[Elections] A simple SqF board elections solution using SqP

[Peter Crowther]

> From: [...] Cees De Groot
> Please. please please please. KEEP IT SIMPLE.

Technologically and organisationally? :-)

> You haven't been tasked with running the next elections for President
> of the United States. We don't need biometric identification, trusted
> third parties or a notary public to count vounts, whatever. Just the
> Net version of 'please raise hands'.
> 
> We are just a bunch of guys in a *friendly* community. We have our
> technical differences, but I really cannot imagine any inside party
> seriously attempting to mess up these elections. Nor an outside party.

Lex, at least, has repeatedly raised the points that (my
interpretation):

1. He regards the current leadership as illegitimate;

2. He does not trust any system for selecting voters if it can be
affected by the actions of a small power group;

3. (reading into some of Lex's messages) I *think* he's concerned about
(2) in the case of SqP because some of the roots are part of the current
leadership.

Lex, these are my interpretations and there's a strong chance I'm
misrepresenting you here; I'd very much appreciate clarification of your
position on these points.

For the record, my own positions are:

1.  The current leadership made a power grab in a power vacuum.  It was
not legitimate by any *democratic* standard*; legitimising the position
of that leadership would be prudent if it is to be able to resist
accusations common to most revolutionary governments.

2. I'm concerned about the selection system in theory, as it clearly
*could* be constructed so as to be biased;

3. I am not concerned with a SqP-based selection system in reality, as
I've seen no indication that the roots are behaving inappropriately.

> A system is needed that is reasonably objective and which stands a
> reasonable chance of listing the right people on the voter's registry.

Where 'the right people' needs definition.

> It's easy to discuss this issue to death, to build in very strong
> safety guards - but let us do something *simple* and just go for it.

To me, the key question when evaluating a possible system is "how likely
is it that the inevitable errors in this system alter the result?"
There will be errors in any system we devise.  One problem I have with
answering that question in the case of Squeak is that we are at the same
time trying to work out who "the right people" are and construct some
form of voting system for those "right people" to return multiple
candidates.  I'd much rather hand the second job off to some body of
people who have done it before, hence my suggestion of a known system
with known properties (a Condorcet variant).  Unfortunately, I don't
think we *can* hand the first job off to people who have done it before,
so we're stuck with having to work it out.

> On 1/8/06, Peter Crowther <Peter at ozzard.org> wrote:
> > Indeed.  For recording and counting votes, however, I'd settle for a
> > system where the system's owner has no interest in tampering
> > with the result.
> >
> Should I take that as an insinuation? I won't. Just let me put
> forward, for the record, that I don't have any interest in tampering
> with the result (as owner of the system where SqP runs). I wouldn't
> know why...

No, you shouldn't, and I should have worded it as '... the system's
owner *can be seen to have* no interest in...'.  Any election system run
by some of the people up for election runs the risk of accusations of
bias, either in voter selection or in counting - see, for example, the
2000 and 2004 presidential elections in Florida, where accusations of
bias were levelled in both cases.  Systems that are clearly run by an
external body are at least freer from accusations of bias in counting.

The castaways are clearly already sensitive to the perceived legitimacy
of their power, hence the long-standing promise of elections in
February.  We should also be sensitive to the perceived legitimacy of
those elections.  They must be seen to be unbiased, otherwise the
undercurrents that are still present in the community from the original
formation of the power structure will remain or get stronger, with an
increased tendency for the community to fragment.  Assuming we prefer
the community to be as large and as cohesive as possible, this would be
a poor result.

Personally, Cees, I'd be quite happy with a voting system you wrote
and/or ran; I can't see any way *in fact* that you'd want to tamper with
it.  But it's difficult to get rid of the possibility of the system
being *accused* of bias in such a circumstance.

Has that made my position any clearer, or has it just made things worse?

		- Peter

* Of course, this begs the question re democracy being the appropriate
mechanism for governance of the Squeak community, yada yada... but this
is called the elections team, so I'm assuming that a democracy with an
elected leadership is considered appropriate.  The community ran under a
very different model for a number of years, and one of the issues
appears to be that it's unclear that an elected democracy *is* an
appropriate model for Squeak.