Re: Case for/data about elections

On Mon, 19 May 2014 17:55:37 +0200, David Singer <singer@mac.com> wrote:

>
> On May 19, 2014, at 10:27 , Jean-Charles (JC) Verdié <jicheu@yahoo.fr>  
> wrote:
>
>> Hi Brian
>>
>> We need data. So far what we have is feelings that the vote is biased or
>> that it isn't.
>>
>> If you are on ac-forum you can see [1] that I asked all the people who
>> are running for the current election to agree on having the results
>> being made public (after the ballot is closed), but so far it has not
>> got a great support, so I fear we once again won't have access to the
>> raw data and therefore people will be able to stand on their position
>> that the system is or isn't flawed
>
> I agree.  I think we have three problems to resolve here before we can  
> make progress:
>
> * not enough AC reps vote at all for the question to be meaningful

What would be a meaningful proportion of the AC voting? Where is the data  
on how many of them do, on which you base your claim?

> * we have nowhere near enough data for us to work out whether we have a  
> meaningful problem

Indeed we cannot determine for certain without more data. Although logic  
suggests we have one, and the circumstancial evidence we have does not  
appear to contradict it or suggest we are somehow a sufficiently  
exceptional community that the logic does not apply.

> * I am not at all sure we are agreed on the purpose of the election;  as  
> I said before, I think Chaals wants it to reflected the (desired or  
> actual) diversity of the membership, but maybe we want an AB that has  
> the people who will best serve, and so on…

This is a false dichotomy. Unless you propose some method other than  
voting to determine in advance who will best serve, it isn't a different  
criterion.

I'm not explicitly looking for some quota-based diversity. I am looking  
for some assurance that most votes of the AC are relevant to the outcome,  
and that the people elected reflect something tending toward a consensus  
of the voters. Neither of those are properties of the current system we  
use - which is why almost 2 centuries of work have been done on voting  
systems which *do* have those properties.

> I agree we can and should solve the data question.

We cannot, except by all publishing our votes, which I think is unlikely  
and would be a bad idea.

W3C Team can. I agree they should. But it appears they do not plan to do  
so.

> The first question, well, I don’t know.

To the extent that elections are now genuinely contested, it appears that  
this isn't a structural problem at the moment.

>  The third we should be settling in the ‘process’ discussion list.

Which is what this thread is doing (among others).

cheers

-- 
Charles McCathie Nevile - web standards - CTO Office, Yandex
chaals@yandex-team.ru         Find more at http://yandex.com

Received on Thursday, 22 May 2014 16:22:36 UTC