Randomness is a crucial component in many efficient algorithms. To
understand when it's necessary and when it can be avoided is a very
fundamental question.  I'll talk about a deterministic analogue to the
random walk and prove that it resembles surprisingly closely the random
walk.  Afterwards, I'll use these quasirandom walks to define a physical
growth model with very interesting properties.  In the third part I will
also propose and analyze a quasirandom analogue to the broadcasting model
for disseminating information in networks.  It achieves similar or better
broadcasting times with a greatly reduced use of random bits.

The talk does not expect any special knowledge about randomized methods.
I'll not give any full proofs, but many figures and several animations to
illustrate the important effects.