When a printer is a gun

Scaramanga's career in hairdressing was short lived

You remember the scene in “man with the golden gun”? Christopher Lee, who plays the assassin, Scaramanga, is having a terse conversation with his corporate backer and whilst he talks he plays with his lighter, pen and cigarette case. The conversation draws to a close and Scaramanga, who has assembled his golden toys into a gun, resigns from corporate service in a brutally abrupt manner. A work of fiction and one with no small degree of panache perhaps, but a not quite a useful insight into the dangers of emergent technology. 

A couple of weeks ago Greater Manchester police, acting on a tip off, raided premises and confiscated a 3d printer and parts of a “printed gun” as part of their operation. The media ignored the rest of the operation and focused on the new technology angle. A printed gun. Of course the story is more complex, the police say they found a trigger and a magazine, the suspect says they are parts of the printer itself. The truth will no doubt come out at trial but it does raise some very interesting points. When does an item become forbidden? When does a design become dangerous? 

How do I recognise a dangerous design when I see one?

This question is fairly easy with modern parts, parts of a car and parts of a weapon are fairly easy to spot when they become dangerous, two springs are innocuous and potentially interchangeable, a seamless tube and a rifled barrel are much more easy to differentiate between and whilst a barrel could make a decent piece of ersatz tube, a piece of tube makes a poor barrel. The differentiator is purpose and quality. Outside of functional differences, the tube isn’t a great barrel because the tolerances employed to make it are lower than those used for the barrel. A barrel is a lot more expensive than the same length of tube as it’s a specific high quality item rather than a generic lower quality one.

The important differentiators are quality and specification. 

But these differentiators only apply in a world where mass produced items are made explicitly for purpose. In a world where parts can be mass produced to easily defined but exacting specifications, using 3d scan/3d print, those differentiators are meaningless. Quality and skill can be recorded and transmitted in a world where plans and specifications can be emailed. Tolerances and machining are down to the quality of the platform, the printer, not the person running it. If you can use Google and turn on a printer, you can print what you want.

What do you need to make a ranged weapon? A means of storing energy, a means to release that energy and a means to direct it towards its target with some degree of accuracy. Modern ranged weapons are expensive, and made to last. Does a weapon you can make yourself have to have the same durability? If you only need to make it work once, what can you make it look like? How modular can you make it and if you are a policeman searching a property, how do you recognise it? The forces of law and order are already reacting in a reasonable but predictable manner, focus on the plans, focus on the printer, focus on notions of DRM controls on parts (building controls into the printers at manufacture, like current limits on resolution to prevent you printing bank notes) and focus on the usual suspects. 

I think these approaches are limited and won’t stand up to the forces of innovation in a modular world. If you want to control forbidden designs, focus on the design not the manufacture. If you want to control manufacture, focus on intent not production. Think analytics and risk rather than blanket prohibition.

If any spring can be a trigger, any carton a magazine, any tube a barrel, any small pot a cartridge, I can’t ban them all. I need something telling me when someone has all four, in compatible scales, together at once. I need something alerting me to an increased risk of “weapon” not “food mixer”.

I need content analytics. 

I can imagine a near future scenario where police are equipped with Google glass like camera eye pieces, automatically counting and tabulating items as a place is searched, looking for form and scale, silently counting, fitting, comparing as the human being gets on with the human tasks. I can imagine a discrete alert once a certain risk factor is passed and the parts of a potential weapon are uncovered. I can imagine a similar capability monitoring the design flows through consumer internet connections looking for designs which could have multiple uses and watching where they are made and how often.

Ethically dubious of course, morally questionable too, but given the degree of monitoring already being undertaken, it can’t be long until someone starts implementing this, if they haven’t already.

Post script. Since I drafted this, another company in the US has demonstrated a fully printed pistol using an advanced powdered metal, laser sintering process. This technique isn’t something which can be undertaken at home and can’t be considered a DIY process… yet.