Floor, Ceiling, and Log
	floor - round down to nearest whole number
	ceiling - round up to nearest whole number

	log10(x) = ?
		What power would I need to raise 10 to to equal x
		log10(1000) = 3
		log10(5000) = 3.7
	log2(x) - What power would I need to raise 2 to to equal x
		log2(1024) = 10
		log2(5000) = 12.29
		log2(1024*1024) = 20
		log2(1024*1024*1024) = 30
	log(x) grows much more slowly than x

	How many bits to represent decimal number N in binary?
		floor(log2(N)) + 1

		1 bit: 0-1
		2-bits: 0-3
		3-bits: 0-7
		4-bits: 0-15
		5-bits: 0-31

	Repeated Doubling: If we set X = 1, how many times can we 
	double X before it becomes greater than or equal to N?
		ceiling(log2(N))

	Repeated Halving: If we set X to N, how many times can we 
	halve X before it becomes less than or equal to 1?
		ceiling(log2(N))

Search(Array, Value)
	return index of Value, or -1
 
Sequential Search
	unsorted data

	unsuccessful search
		always n
	successful search
		worst-case: n
		average-case: n/2

Binary Search
	iterative

	unsuccessful search
		always floor(log2(N)) + 1
	successful search
		worst-case: floor(log2(N))
		average-case: floor(log2(N)) - 1

	If binary search is so fast, why would we ever need 
	any other search algorithm?
		It assumes a sorted array, which is hard to
		maintain efficiently if the set of values is dynamic

	recursive

	modified binary search that returns the position where the 
		value would have been if it were in there
		(final value of low)

	Java binarySearch

Boggle
	dictionary implementation
	prefix pruning