[MG] Mixing proposal tuning and voting creates consensus

[Scott Raney]

On Sun, May 11, 2014 at 8:22 PM, Bentley Davis <Bentley at bentleydavis.com> wrote:

> I've read through this thread and most of "The Plan"
> http://www.metacard.com/ThePlan.pdf and I can't get a clear indication how
> "The Plan" is different from what Ed is proposing

I think there are two big differences:
1) Ed (et al) proposes rating and voting to be the same process.  I
separate them, not only into separate phases, but also that in the
first phase people don't even rate the proposal directly, they add
comments and rate only those comments (i.e., it's preparation of a
voter's guide).  Only at the second phase does everyone vote on the
actual proposal.

2) I am keenly aware of three practical/engineering constraints that
don't seem to concern Ed:

  a) To control the authoritarians we need to have high (i.e., more
than 1/3) participation rate on the final vote.

  b) Stuff needs to get done, even if there is no "consensus".
Letting a proposal that has simple majority support but no consensus
sit in limbo forever is not an adequate design.

  c) We need to reduce the time commitment at all levels as much as
possible, which means we need high levels of information filtering so
that people can make an informed decision on all issues that affect
them in an hour or so a week.  Meaning that of the thousands of
proposals that will be made the majority of people should only see a
couple of issues come to a vote each week.

> Who can make a proposal?

Anyone.  I'm still working out how a *group* can make a proposal (so
that we don't end up with the problem we found in Loomio where only
you could edit your proposal).

> Who can modify a proposal?

The original proposers only.  Modifying a proposal triggers an email
an request for review to everyone who had commented or rated the
previous version.

> What is the trigger for a proposal to come up for vote?

An empirical question, to be refined after we see how it actually
works.  As I said before I'm currently thinking there would be several
constraints:
1) A certain time has elapsed (a month?).
2) A certain number of people have entered pro/con statements or rated
someone else's.
3) The ratings on the top 10 "pro" statements are a full point (out of
5?) higher than the cons.
4) The proposers release it for this.

Note that if #1 is met but others aren't I'm thinking to allow
forking.  We could allow that at any time except the fork shouldn't
have to start from scratch, i.e., everyone who commented or rated
would be notified of the fork.  We need to social engineer this to
some extent to keep an idea moving forward yet not burden everyone
with the prospect of keeping track of a large number of forks of
proposals on it...

> What is the criteria for a proposal to succeed a vote?

I'm still liking a simple majority of the voters or yes vote from 1/3
of the registered voters or their proxies, whichever is larger.  But
I'm flexible on the exact threshold (25% might very well be
sufficient).
  Regards,
    Scott