On 12/05/2011 02:13 PM, Chí-Thanh Christopher Nguyễn wrote:
> Zac Medico schrieb:
>>> 2011/12/4 Chí-Thanh Christopher Nguyễn <email@example.com>:
>>>> I already stated that I would leave the call to zmedico as lead
>>>> developer of portage. Giving him the arguments and the data to make the
>>>> decision, but watching whether his reasoning is sound. That means
>>>> clearly saying where it is based on fact and where it is based on opinion.
>>>> (I wouldn't mind if he made an entirely opinion based decision, as long
>>>> as he didn't claim it to be fact-based)
>> I think it's difficult or impossible to get an reasonably unbiased
>> sample of the population of Gentoo users as a whole. Voting is only fair
>> when you can get a reasonably unbiased sample of the whole population.
> This is something I can agree with.
Yeah, I think anyone with a cursory knowledge of statistical sampling
can agree with this. On the other hand, it seems apparent that lots of
people are completely oblivious to the challenges involved in obtaining
reasonably unbiased statistical samples.
>> Due to human nature, people who are unhappy with the change in defaults
>> are more likely to express their opinion publicly than people who
>> welcome the change in defaults. This tends to bias statistics in favor
>> of the unhappy people.
> But this is precisely what I meant above and in my older post. You have
> no idea to which extent this effect is present, much less whether other
> effects which could introduce bias in the other direction outweigh your
> "unhappy with the change in defaults" group.
> You keep on repeating this as if it were fact, but have no data to base
> this claim on.
One person's common-sense behavioral model is another person's baseless
> One way to investigate would be to sample statements in the forum
> thread, and determining how many responded with personal preference and
> how many with practical arguments. It would still have to be accounted
> for those who try to rationalize their pre-conceived opinion with ad-hoc
> arguments, but better than nothing at all which is the current case.
Sure, but that seems like more of an academic exercise than a practical
one. I think we'll better of with a judicial approach, where a group of
judges weighs a set of pros and cons. Gentoo's council is the closest
thing to that we have to a judiciary.