17 Nov
2005
17 Nov
'05
10:09 a.m.
John McKay wrote: << . . . perhaps someone can tell us when numerical approximations are useful? . . .
Sure: * When you're experimenting with concrete cases to try to find or confirm a pattern. * When for similar reasons you want to see a fairly accurate visual depiction of a mathematical object or of some statistics related to one. * When you want to display a result that cannot to your knowledge be expressed in closed form. (And, I'm sure, for many other reasons.) --Dan