Should We Allow Copies of Analogue Objects?
I write a lot about copyright, and the right to share stuff. In particular, I think that for digital artefacts – text, music, video etc. - free software has shown us that there is no contradiction between allowing these to be copied freely and creating profitable businesses that are powered by that abundance. What has to change, though, is the nature of the business models that underlie them.
The parallel between digital content and software is obvious enough, which makes it relatively easy to see how media companies might function against a background of unrestricted sharing. But we are fast approaching the point where it is possible to make copies of *analogue* objects, using 3D printers like the open source RepRap system. This raises some interesting questions about what might be permitted in that situation if businesses are still to thrive.
On Open Enterprise blog.
Posts
9 comments:
Posting here as this comment seems to have gone over 1000 characters!
I suspect you're looking at the wrong products. Stick with me here...
The key is the use of a digital pattern for driving the 3d printer.
3d object scanning isn't relevant in the beginning --- that's the equivalent of creating a Beatles CD using the songs taped off the radio. It's sufficiently different from the original to not worry to original manufacturer.
Today we already have close to zero cost of production for most analogue goods (clothes, bags, tennis rackets, ironing boards etc. etc. etc.). Most of these things are not safety or even quality critical products. The manufacturers, wholesalers and retailers make money from the constraint of supply cost which they overcome by volume. High value low volume goods achieve that status through branding rather than superior content. The existence of copies is controlled by the fact that there are only a limited number of sources and these can be pursued and shut down. There are no easily used design patterns for reproduction.
Introduce industrial 3d printing and companies will start generating more and more products using 3d printers driven by digital design patterns --- market pressure will force this once 3d printing is cheaper than other production methods. At that point these digital patterns *will* exist. The only difference with the purely digital world is that for recorded music (say) the pattern *is* the product. But, it only takes one copy of the design pattern for an analogue product to get out onto the net and that won't matter.
Now introduce home 3d printing. If they can get the pattern people can print their own goods at home. All they will need is the downloaded instructions and raw materials. Quality will not be such a big issue. Did people fret over the quality of MP3 files being lower than CDs? The price difference is way bigger than the quality difference. For most products the quality will be good enough.
OK, so now try thinking the example through again, but substitute Aston Martin for Louis Vuitton handbags and Airbus for De Beers diamonds both of which I'm sure generate much higher profit margins. Everyone with a 3d printer can now have an LV handbag and a diamond as big as a golf ball for next to nothing.
The brand value of those goods will be massively degraded. Louis Vuitton and De Beers will claim intellectual property rights on their handbags and 'real' diamonds and howl loudest for legislation to make possession and sharing of their patterns for 3d printers ultra illegal. Where does this go from here...
Smart and motivated people will start to the reverse engineer the patterns for these analogue goods? DMCA extended to forbid this? Ordinary people jailed for being found in posession of a fake handbag? Is this coming?
@guy: lots of interesting thoughts. I'll post a longer reply to this soon...
I am an amateur futurologist. I enjoyed your article, but did not find your argument compelling, even though I expect eventually (on a scale of decades to centuries) analogue copying (or what Star Trek calls "replication") will necessarily be safe and free.
"What I liked about [the questions] is that they start from a very particular case and widen the scope more and more ..."
But not beyond the example of an Aston Martin, nor did you move (much) beyond that example either. Your argument seems to rest entirely on safety. As I look around my room at some simple manufactured obects - picture frames, a broom, bookcases, a vase, an empty pizza box (of course) - I see very few which would be dangerous if manufactured poorly.
While I agree that 3D printers (probaby based by then on grey-goo technology) will eventually evolve to the point where they can build a working car (or all of the parts for a working car), I don't agree with what you implicitly extrapolate will *at that time* be the cost and reliability of the build. In particular, to use your example, what happens when it becomes cheaper for Aston Martin themselves to "print" a car rather than "manufacture" it? Even assuming they continue to have the market advantages of exciting design and ingenious and reliable engineering, if replication ever becomes technologically and economically feasible, they will jump on it as quickly as the pirates. Then the "analogue original" will have a digital blueprint, in effect meaning that making an "analogue copy" will be equivalent to obtaining a digital copy of the blueprint (from The Pirate Bay), and then rendering it in an analogue form, just as we render digital music into analogue sound waves in order to use it.
I think a more fruitful paradigm would be the copying of board games. Some games, like Monopoly and Scrabble, can be programmed without loss of functionality, of course. But Mousetrap relies for its fun on the thrill of watching the analogue trap being built and sprung. I don't know RepRap's detailed spec, but it looks to me that it could make the plastic parts for Mousetrap today. (I remember seeing a 3D printer on Tomorrow's World in the '80s which used a tank of liquid hydrocarbons which could be "fixed" using laser light into plastic shapes, including the classic ball-in-a-cage; I wanted one so I could build Rubik Cube variants without woodworking skills.)
I think it's fairly safe to assume that the route from factory plastics manufacturing to home producation will follow the same price curve as printing and CD making have. Of course in the latter cases, the final step to a physical object has mostly been abandoned in a way that analogue copying never can be (well, not until we become brains in jars - I should get to patenting - or is it copyrighting? - those jars right now).
When that happens, there will be plenty of open-source chess-set blueprints to download. But people will also write open-source specs to print accurate facsimiles of intellectual property like Mousetrap, and then later steal and freely distribute actual digital blueprints.
I look forward to the opening of those cans of worms with giggling Schadenfreude.
@igblan: more great stuff...I'll have to have a think about all this - and then I'll post soemthing in reply.
Yup, that's it. A printer is a digital player like any other. For music you put air in, you get music out. With a printer it's paper and ink, a reprap uses plastic. It's the digital instructions that describe the output where the power lies --- and the property disputes. The recorded music industry (as traditionally structured) is just the first casualty of this.
As igblan says, companies will exploit this technology before consumers. They could well try to protect their digital designs (e.g. encrypt with some sort of machine key so that *only* their printers can manufacture the goods... until someone cracks it).
Board games are a fair example of things you *could* copy, but you probably wouldn't print your own copy of MouseTrap because it isn't disproportionately expensive to buy anyway. Digital music was disproportionately expensive (£16 CDs half full of tracks you didn't want). My guess is that it's designer fashion goods with brand inflated value that will suffer first... how about designer spectacle frames?
Glynn, your basic point is that 3D copies are not necessarily detrimental to incumbent producers because consumers want assurance about the provenance of the products they buy. An interesting argument. It is also a pure trademark argument in the sense that trademarks are there to allow consumers to ascertain the provenance of the product they are buying. That is what is protected by trademark, in order to reduce transaction costs. However, the designs of many products are trademarked themselves. Classical examples are the classic Coca-Cola bottle, but also the Porsche 911 design. In a world where you can make 3D copies (we are nowhere there yet, I can assure you as a Makerbot Cupcake CNC owner struggling to get it to work), you may end up with confusing consumers by the very design you are copying. And there I start having problems with your argument.
In 1995 I took a tour of a “lights out” printed circuit board (PCB) facility in Virginia owned by a joint venture of GE and FANUC. What was amazing was that circuit designers from all over GE could literally e-mail their designs (actually, electronic specifications of their designs) to the facility, with instructions on how many copies they needed. The facility would put the job in a queue. When the job reached the head of the queue production of the board would begin. Nine 3 foot by 3 foot circuit board blanks would be photo etched to lay down copper traces for however many circuit boards would fit (the facility was capable of “printing” any size circuit boards from 2 inches by 2 inches up to 3 feet by 3 feet). The 9 layers of circuit boards would be sandwiched together, bonded under a press. Then through holes, vias, and mounting points would be drilled by a robot. The boards would then be cut apart to specification by another robot. They would be picked up by robots which moved them to an assembly line of “pick and place” machines which would put all the electronic components on the boards. Next it was off to a wave solder machine and a cleaning station. A circuit board tester would verify that the assembly process was done correctly – if not the board was re-queued for production and the failing board discarded as scrap. Finally, the completed, tested boards were placed in electrostatic packaging and shipping boxes. A shipping label automatically applied and then the boxes were stacked for UPS. The board would be in the hands of the designer within 48 hours from the time the “submit” button was pushed.
They had a similar facility down the road which would make plastic and metal parts to specific mechanical specifications, using numerically controlled milling machines. I did not get a tour of that plant, but I’m told it was quite similar to the PCB plant.
15 years ago a designer at GE could literally “print out” 5 copies of a circuit board and the casing and be holding the physical products in a day or two. Marginal cost of these is quite low.
I think Guy has it right – the scanner is an irrelevant detail. Lots of physical goods today have a “source file” which allows any one of a number of factories to realize that electronic specification into an analogue item – perfectly.
I suspect that the Aston Martin and the A380 are both fully captured digitally in a mechanical CAD package as well.
Consider the Apple iPad – it has 4 major components, the software (an abundant good), the printed circuit board (whose design specification is an abundant good), the case (whose design specification is an abundant good) and a few custom integrated circuits on the circuit board (whose designs are also abundant goods). 3 of the 4 components could be reverse engineered quite easily. The custom ICs are very hard, but not impossible, to reverse engineer. The physical items are probably not even manufactured by Apple employees – they are likely subcontracted out. The custom ICs are quite likely second sourced – that is, built by 2 different companies in 2 different factories to assure that Apple does not create a monopoly supplier.
With respect to the iPad, Apple is merely an IP company.
You note in your blog that: Upon reflection – thanks in part to these questions – I don't believe that we ever will be able to make an *exact* reproduction of something like an Aston Martin (or even of a spoon, come to that) using 3D scanning and printing. What we will be able to do is to make a copy that is as close to an exact reproduction as you like – depending on how much money you want to spend (and obviously the closer you get, the more it costs.) Remember: these are analogue, not digital, copies.
What if they were digital copies? Does this change your conclusion? If one is able to get hold of the digital source for these physical items one can make an *exact* copy – well, for certain types of physical goods today.
@original: thanks for that fascinating anecdote. I'll try to answer your point when I write my post in reply to the others (which it relates to, obviously.)
@all: I've put up a post on CWUK trying to answer some of the excellent points you raise:
http://www.computerworlduk.com/community/blogs/index.cfm?blogid=14&entryid=2971
Post a Comment