On: Ken Thompson's theoretical C compiler backdoor

[aston]

Not theoretical. Check out the (awesome) paper: "Reflections on Trusting Trust"

http://www.ece.cmu.edu/~ganger/712.fall02/papers/p761-thompson.pdf

edit: It's linked in the Scienceblogs article, but it ought to be the main feature.

[jey → aston]

Here's the talk in HTML: http://www.acm.org/classics/sep95/