evolution

hi all.

for the evolution analogy to realise its real power, is it better that all repraps are slightly different, so that the phyla can diverge into different species, the more successful of which will eventually dominate.

if the reprap is right now at the bacterial level of evolution, it's only job should be to make more repraps.

but as it becomes more numerous, as an organism, it should start to give back to us something we need to survive, otherwise, why would we bother to nurture it?

whether that killer ap is safe drinking water for the third world, or cheap solar cells for the first one remains to be seen.

perhaps those two things will be produced by two different species of reprap, existing in different environmental niches.

thoughts anybody?


dunc

It's a good idea, and I think it's happening already. There are a number of different variations on each of the key parts of the system, such as the Cartesian positioner, the extruder, the control electronics, and the design software. Most importantly I see a small but steady growth in experiments with new types of materials, as this is (IMNSHO) the real key to everything.

I agree with Colin (surprise!) there is lots of "mutation" in the mix right now. In fact if we have a problem it is the lack of a solid, readily available, benchmarked reference version. While this is probably a function of the current pre-alpha state of the platform, but to the best of my knowledge no two people currently have working comparable repraps. I hope the availability of cast parts from Bi&By will help alleviate this situation but it is currently the case that we can't really say what an "average" RepRap can do.

This is one of the reasons I would like to create an accurate CAD model. That's not as good as an actual model that is The Official Reference RepRap, but it will go a long ways towards organizing and categorizing the natural "mutations" that builders will introduce.

"we can't really say what an "average" RepRap can do."

true,
but with adrians "water filter", i think we can see what a "successful" darwin should be able to do.

there will be machines unable to pull this feat off, and unless they can do something else useful for us, they will quite rightly die off in an evolutionary cul de sac.

perhaps it would be useful for the website to proliferate into other, separate websites, in different languages, specific to different parts of the globes cultures, needs, and local areas of expertise.

if, eventually, an indian car mechanic could perhaps design a third or forth generation darwin, which could deal with its own electronics, AND help repair cars, i would imagine such a machine would spread throughout the third world far faster than any charity backed medical giveaway machine, no matter how amazing or useful.

in another part of the world, you might find a big reprap making lightweight composite building materials out of recycled paper. it can't make it's own electronics, but exists in symbiosis with it's circuit board printing little brother. especially if it can make more reprap chassis.

as colin says, materials may be everything.

possibly the core specification of repraps is something as fundamental to us as "quadraped's are succesful", and "central nervous systems are succesful" are to the animal kingdom. from those common traits, there still arises huge diversity.

at the moment, i'd say adrians darwin looks succesful to me, and as such is worth copying, seeing as i'm just starting out.

thanks for the replies.

dunc

What I think is most important is being able to take a part designed and built on one RepRap, and build a copy of it on any other, and secondarily, modify the design and make copies of that.

What I haven't really thought through is how much materials are implicit in design and control. It makes sense to build a mechanical part in CAPA (for design prototyping), in wax (for investment casting), and laser-sintered metal (for direct manufacture, if your machine can do it), or even to drive a milling machine. PCBs can be subtractively etched chemically or mechanically, or perhaps manufactured via a deposition process.

What I do know from my own very limited experience with CAD-CAM-CNC machining is that somewhere in the chain, the limitations and capabilities of the machine tool start to influence the process very heavily. Because I am an all-in-one shop, I design things with the limitations of my machinery in mind, and because my machines are relatively small and weak, this frontier comes pretty early. It's probably too early to need to think about this, but some sort of way to categorize and standardize machine capabilities might come to play a very important role in making the real source code--the stuff RepRaps build--more successfully cross-platform.

I guess that if it needs us it's more like a virus with our hands taking the role of the ribosomes and us being the host cells. While viruses sound like nothing you want to associate RepRaps with, not all of them are bad, they're part of our world. We have the DNA for many virusses, called "endogenous retrovirus" and they seem to play an important role in our evolution. Calling them good or bad is oversimplified, they shape development. Read Greg Bear's great book "Darwin's Children" for some fascinating fiction about this.
[www.amazon.com]

It would be stupid to claim that any virus is bad. I see RepRap as a tool that scales well, that creates solutions to local problems and very important to innovation and creativity of humankind.

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]