[MG] "rough consensus"

[Scott Raney]

As I was reading that vTaiwan writeup, I came across this reference,
regarding "humming" as a polling tool:
https://tools.ietf.org/html/rfc7282

I have to say, that's about the most pitiful example of amateur social
engineering I've ever come across. Not only is it internally wildly
inconsistent (it first claims "no kings", then assigns the chair
absolute power over not only the final decision, but indeed the entire
course of the debate), and woefully ignorant of political practice and
terminology (it's called "log rolling", not "horse trading"), but it
simply ignores the possibility that there are better ways to make
decisions, particularly in highly technical fields. And this is in
about the most homogeneous environment you can possibly image: It
makes the Quakers look like a herd of cats! But perhaps most pitiful
is the fact that this group, which is 100% internet masters, is still
using email and in-person meetings to hold debates and votes. Argh.

As to that "better way", even Loomio and Vilfredo and Pol.is would
provide that: A way for people to comment on a proposal and attempt to
promote, change or in some cases justifiably delay consensus by
recruiting defectors from the other side with quality arguments. I'm
not planning to include a component like that in AutoMatch, at least
in the first version, because I think it's a lot less likely to be
effective in a heterogeneous environment (especially an ideologically
heterogeneous environment), but see it as an obvious solution for the
IETF. I'm reminded of David Brin's proposal for "Disputation Arenas",
which would be perfect for that kind of problem:
http://www.davidbrin.com/disputation.html
Regards,
Scott

[max stalnaker]

Maybe you want complicated? Maybe humming has a specific cultural
advantage. Maybe occupy oregon meetings have an absolute dictator for
three hours. Under RRO the chair can do no wrong if he has two thirds
support. And RRO is too complicated when there is conflict for most people.

And people do not want to spend all day on decision queues.

Post by Scott Raney
As I was reading that vTaiwan writeup, I came across this reference,
https://tools.ietf.org/html/rfc7282
I have to say, that's about the most pitiful example of amateur social
engineering I've ever come across. Not only is it internally wildly
inconsistent (it first claims "no kings", then assigns the chair
absolute power over not only the final decision, but indeed the entire
course of the debate), and woefully ignorant of political practice and
terminology (it's called "log rolling", not "horse trading"), but it
simply ignores the possibility that there are better ways to make
decisions, particularly in highly technical fields. And this is in
about the most homogeneous environment you can possibly image: It
makes the Quakers look like a herd of cats! But perhaps most pitiful
is the fact that this group, which is 100% internet masters, is still
using email and in-person meetings to hold debates and votes. Argh.
As to that "better way", even Loomio and Vilfredo and Pol.is would
provide that: A way for people to comment on a proposal and attempt to
promote, change or in some cases justifiably delay consensus by
recruiting defectors from the other side with quality arguments. I'm
not planning to include a component like that in AutoMatch, at least
in the first version, because I think it's a lot less likely to be
effective in a heterogeneous environment (especially an ideologically
heterogeneous environment), but see it as an obvious solution for the
IETF. I'm reminded of David Brin's proposal for "Disputation Arenas",
http://www.davidbrin.com/disputation.html
Regards,
Scott
_______________________________________________
Start : a mailing list of the Metagovernment project
http://www.metagovernment.org/
Manage subscription: http://metagovernment.org/mailman/listinfo/start_
metagovernment.org

[Scott Raney]

Post by max stalnaker
Maybe you want complicated?

Things should be as simple as possible, but no simpler. If it takes
technology to enable humans to make collective decisions in groups
larger than our built-in programming was designed to handle (around
30), that's what we'll have to use...

Post by max stalnaker
Maybe humming has a specific cultural
advantage. Maybe occupy oregon meetings have an absolute dictator for three
hours. Under RRO the chair can do no wrong if he has two thirds support.
And RRO is too complicated when there is conflict for most people.

Adding the "two thirds" support requirement makes it a fundamentally
different process from what the IETF is using ("absolute dictator"
truly means "absolute" there, albeit with the possibility of appeal
and review at a later date).

Post by max stalnaker
And people do not want to spend all day on decision queues.

Agreed, and with well designed decisionmaking software, they won't
have to! Having large numbers of people gather in a room to make
decisions, or indeed even disseminate information, is so 20th
century...

Thinking more about Pol.is and how it is actually used (especially
with vTaiwan) the fundamental limitation of its design is becoming
clear to me: It truly is just a polling tool because it is essentially
a one-way process. It is *not* being used to disseminate information
or encourage debate, but merely to collect existing or
spontaneously-formed opinion. If it was designed to disseminate
information and *facilitate* consensus forming the posts would contain
argument and fact, which near as I can tell they almost never do
(almost all are "tweet" sized, and we all know how Twitter has ruined
the process of debate: FaceBook is a veritable Disputation Arena by
comparison).
Regards,
Scott