[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: On linux kernel packaging issue, assuming that user is never right

On Sat, 08 Nov 2003, Mathieu Roy wrote:
> People knows all about placebo effect, but do you have any evidence
> that there is nothing more than placebo effect?

It's normal for the person claiming that there are two populations to
provide appropriate data and statistics to back up the claim.[1] By
default, we generally assume that two groups come from the same
population without evidence to refute it.

> If you compile a kernel without lot of modules for hardware and stuff
> you do not have, there is nothing weird to suppose it may have some
> consequences at later point, by having a kernel size reduced by 25%. 

You can suppose it. But you shouldn't wander around claiming it to be
so without actually running the tests and providing the data.

> One more time, it means taking users for loosers. [...] There may be
> also real hardware failure/misconfiguration. 

Without appropriate tests and measurements, there is no way to
distinguish between real change and perceived change. Since we have
the tools available to quantitatively measure the change, why not go
down that route and actually measure it before making these claims?[2]

Don Armstrong

1: Quite obviously, because H_o cannot be "proved". We can only fail
to reject it.

2: If you do, please do a better job than most benchmarks that I've
seen. Hint: If you don't have replicates of your sampling, your data
is useless and says nothing.
Fate and Temperament are two words for one and the same concept.
 -- Novalis [Hermann Hesse _Demian_]


Attachment: signature.asc
Description: Digital signature

Reply to: