Friday, February 17, 2006


Reprap Musings

Imagine a world where reprap has caught on, say 20XX. Substrate materials have become extremely common. Due to economies of scale, its dirt cheap. Infact, some people toss out broken projects, opting to print out a newer version instead.

Then, one could create a reprap project that was essentially a scavenger. It roams around, searching for consumables. It would gather enough material to make a copy of itself, then replicate. If done in the correct fashion, it would be an amazingly efficient recycling program. Mouse size robots scurring around, searching for bits that can be recycled, then taking them back to home base. They could use accurate positioning and networking to pinpoint where sources of substrate are, then devour it, ant style.

Since you would only need a finite number of mice to cover a given area, there would undoubtedly be an overstock of workable material. Combine that with the robots being able to recycle most of their parts, it would be very self sustaining. Given the open nature of the project, it wouldn't be hard to imagine a benevolent individual offering free or low cost prints of reprap machines, or any other useful device. That would be very cool.

Obviously, there would be need to be certain guards in place. Ownership of resources would come into play, and also safety of the public. There are currently laws that cover these things. They carry serious punishment if you break them. If your mousebots eat some guys car, he will probably sue you. Likewise, if they eat the actual guy, you will definitely end up behind bars or worse. Even in the extreme event of someone releasing a intentionally malvolent replicating machine, there would be a hundred times as many normal, smart people out there who enjoy not being eaten or killed by robots and could come up with a way to disable it.

You're talking about the macroscale equivalent of grey goo.

This is whay I think it behooves makers of exponential manufacturing devices to include ways for users to recycle their stuff rather than throwing it out. If it represents a modicum of value, it won't end up on the curb, where free-range bots could get at it.
I've spent a while musing about the future of this technology as well. At present, a (very) rough symbiotic balance could probably be said to exist between material producers (miners, etc) and manufacturers - minders are paid ultimately in products made from the resources they dig up, and the manufacturers are paid with raw materials to make more. In a world where RepRap technology has taken over, manufacturing entities would effectively be decimated, if not actually destroyed (it will take a while for fabricators to evolve to the point that you could make, say semiconductors on a small scale at home, though I suppose thermionics might turn out to be dead easy), and of course a miner would be able to pour his hard-earned resources into his own fabricator if he wanted. Clearly the current exchange system could not exist, but what could go in its place?

The issue of ownership is a difficult one. Personally, I rather feel that someone can only truly be said to own something that he created himself - if I design a new machine and build it, I own it. But what if someone else builds the same machine to my design? Do they own it, or do I? Working from the hacker saying "Information wants to be free", and the general observation that useful information does maximum good to a society when it is as widely distributed as possible, it would be socially irresponsible to restrict use of an idea, say through a patent - charging more than its worth could be said to be tantamount to holding society's future to ransom, but how could one actually derive its true value? Personally I don't really trust the almighty "Invisible Hand". Also, with anyone able to make their own copy of something, most patents would be entirely unenforcable, in rather the same way as P2P has made copyright infringement difficult to control - it's easy to track a large wave of unauthroized product to a single factory, but virtually impossible to pin down every individual making one copy - and even if you could, would you arrest a significant swathe of the population?

Returning to ownership of produce, can any person truly be said (or perhaps the right word is "allowed") to "own" land, or more exactly the resources in it? Anyone with the right equipment could theoretically extract resources from it, and from the description above be said to own them, but conventional law & thinking says that he does not own the lump of ore or whatever he dug up, but the landowner does.

I'm not proposing a solution here, I'm just raising a few questions that bother me and, I hope, furthering discussion. I'd write more, but I need to get back to work now.
This concept is basically what I had in mind when I posted this entry ( ) on my blog. Devices with a distinct purpose. They could scavenge an area, retrieve materials, deliver to a localized Processor, then use the converted material... for fuel, creating things like homes, whatever. In effect, they generate a "subGoo".

After a structure is completed, they cannibalize each other until only a few remain to ensure the structure is maintained, for recycling, to replicate if necessary. The Processor remains the "heart" of the system. It's very much like a termite colony with humans involved in a symbiotic relationship with the machines.

wrt IP ownership, imo, like it or not, it's history.
I don't see reprap having time to evolve to this level of sophistication before we have full blown molecular nanotechnology.

But, Tom, there is an answer to your concern, and it's the use of "proprietary" rather than patented technology. You have to have quite a bit of information about something to duplicate it, and getting that information out of an exemplar isn't necessarilly the easiest thing in the world, especially if it's made with the intent that you NOT be able to get it out.
***I don't see reprap having time to evolve to this level of sophistication before we have full blown molecular nanotechnology.***

You never know. It might or it might not. I've been waiting all my life for fusion power.

I really like the idea and promise of nano. That said the pieces that are necessary to make it work just aren't there yet. Build-time problems kill nano in all but a very few commercial applications.

My son's university, UC Santa Barbara, is a good example. They have kept a handfull of Nobel Prize winning physics types on their staff for decades and have a world-class reputation in physics. They're also building a rather substantial building to house JUST their nano work. My son has been snooping around looking for where he can get his foot in the door over there. The building is not done and it isn't clear at all who's going to be working there when it is. He hasn't been able to locate any professors who are committed full-time to nano at the university.

In the science park zone outside the university there aren't any nano startups, either.

I like nano, but I wonder if I'm going to live to see it happen. I'm not that close to dying, either. :-S
I think that, if you want to have an idea what will happen, you have to look at things game-theoretically and ask, "What is the evolutionarily-stable strategy, here?"

Take the concept of ownership. At the moment a person can own two types of things: a material object like a cat or a truck, or an idea, like a patent or a trademark. But the range of material objects that one can own is limited by practicality - you can't own Pluto, or the third oxygen molecule from the left at the base of the Great Pyramid. And owning an idea has always been a bit dodgy, as a simple though experiment demonstrates: imagine going up to a stranger and asking for a cheque for $1,000. You would expect a brief answer in the negative. Now imagine going up to a stranger and asking them to tell you their most interesting idea. You would expect to be locked there half an hour later concocting ficticious urgent appointments.

People don't instinctively think of ideas as property, whereas the youngest child will yell when they are playing with something and it is taken from their hands.

The reason, of course, is that matter is zero-sum, which is why we have a conservation law for it at the very foundation of science. Ideas are not.

The big trick of life is to make matter as close to being like ideas in this regard as possible, with universal exponential growth. RepRap should do the same for artificial products.

I've said it before, but our oldest industry teaches us what we need to know about these systems: agriculture has always dealt with self-replicating exponentially-growing machines. And it has always re-designed them using selective breeding. RepRap will just make manufacturing like agriculture.

"RepRap will just make manufacturing like agriculture."

Especially if it is eventually capable of using organic feedstocks as the thermoplastic. THAT would be amazing.
I'm out of my field here, but I seem to recall hearing that just about any economic system there has ever been generally doesn't quite know how to handle things like agriculture...
"You never know. It might or it might not. I've been waiting all my life for fusion power."

The military have working fusion power. And on a mighty big scale, I might add. That's the chief reason we don't have civilian fusion power: The only proven route to it has been foreclosed for non-technical reasons.

I saw it in college, a quarter century ago: An engineering study on a fusion power plant based on small nuclear explosives. Looked quite feasible, though to be sure you wouldn't need many of them to supply the nation's juice, even now. LOL
I think it would be pretty trivial to make incentives for people recycling, I mean, these ideas essentially say that garbage is just as valuable as regular matter. No one is going to want to throw anything out because it can be useful! So I don't think this will be much of a problem, I think resources would be readily available and no one is going to have to go around thinking up novel ideas to get feedstock for their machine. There simply is just so much that you as an individual can use.
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to
Posts [Atom]