In message <[EMAIL PROTECTED]>, John Sadler <[EMAIL PROTECTED]> writes
And these are corroborated by who ? Which version is involved and what level of arithmetic.? Has the allegedly faulty version been upgraded since the fault was reported ? Has it ever been reported ?There is some correspondence on EDLUG suggesting that the statistical functions in Excel are flawed. The commentator wondered how many corporations are going to find there projections do not work!
I seem to be becoming a Microsoft apologist which is not a position I choose to have but I do find it vaguely annoying when people come up with unsupported statements because they fit their arguments. There was a similar story going around about the Pentium 1 processor's maths ability which caused great jollity amongst the Intel hating farternity (sic) but, in reality affected a very small number of calculations and was easily fixed. It was, nevertheless, heralded as a major problem at the time. I doubt if any company would use an Excel spreadsheet for anything really important or complex mathematically and, since I use them all day at work for many different forms of calculation and have never found an error I would say that it is unlikely that it is important.
To be fair to all sides, the mathematical errors ( if any ) would not be noticed by ordinary users.
We would need more detail on which 'statistical functions' in Excel 'are flawed'; and it what circumstances this would manifest itself.
Which will probably only be of interest to mathematicians :-)
-- Malcolm Cadman _______________________________________________ QL-Users Mailing List
