Open Letter: Why I, Kevin Kofler, am not rerunning for FESCo

Stephen John Smoogen smooge at gmail.com
Tue May 4 03:34:29 UTC 2010


On Mon, May 3, 2010 at 9:13 PM, Mail Llists <lists at sapience.com> wrote:
> On 05/03/2010 10:01 PM, Jesse Keating wrote:
>
>>
>> Of course the sample is biased.  It's a sample of people who frequent
>> the forums, that's a self selecting group of people, by no means a
>> worthwhile representation of the Fedora user base as a whole.
>
>
>  FYI - Not true - I joined the forum for the sole purpose of voting in
> that poll - everyone I know who voted joined the forum for the same
> reason. The mailling list brought it to my attention.

Actually what you did is considered the classic example of skewing the
data by self-selection at this point it is no longer polling but
'voting'. It is the reason why many polls will say 'unscientific
survey' when done this way... you can not apply standard tests and get
answers that would be found scientifically valid.

For a poll to be statistically valid, the population must be randomly
polled. The population polled may be a 'self-selected' target (eg
those who have registered to vote versus the general population) but
it needs to be a semi-closed set. [EG polling surveys of registered
voters after the deadline to register is considered to be more
accurate than before so because the population will not suddenly grow
afterwords.]

Listen, I would have to say this no matter what the survey came out
was. It was seriously flawed in ways that make using it worse than
having no data at all. It also makes it harder to do a proper survey
because of the emotional baggage that various people have attached for
or against the first one.

>  We all follow the mailing lists - as presumably does FESCO - so the
> sample is less biased than is being suggested.
>
>   FESCO is not voted in by Aunt Tilly ... are they?

They aren't voted in. The range voting method does not vote people in
or out.. it determines who the majority of people are most likely to
'live' with. Basically it tries to remove the emotional political ends
and find who the 'silent' majority want. How much it does that is
dependant on other factors but that is what it (and the Debian voting
system) aim for.

In the end, being elected by a range vote is less of a mandate than an
acceptance.

-- 
Stephen J Smoogen.
“The core skill of innovators is error recovery, not failure avoidance.”
Randy Nelson, President of Pixar University.
"We have a strategic plan. It's called doing things.""
— Herb Kelleher, founder Southwest Airlines


More information about the devel mailing list