I’ve been thinking a lot about space travel recently, particularly the recurring theme in the sci-fi world of ’seed ships’, ships that take hundreds of years to reach their destination, carrying everything they need to start a culture from scratch. They locate a habitable world first, then load up and go.
There is this incredible (and real) device called a 3-D printer that allows you to build parts, any part, from plastic or metal dust.  There are several ways you do this, but my favorite is where they lay some dust down, then shine lasers from different angles to intersect at the spot of dust they want to melt.  The liquid dries quickly, but by sprinkling down layer after layer of dust, and melting into liquid only the parts they want, they can ‘extrude’, or melt into being, an entire part.
I was thinking about the fact that you are going to want to replace ANY part on the spacecraft at will during the hundred year voyage to your new home, so a great idea would be to make every part on the spaceship out of material that can be used in a 3-D printer, and store digital copies of every part you make the spaceship out of.  Then, as long as you take enough metal powder with you, you could remake the entire spaceship!  And if you were really clever, you could plan on stopping by asteroids on the way made up of the metal type you use.
The trick would be how to design each part so that they could be made piecemeal in the (relatively small) printer and then assembled within the spacecraft.  Also, electronics are normally made from HUNDREDS of different materials, which would be too complex.  You would need to rethink all of your electronic and mechanical design to limit the number of elements used to just a handful, OR you would need a way to convert basic, plentiful elements into all the elements you need.
The awesome thing is that given sufficient interest we could build a spacecraft with all the properties described above.
The only problem is that a) our propulsion technology is so limited it would take hundreds of thousands of years to get to the nearest star, and b) without some form of artificial gravity the passengers would all die of various illnesses after several decades in space.

Bummer.  Give us two, three hundred years though, and we’ll be so ready.

In the random life of Arin, I just met James Cameron.  He was getting a tour of JPL, so my crew showed off our formation flying robots.  We gave the whole spiel about how they are designed to detect Earth-like planets around other stars, and he was very interested, said we “were doing great work”.  He is damned sharp, faster and quicker to get each concept than our academic peers who we normally give tours to.
I’m officially impressed.

heh, he also said: “you guys are almost doing the ‘pre-research’ for Avatar.  In order to have any reasonable time frame of exploring other stars, 150-200 years, you’d need a pressing reason to do so. Avatar was based on the idea that this had already happened, and that’s exactly what you are trying to do- find an Earth like planet.  Very good work.”

Like I said, a cool guy.  I did NOT tell him I haven’t actually seen his movie yet.