Summary
	l A simple fast min-cut algorithm » Random selection avoids rare problems l Generalization to near-minimum cuts l Bound on number of small cuts » Probabilistic method, backwards