Quote:
Originally Posted by Moogs
Enki: I'm obviously not a big spreadsheet geek but your comments are surprising. I would think by now all things computational would be uber-stable in Excel. I mean every business on the face of the planet, including engineering businesses, etc, use Excel. Or at least I thought so. You'd think math would be the very easiest thing to fix, bugs-wise, but I guess not?
Maybe Apple's version will be better. And yah personally I wouldn't use a lot of statistical stuff I would just get Vista office for straight 1:1 compatibility with the rest of planet earth. And there's also that free version of office you can download that mirrors a slightly older Windows version I think.
|
It's not that the functions mathematics are "buggy",in the broken sense of the word, it's that they use some odd nonstandard interpretations of how to implement many of them. Since they aren't adequately documented and fully modifiable, you gets what you gets whether you want it or not. Well I don't want some oddball implementation of a function that I don't trust. The graph plot stuff is class A-1 hosed, guaranteed-pain-in-the-ass though.
Most users are oblivious and just always use the default XL functions, which is fine as long as they don't compare with the exact same numbers to a non-Excel function that isn't oddly implemented. Or I often hear "there's just a little typo in the data somewhere, but we are close enough". Riiiigggghhhht. And we wonder why 75% of all businesses fail in their first year. (Not XL's fault directly, just users ignorance in general).
I'm merciless about correctness and cross-checks. Others like myself accept XL for the great tool it is, but acknowledge that once you get too far into provided formula land you give up too much control and verification of the result. The oddball functions have never been updated as far as I can tell despite protests a good number of years ago, that leaves me with little confidence in the rest of the non-trivial ones.