[Foundation-l] Wikimedia Board Elections

[Aphaia]

On my wikibreak, yesterday, I was worried how we detect open proxies
... the strongest tool I have is not batch-processing type..

On 9/23/06, Ronald Beelaard <rbeelaard at gmail.com> wrote:

> I presume the standard check user tools are inappropriate for that.

Agreed.

> The reason I post this (rather critical) question, is the
> aforementioned work on open proxy detection. With just a list of IP's,
> I could easily scan those IP's on being an open proxy or not. If
> detected that is positively yes, if not detected the likelihood of
> being OK is pretty high as the program also takes into account the
> daylight periods in the respective time zones of the IP's.

Your proposal is attractive for me but I am strongly hesitant to give
you raw data. Not depending  I and / or the other Election officers
know you, I simply fear the current version of Wikimedia Foundation
private policy doesn't permit us to give such raw data to the third
party. (Brad?). I am afraid my counterproposal sounds rude, but would
you please consider giving (one of) us your script/tool for detection?

> An even more easily obtainable by-product of such an exercise could be
> a frequency distribution of votes per country and that might be more
> useful for understanding the origin of votes than just the language.

I think it would be interesting, but "the origin of vote per language
community" and "per country" are not compatible in my impression. Some
Asian language community editors could be in the USA, for example. And
"involvement of editors who are active on a certain project" and "...
in a certain land" might be perhaps related, but not equal.

Sincerly,
--
Kizu Naoko
Wikimedia Foundation Election Committee, 2006