclosed idea, as long as the idea was not already disclosed publicly earlier. Only in the case of alleged infringement, would intellectual property experts, lawyers, and the judicial system be brought to bear on the case with all the costs, time and complexities involved.

The key idea here is to offer "simple, agile and cost-effective intellectual property protection"; that is, making it easier to obtain patents, albeit lightweight ones. But in doing so, it will remove one of the few remaining barriers to patent applications, which inevitably will mean that every patent troll in the world will file thousands of trivial claims, since it will take so little effort or money to do so. It will give rise to the equivalent of patent spam.

Worse, these patent spammers will then proceed to sue huge numbers of inventors - and users - of objects made using fabbers. In fact it will become exactly like the world of copyright today, where tens of thousands of letters are sent out to alleged infringers, threatening to sue them but offering them a special "low-cost" way of settling.

Even more damaging, such a lightweight system will create a patent thicket around objects made with personal manufacturing systems that even nanotechnology will be unable to pierce. Again, we already have an all-too concrete example of what happens when it is easy to obtain patents for key ideas that are often indispensable for all users, in the world of software.

Before software could be patented, programmers wrote software by drawing on the commons of software techniques - and adding to that commons themselves. They didn't need to worry about "infringing" on someone's idea, because you couldn't get patents on ideas - just copyright on implementations of those ideas.

Now, with software patents being awarded in increasing numbers, things are so bad that it is probably impossible to write any non-trivial program without nominally infringing on someone's patent. That proliferation has led to dense software patent thickets, most notably in the world of mobile phones, where multiple companies are suing each other, wasting valuable resources that could have been devoted to creating more innovative products, not paying lawyers' bills.

The example of software shows us that patents simply do not work when applied to the digital realm. Indeed, the software industry spends far more money litigating software patents than it actually earns from licensing them. The fascinating book “Patent Failure: How Judges, Bureaucrats, and Lawyers Put Innovators at Risk” by James Bessen and Michael Meurer (Princeton University Press, 2008) quantifies this: the aggregate annual patent profits for software in the US were $100 million for the years 1996-9; the aggregate litigation costs for the same period were $3,880 million. There is no evidence that things are any better now.

It would be utter folly to import the worst features of the existing digital world into the new one that is emerging. Micro-patents will not foster progress and innovation: they will actually make things worse than the current situation, hard though that may be to believe.

The only solution is to have not "micro"-patents, but the limiting case where the size of the patent tends to zero - that is, none at all. Then, companies and inventors would compete not on the underlying ideas (which patents try to capture and monopolise), but on their *implementation* of them.

As well as avoiding patent gridlock, this also addresses issues of copying and counterfeiting, since people will pay more for otherwise identical products when they come provably from a trusted supplier, and also of safety, since it rewards better-quality products (not just patented ones).

As such, it's patently better than "solutions" based on intellectual monopolies that won't ever work (as proved by copyright and its attendant legislation, which have failed to stem the flood of shared music and video files in the slightest) and will actually lead to a net loss for companies forced to deploy them (as software patents and their thickets demonstrate). (Via Shapeways.)

Follow me @glynmoody on Twitter or identi.ca.