I only made 770 (67th percentile) on the practice GRE CS subject exam I took today. I guess I need to study up on graph theory and grammars over the next month. The computer architecture questions were all fun though.
Stupidest question of the exam, which I “missed”:
Which is the closest to a perfectly secure encryption scheme:
a) Caesar cipher (heh)
b) DES
c) Enigma
d) One time pad
e) RSA
I take issue with the supposedly correct answer of “D”, since yes, in principle, it is perfect encryption, but in practice, it sucks. Knowing OTP is flawed has to be more useful than knowing that it could be perfect in circumstances that are never true. If only there were an option in the list which provided good encryption and also solved the key distribution problem!
The engineer in me is at times in conflict with the “computer science == astronomy” crowd (yeah, I’m looking at you Dijkstra). Which is why every time I read someone’s article on here’s-how-to-do-X-in-unreadable-haskell my eyes glaze over. Maybe I’m not smart enough, but I would take the fact that there are now a zillion blogs trying to explain monads as evidence that the functional people made pipes too confusing.
Oh, by the way, here’s a secret: CPUs are imperative!