Hi!
Interesting. I appreciate the ideas you've put on
the table.
I am glad about that. I worried a bit that I threw too many of them out. :-)
I'm happy to start discussing and working on this.
Great!
1. Way too difficult to understand (as you point out).
Maybe, or maybe not. I see here two approaches:
- have a user interface where people who are interested in how
everything works and how it is behaving can look into it in real-time
and monitor how things are happening - so once the system is live, I
think people can just learn by observing how it works
- have a user interface where people who just want to make decisions,
can just make them and don't have to worry about how computation is done
in the background, what they worry and experience is that decisions are
made better then before, and this is what counts
:-)
2. Delegation of voting allows budding politicians to
gather power by
collecting delegated votes. In a large system a handful of these people
would likely end up controlling everything.
I don't see this as a bug, but a feature. This is part of democracy. It
does not change the fact that popular or good talking people will have
more influence in the group. But what it does change is that this
influence will be explicit and not implicit. And while those people will
still be able to have influence, in my opinion, other people, who maybe
don't want to be exposed and are not exposing themselves, but people
still trust them, will get more influence.
So my look at this is, that the system is at worst as bad as what we
have currently (media and people of power manipulating/influencing). But
because you empower also those, who are traditionally not empowered, you
improve the results.
In some way it is Plato's idea of forcing philosophers to guide the
group. They, if they are true to themselves, don't want this power. And
through such system you allow them that they vote as they would vote
just for themselves, maybe they don't even know that they have more
power, but in fact they have been given trust by people.
And don't forget. Delegation is active only when a person does not vote
him or herself. So nobody can gather power if people don't allow that.
If people allow that - isn't this democracy then?
So what I hope to achieve in practice is that for mundane tasks only few
people have to vote and the thing is done. Not whole community have to
be involved in some simple decisions. But when something big is there,
community can get naturally more involved and start casting votes
personally.
So you get both speed of more centralized societies for not important
issues and direct democracy for more important issues, where community
can decide on issue to issue basis where it wants to be on this spectrum.
3. Does not take into account other types of voting
such as "everyone
selects all the choices they are ok with".
It does. The system is orthogonal to this. You can have rankings, or
multiple choices.
4. Citations or references to existing research. There
are people who
dedicate their careers to studying this kind of stuff.
Yes. I am studying this as my research. What I sent was description of
the way how I think voting scheme could be implemented. If you want to
read more about related work, you are invited to read this report I had
for one class last semester:
https://github.com/mitar/theory-project
4. No simulations. How can you know your program works
if you haven't run
it?
In above repository there is some basic code I was using to play with
idea. But I agree, what is missing is a real program people could try
out. So this is what I plan to do as my next step, make a simple
webportal where a group can register, define delegations and use it to
compute decisions. At least the association we are planning to establish
in Slovenia is centered around radical transparency, so votes and
everything was meant to be public anyway. Because the real issue I don't
yet have an answer to about my proposed schema is how to assure privacy
of data (of votes and of the delegations themselves - such social
network of delegations is highly sensitive) while keep security. For
small communities like small associations this might not be big problem
because mostly this is already known, who trusts who. So the system is
there then just to facilitate faster and easier decision making.
But yes, if it would be already done thing it would not be so
interesting, wouldn't be? So, I am inviting all of you to hack also the
way we are doing decision making. I am not sure if it will work, I
believe it will, but of course I cannot promise you anything. This is
why I wanted first to try it in our own association, to eat my own dog
food. But maybe somebody else is also interested in joining for dinner.
:-) Sharing is caring! Of dog food as well. :-)
I realize that what you pitched is more of an initial
concept than a
completed voting system design and that some of these issues are only
issues because you haven't gotten around to solving them yet.
What I sent you was just a description of the idea. How I envision the
end product you can see in this document:
https://docs.google.com/document/d/1uuQsYeWYU4-dTkuk9YKhACWZIgE5GD7rcmsLTKt…
My next step
is to develop a simple app to be used inside the
organization. If you are interested, we might do it together.
Hm. I guess that's one way to test your system :-)
Yes! No need to theorize, hack, create a system, and let's try it!
Do you have some links to those systems?
There are in the report above (links to papers). The ones I would
recommend that you check out (which are open source projects with real
developed code and not just theoretical concept of how it could be) are:
http://liquidfeedback.org/ (used by German Pirate Party)
https://adhocracy.de/ (a more user friendly version of same idea)
https://www.loomio.org/ (New Zealand approach, no delegation, but nice
simple interface and normal voting)
A nice video about liquid democracy:
http://www.youtube.com/watch?v=fg0_Vhldz-8
Mitar
--
http://mitar.tnode.com/
https://twitter.com/mitar_m