[OSGeo-Discuss] Bug report response time
Mateusz Loskot
mateusz at loskot.net
Fri Nov 12 17:04:42 EST 2010
On 12/11/10 08:01, Paolo Cavallini wrote:
> Il 11/11/2010 23:12, Mateusz Loskot ha scritto:
>
>> Statistics is evil. As long as the methodology is right, you can
>> proof any bullshit you like, or shoot yourself in your the foot.
>> So, I'd be careful.
>
> Sorry, I do not agree. Statistic is good, if (as anything) you use
> the correct method. Using single cases, as you did, is not using
> proper statistics.
I used simplified example, though well expressing my concern.
Imagine each single "proof" is a sample of 1000 cases. Given enough
amount of time, I can find it.
> If you analyse averages or medians, quartiles and standard
> deviations, you get meaningful tests of your hypotheses.
I can use the same techniques to calculate the statistics to prove
there is a relation between number of bugs reported to GDAL Trac and
number of sunspots observed within a period of time.
"Statistics is not math", my math teacher used to say.
George Cobb "In mathematics, context obscures structure. In data
analysis, context provides meaning"
David Moore "Mathematical theorems are true; statistical methods are
sometimes effective when used with skill"
Statistics is bad.
Best regards,
--
Mateusz Loskot, http://mateusz.loskot.net
Charter Member of OSGeo, http://osgeo.org
Member of ACCU, http://accu.org
More information about the Discuss
mailing list