DNA is a fabulous molecule. I wouldn't be where I am without it. Obviously, anything with the ability to make everything from seaweed to the King of Sweden has a lot of potential, so it's no surprise to find people making computers out of the stuff. The latest effort has made a general-purpose computing engine out of the stuff, at least in theory, and while it's still far too much trouble to cook up a word processor in organic gunk one still gets that shivery feeling that we're about to kick off yet another deep revolution in the way we live in this world.
I hope someone's keeping an eye on the crossover between DNA programming and the work done in understanding how changes in DNA result in new species. There's a lot we don't understand about the way evolution actually happens, and some evidence suggesting that the DNA replication mechanism has the ability to code not only mutations but the conditions under which they were successful. If similar conditions turn up again, then the mutation gets retried -- at least, that's one theory.
Instinctively, it seems that discoveries in DNA computing should shed light on what's going on in DNA evolutionary biology -- and vice versa. Science is often very bad at cross-disciplinary collaboration: I do hope that's not happening here. The benefits for both fields from sharing data could be extraordinary.