Because it's a solvable problem.

There seem to be a couple of problems inherent in the tech field of prosthetic design. First and foremost of them is that comparatively few people need artificial limbs, so not enough of them are manufactured at once to bring the cost down. A second problem is that because so few people tend to need them, designs don't seem to improve very rapidly. When enough of anything are not constructed, there isn't enough pressure for bugs to be ironed out rapidly, nor for designs to evolve in positive directions so relatively simple advances may not appear soon. Business and industry meet science and technology, what can I say?

Which brings me right along to the phenomenon of curious hackers tinkering because they enjoy it. Easton LaChappelle has been building robotic manipulators since the age of 14 for the sheer fun of it. Lacking a background in software development or engineering, he taught himself what he needed to know and won third place in the Colorado State Science Fair in 2011 with the first version of his design. A chance meeting with someone at last year's science fair caused him to blaze the trail still farther: Top of the line prosthetic limbs cost tens of thousands of dollars but they don't grow as the body does so they need to be replaced at full (or highter cost). So, Easton designed a third iteration of his robotic arm, printed as many parts as was feasible on a 3D printer, and used off the shelf components for the rest. Total cost of the prosthetic he's designed? $250us. Easton's v3 is said to be amazingly capable for an artificial limb, and to facilitate that he built a novel user interface system that seems dauntless at first, until one considers the learning curve of a pre-paid smartphone from the grocery store. A consumer EEG is used to actuate functions and control movement of the arm (hey, don't look at the screen that way, they're out there and a lot of fun to hack around with...), and flexing muscles in other parts of the body selects gross movements (such as flexing the elbow and turning the wrist). NASA, impressed by his work, has tapped him to intern on the Robonaut project headquartered at the Johnson Space Center.

A few states over, five year old Liam Dippenaar was born with ambiotic band syndrome: The fingers of his right hand never developed. Two hackers, one in the state of Washington and the other in South Africa collaborated over the Internet and designed an open source prosthetic hand for the youngster that can be manufactured on a 3D printer and assembled with off-the-shelf parts. Using Skype to communicate, an open source CAD package called OpenSCAD, and a pair of Makerbot Replicator2 3D printers they were able to shorten their design-fabricate-test-debug cycle to 20 minutes per iteration, which I think might set a record for hardware development practices. Anyway, Liam's mother contacted the duo after learning of their project blog, and they sent Liam a prosthetic hand at no cost sized appropriate to the youngster. If you are interested in helping to fund their research they have a page at Fundly where you can contribute. Also, to help as many people as possible around the world they've posted their designs to Thingiverse under a Public Domain license, meaning that anyone and everyone is free to download the designs and print them, modify them, use them, sell them, and contribute their improvements back to the community if they wish.

Just in case Thingiverse is blocked by overzealous web filtering software in your area, I've mirrored the files here (20130302) so they will be another copy floating around Out There Somewhere, should someone need it.

One of the problems that isn't easy to solve about space travel is carrying enough supplies with you. It's all well and good to have a small stock of spare components and tools when setting up a colony of any kind (doubly so when it's a proposed lunar colony) but mass is hard to move and fuel is a concern. So, how would hypothetical lunar colonists manufacture replacement structural components or tools without waiting days or weeks (because space travel is anything but agile)? At Washington State University a small group of researchers studying 3D printing published a paper on this problem in a journal of rapid prototyping technologies. As the name implies, they've worked out a way in which it might be possible which involves using industrial lasers and lunar regolith as feedstock. Regolith, which most of us think of as "moon dust" contains significant amounts of metallic iron and aluminum which could be easily smelted and shaped. Lunar regolith is very well studied, so much so that NASA has closely replicated it in bulk for experimentation, so the research team was able to get their hands on the simulant and try the laser printing technique they'd developed. Lo and behold, they were able to fab simple shapes out of the smelted dust. The parts aren't said to be very pretty but as a proof of concept seem solid. NASA is testing them at this time.

And now, for the obligatory bit about 3D printed weapons, because everybody else is doing it.

It's been said numerous times that the descent of the human race from simians has resulted in a fairly simple algorithm being hardcoded into their DNA. It goes a little something like this:

Can I eat it?
Can I fuck it?
Should I kill it?

This algorithm seems to be so deeply embedded that it somehow winds up in everything created by the human race, case in point, the humble newspaper. So, it came as no surprise when 3D printing enthusiasts started fabbing weapons components. The Wikiweapons crew continues to advance the state of the art in 3D printed rifle components by testing and releasing meshes of an AR-15 receiver that successfully fired over six hundred rounds of .223 ammunition before ending the test because they ran out of bullets. They're also perfecting and releasing meshes for magazines for many different firearms which have been tested to hundreds of rounds each.

Yay, domesticated primates for deciding that killing is a better alternative than fixing things.