Fixing Statistical Errors in Spreadsheet Software:
The Cases of Gnumeric and Excel
B. D. McCullough - Dept. of Decision Sciences
LeBow College of Business, Drexel University, Philadelphia
http://www.csdassn.org/software_reports/gnumeric.pdf
"The open source spreadsheet package "Gnumeric" was such a good clone
of Microsoft Excell that it even had errors in its statistical
functions similar to those in Excel's statistical functions. ...
Gnumeric v1.1.2 has largely fixed its flaws, while Microsoft has not
fixed its errors through many successive versions.
... use Gnumeric rather than Excel."
[ tests & charts clipped ]
"One might have wondered whether there is something inherehently
difficult a fixing statistical bugs in spreadsheet package. It is
reasonable for the developer of a spreadsheet package not to fix errors
in a timely fashion? The answer to this question is seen to be, No. A
related question is whether amateurs working on a volunteer basis, with
no budget for R&D, could do a better job than professionals who work for
pay and have an R&D budget) when it comes to fixing statistical error in
a spreadsheet. The answer to this question is, Yes."
--------------------------
-- The above paper is referenced in discussions of Proprietory
Software-vs-OSS at http://cgi.linuxfocus.org/cgi-bin/lftalkback?anum=345
It followed a Linux Focus article "Linux in Science - Or How a Useful
Neural Network Utility Was Developed" at:
http://tldp.org/linuxfocus/English/September2004/article345.shtml
---------
To unsubscribe, send email to <aklug-request@aklug.org>
with 'unsubscribe' in the message body.
Received on Sat Sep 4 21:12:11 2004
This archive was generated by hypermail 2.1.8 : Sat Sep 04 2004 - 21:12:13 AKDT