Re: 0.99999998 (was: Unknown SQL)

>IEEE gets more complicated than that! In IEEE arithmetic, -0 and +0
>are distinct values. So 1/-0 == -INF, 1/+0 == +INF. And yet -0 == +0
>in all tests.
>
>I doubt many people will have problems with this particular thing in
>conversion, but there is a more general point, that is that equality
>is not as simple a test as it may first seem.

The fact of the matter is: -0 does equal +0. I don't see what the problem is.

>Say if two banks, A and B, merge, then if a customer of bank A is
>called 'Mrs Robinson' and a customer of bank B is called 'Mrs
>Robinson', then although the strings match textually, there is no
>guarantee that they are one and the same person. You cannot just
>import the data.

Well, actually you can just import the data. Since each Mrs. Robinson has some form of customer ID at one of the two banks and since the import process must translate these customer ID's into non-overlapping sets in the merged result, one can just import the data. Your initial assumption that banks will identify customers by honorific and surname is absurd.

Regards,
Bob Received on Sun Jul 22 2001 - 01:29:21 CEST