ext_8821 ([identity profile] lederhosen.livejournal.com) wrote in [personal profile] lederhosen 2005-08-01 09:41 am (UTC)

1^inf is bad in the same way that 0*inf is bad. In fact, if you take the log of (1^inf), that's exactly what you get.

The catch is that although it is often convenient to manipulate 'inf' as if it were a number, it's not. An expression like '1/inf = 0' isn't really saying that inf is the reciprocal of 0; it's more shorthand for 'as x grows arbitrarily large, 1/x tends to 0'. Most instances where inf appears, we're really talking about limits or some such, and all the operations involved are really on *finite* numbers (though possibly an infinite set of all-finite numbers).

There are situations where it's consistent to say that 1inf = 1, just as sometimes it's consistent to say that 0*inf = 0. (limx=>inf0*x certainly equals 0.) But there are others where it breaks down; in the neighbourhood of x=1, y = inf, the function z=x^y is horrendously discontinuous.

Post a comment in response:

If you don't have an account you can create one now.
HTML doesn't work in the subject.
More info about formatting

If you are unable to use this captcha for any reason, please contact us by email at support@dreamwidth.org