[Cryptography] cheap sources of entropy

[lists at notatla.org.uk]

Dan Geer writes:

> > One well-calibrated well-defended well-monitored entropy source
> > makes incomparably more sense than an arbitrarily complicated
> > conglomeration of sucky sources.

> Recalibrating first principles for a moment, please.  My understanding
> is that a mix of N bit streams will be truly unpredictable if any 1 of
> the N bit streams is truly unpredictable.
> 
> If that is incorrect, what am I missing?  (RTFM is entirely acceptable
> and even gracious if accompanied by a pointer to TFM to R.)

Details .. you are perhaps assuming each of the N streams knows nothing
about the others and the part that combines them is trusted to do that
simple task.

That's not always easy to prove - for instance when your mixing at a CPU
uses an on-chip RNG as one of the sources.