An interesting article on the development of long-term, "Social Infrastructure" software. Give it a read. The article also brings up a point I'd like to make myself.

My personal view: Society today is heading for problems directly caused by over-reliance on computing. Am I being a Luddite? No. Technology is wonderful. But the global society today might be relying far too much on what is really a pretty nascent set of technologies.

Think about it: computing only started about 60 years ago, and it has only been a true part of the world's "Social Infrastructure" for less than a decade. Computing technology's every growing pain is going to reverberate through society, and this is going to lead to real social problems.

The solution? It would be immensely difficult to reverse society's dependence on computing. The can is open, the worms are everywhere, so to speak. Economic efficiency and the potential for free, open communication are of enormous value. The benefit side of the cost/benefit analysis is clearly huge. The solution then, must come from minimizing the cost side of the equation, through responsible engineering and the development of a global society robust enough to deal with the problems that are bound to arise as technology matures.