Interesting, nice foresight. Remind me of my college days when I could design stuff like DDoS resilient distributed voting systems without the burden of having to build and test it (although I would not have gotten away with that in my MSc thesis), truly understanding the issues, or being in a constant inner-struggle over the purpose of computer science.
the paper is at <a href="http://dspace.mit.edu/bitstream/handle/1721.1/41756/226233316.pdf?sequence=1" rel="nofollow">http://dspace.mit.edu/bitstream/handle/1721.1/41756/22623331...</a>