Some more thoughts on controlling the spread of dangerous information online

Was it appropriate of the New South Wales parliament in Australia to recently decide to ban the possession of blueprints for the digital manufacture of firearms (using a 3D printer or an electronic milling machine, as depicted left)? In their recent entry for this blog, Wouter Tebbens and Marco Fioretti forcefully argue that this measure is unlikely to have much effect, considering how difficult it is to keep files of any kind off the internet when there is a real demand for them. They also point out that determined people can already build their own untraceable guns without using digital blueprints, by following instructions posted online or by attending “build parties” where the required equipment is made available to them to build their guns, using various parts that can be purchased legally without the need for any background checks (see for example this testimony).

There is no doubt that the effects, up to this day, of the measures that have been taken to stop the spread of digital gun blueprints are not particularly encouraging. In 2013, after Defense Distributed, the makers of the 3D-printed Liberator pistol, had to remove the blueprints for the gun from their website (where they had been downloaded more than 100’000 times in the space of just two days) at the request of the US State Department, these files quickly resurfaced on the Pirate Bay, where they can still be found. Even more disturbingly, the Pirate Bay also hosts blueprints for more powerful and dangerous guns such as the AR-15, a semiautomatic rifle (the Liberator, by contrast, is a single-shot handgun).

Despite these poor omens, however, I am still not convinced that measures meant to regulate the possession and dissemination of digital gun designs must necessarily fail to have any interesting impact. Importantly, this will depend on the number of people who might illegally own such files or upload them to the internet at any point in time. Tebbens and Fioretti point to the failure of the war on file-sharing. However, it is worth emphasizing what a massive hydra the copyright industry has been dealing with in that case: speaking solely of music files, an estimated 40 billion were illegally shared in 2008, according to a report by the International Federation of the Phonographic Industry (IFPI). It seems unlikely that the illegal sharing of digital gun blueprints would come even close to such numbers, at least as long as it remains confined to its current target demographic. Luckily, many more people seek to access music, software and film than seek to acquire untraceable weapons.

To take now a different comparison point, the Center for Problem-Oriented Policing states that “at any one time there are estimated to be more than one million pornographic images of children on the Internet, with 200 new images posted daily”. These are obviously shocking numbers; still, they are clearly not of the same magnitude (fortunately) as those I have just cited about the illegal sharing of musical files. Partly as a result of this, while we can agree that it is not possible to completely eliminate child pornography from the internet, regulatory measures aimed at preventing such material from being produced and distributed are nevertheless worth implementing. As I mentioned in a previous blog entry, the fight against child pornography has arguably managed to reduce the amount of such files available online (and made the remaining ones more challenging to access for prospective consumers), has led to some perpetrators being arrested and charged, and provides some degree of deterrence. (Despite boasting a no-censorship policy, the Pirate Bay itself removes files containing such material when they happen to get shared through the site – at least when faced with pressure from the authorities.) Why couldn’t the same be true of measures (like the NSW ban) designed to exercise some control over the possession and distribution of blueprints for digitally manufacturing weapons?

To clarify, my purpose in using this comparison is not to suggest that sharing digital gun blueprints online is ethically as objectionable as uploading files depicting child abuse. I merely wish to indicate reasons for cautious optimism about the viability of regulatory measures concerning the former. Furthermore, I certainly agree with Tebbens and Fioretti that such measures could raise privacy concerns, depending on how they were implemented. The same is actually true of the fight against child pornography. Yet opposing unacceptable intrusions into the private sphere (e.g. in the absence of any incriminating evidence) on the part of law enforcement agencies does not, in either case, entail that we should oppose all legal prohibitions on the possession and distribution of the incriminated material. A measured application of such prohibitions seems preferable.

Admittedly, this proposal raises a number of further issues. For instance, as Fioretti and Tebbens point out, there are already a number of websites containing instructions for building DIY weapons without using digital devices. It might be argued that banning the online distribution of digital gun blueprints (outside of restricted circles accessible only to people with the relevant permit) would commit us to also banning such webpages – something that might seem both excessive and unworkable. Properly addressing this issue would require a separate blog post; for the time being, let me just note that even if there are similar grounds for regulation in both cases, figuring out how to control the spread of digital blueprints appears to be a more pressing matter, as digital manufacturing already allows people with little to no technical know-how to build fully functional guns. Building firearms the traditional DIY way, on the other hand, is much more technically challenging, and the resulting weapons are often unreliable, posing as much of a threat to the user as to his target. This potentially gives DiDIY weapons much greater appeal than DIY weapons for people – including those with evil intent – wishing to avoid registration and background checks.

At this point in time, such specific worries about digitally manufactured firearms may seem premature. For example, the vast majority of the perpetrators of recent mass shootings in the United States obtained their guns legally and passed federal background checks, despite the fact that some of them had documented mental health problems or criminal histories. However, as digital devices enabling the manufacture of items like firearms continue to improve and start spreading across the world, it would seem desirable to act preventively. While I do agree with Tebbens and Fioretti that regulatory measures aimed at controlling the possession and distribution of digital blueprints will not solve the problem on their own, it still seems to me that they could be part of the solution – depending, again, on what the size of the challenge turns out to be.