Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Quadratic voting interface for delegations #247

Open
santisiri opened this issue Mar 22, 2018 · 21 comments
Open

Quadratic voting interface for delegations #247

santisiri opened this issue Mar 22, 2018 · 21 comments

Comments

@santisiri
Copy link
Member

An early prototype for enabling quadratic voting features applied to delegations on Sovereign.

@virgile-dev
Copy link
Member

Hey @santisiri
Why only on delegations? I would have implemented it for voting as well.
What do you think ?

@santisiri
Copy link
Member Author

The goal of quadratic voting is to prevent monopolies. A proposal cannot be a monopoly per se, having powerful ideas matters. The goal is to prevent excessive influence from the players not the content.

@virgile-dev
Copy link
Member

The few papers I read is often applied to voting arguing that it is the best way "to supply better valuations that aggregate private information of dispersed multitudes." ( Eric Posner

I'll get a crack at mocking up something for delegation, but I don't why voting on issue shoudn't benefit from it mitigates the impact of the powerful.

@virgile-dev
Copy link
Member

virgile-dev commented Mar 30, 2018

@santisiri This a quick one just to make sure I understood the principle.
quadratic-delegation

@santisiri
Copy link
Member Author

I like the mockup, yet I think we need to convey more clearly why the additional cost. Reminds me of how ETH uses Gas for computations.

The idea I had in mind is simply not allowing intermediate values in the slider.. the slider would go from 1, 2, 4, 8, 16 and have some kind of gradient to give the idea that it is a quadratic vote this time (not a standard one).

@virgile-dev
Copy link
Member

Ok ! will work on another iteration based on those principles.
I think maybe we should still have a little debate on weather it's a good idea to implement quadratic voting on ballots. What are you thoughts @DemocracyEarth/sovereign-development ?

@AlexJupiter
Copy link
Member

AlexJupiter commented Apr 6, 2018

@santisiri @virgile-dev I don't think it should be understated how important and difficult the implementation of quadratic voting will be. I am concerned that this is going to add another degree of complexity to an already complex product.

These complexities we will solve, however it's going to take time, and we have to pick our battles carefully at this stage.

A couple of questions I have before offering a full opinion:

  • Has there been any pilots or any research into the implementation of quadratic voting?
  • If we were to work on this now, it would take resources away from the higher priority items we have identified in our product roadmap. Where would this sit in terms of priorities?

@AlexJupiter AlexJupiter added this to To do in Propose v2 Apr 16, 2018
@AlexJupiter AlexJupiter added this to To do in Delegate v2 Apr 16, 2018
@virgile-dev
Copy link
Member

@AlexJupiter The only quadratic voting implementation I know of is this one (no demo available)
https://collectivedecisionengines.com/quadratic-voting.html

I don't think we should tackle this right now. Let's prioritize the /peer view

@paula-berman
Copy link

paula-berman commented Apr 20, 2018

  • What is a monopoly in a liquid democracy? A leadership backed by sovereign individuals who can revoke their delegations at any time? If a populist leader amasses too much power and than doesn't deliver to his sovereigns' (let's resignify sovereign to the old "ordinary" citizen) satisfaction, than he will be removed from power in a minute. Look at Trump's approval ratings and imagine that reflected in delegations.

  • I don't like thinking that I will be taxed for doing the responsible and effective choice of admitting my own complete ignorance on most matters in the world and delegating to the appropriate people that I choose and trust.

  • I also am uncomfortable with the idea of taxing users unless its a necessary fee for the maintenance of the system.

Looking forward to see the counter arguments....

@AlexJupiter
Copy link
Member

I don't like thinking that I will be taxed for doing the responsible and effective choice of admitting my own complete ignorance on most matters in the world and delegating to the appropriate people that I choose and trust.

I think this is an excellent point in my opinion!

@santisiri
Copy link
Member Author

santisiri commented Apr 20, 2018

What is a monopoly in a liquid democracy? A leadership backed by sovereign individuals who can revoke their delegations at any time? If a populist leader amasses too much power and than doesn't deliver to his sovereigns' (let's resignify sovereign to the old "ordinary" citizen) satisfaction, than he will be removed from power in a minute. Look at Trump's approval ratings and imagine that reflected in delegations. I'm not worried. We should create a system that treats Sovereigns like Sovereigns and trust and empower them with freedom, flexibility, liquidity.

Check the experience of the Pirate Party in Germany that we refer in the paper. That's a monopoly in a liquid democracy. The purpose of democracy is to keep an incentive for people to participate. That monopoly led to the end of that experiment. The example of Trump's approval rating is interesting but I don't see exactly how it connects to the proposed idea. In the case of our type of democracy (which is token based), could mean having a Zuckerberg.. his approval ratings also zuck (pun intended) but he's still controlling the network.

Also the purpose is enabling a conscious thought on how participants state preferences in the system. Quadratic voting is able to capture high-signal regarding where you are putting your priorities as you participate in the collective decision process. Which can help to the idea of having delegates empowered for specific topics (and avoid great generalists running the show as is politics today).

I don't like thinking that I will be taxed for doing the responsible and effective choice of admitting my own complete ignorance on most matters in the world and delegating to the appropriate people that I choose and trust. I also am uncomfortable with the idea of taxing users unless its a necessary fee for the maintenance of the system.

I think this is a good point, agree with @AlexJupiter. I think we are all aiming for "empowering the best minds in the right topics". Since quadratic voting forces prioritization, could avoid formation of generalists.. but I might be wrong. Definitely this is the right question to pursue. Also.. I was thinking about this specifically for delegations.. but maybe could be the other way around and be for voting haha. Anyway, the idea is to always provide the options to the real sovereign: the users.

Looking forward to see the counter arguments, but as of now I personally think quadratic voting is a bad idea.

Anyway, I think this is a toggle feature that has little cost to implement but valuable insight for research. Check the paper if you haven't already.

@santisiri
Copy link
Member Author

@santisiri @virgile-dev I don't think it should be understated how important and difficult the implementation of quadratic voting will be. I am concerned that this is going to add another degree of complexity to an already complex product.

I think this is a good concern. I haven't advanced on its development so we could push it forward to a future milestone as we'll focus on Venezuela.

@santisiri
Copy link
Member Author

I'll move quadratic voting to Milestone 1.0.0 and lets take it from there.

@paula-berman
Copy link

paula-berman commented Apr 21, 2018

The example of Trump's approval rating is interesting but I don't see exactly how it connects to the proposed idea. In the case of our type of democracy (which is token based), could mean having a Zuckerberg.. his approval ratings also zuck (pun intended) but he's still controlling the network.

@santisiri What I am implying here is that in the way I see it, in a liquid democracy zuck's control of the network would be pending on his approval ratings (liquid delegations). No one gets a blank check to do whatever they want, delegates can have their sovereigns revoke their delegation at any time. So the ability to revoke can also be considered a mechanism against the formation of monopolies.

From the article you linked:

screen shot 2018-04-20 at 21 44 35

The Pirate Party example also demonstrates how users were delegating in a conscious way and "promptly withdrew their votes" once Haase voted in a way that they weren't happy with. Haase doesn't strike me as the populist leader, his delegators seemed to be critical thinkers who were monitoring him closely, and I am actually encouraged to see that this system enables someone that is a profoundly engaged scholar such as him to arise to power, which usually doesn't happen in representative democracy. LD is about empowering the smartest minds in the system, and one might argue that it did just that in Germany.

Now, even if that is the case, that is still a very specific universe and we can't affirm that this is how things will always play out. Nevertheless I think that before discussing quadratic voting we might want to investigate how liquid democracy performs in terms of formation of monopolies, to see if it is really an issue or not. Also think about how quadratic voting interacts with LD. Student-led pilots might be greatly informative in that sense.

@santisiri
Copy link
Member Author

I think the angle on...

  • Revoke as a control mechanism
  • Ability to not be conditioned if you really want to delegate power to someone.

... are strong arguments.

The goal is to prevent formation of monopolies. Revoking allows it but it focuses on the punishment side of things if the empowerment goes wrong. Whereas quadratic voting directly prevents it. I think we can keep discussing this feature and put it on hold in terms of implementation. It would be interesting nonetheless to find a context where we can test it out.

@paula-berman
Copy link

paula-berman commented Apr 21, 2018

I have introduced the topic to the students that I am working with. We could potentially deploy a separate version with quadratic voting for their pilots starting in the next school year and then compare the emerging dynamics in terms of formation of monopolies with the main version that is just based on liquid democracy.

I wonder if it would be possible to eventually merge the two, so we don't miss on the network effects of the work done by the students?

@marc-wagner
Copy link

Hi, a thought on quadratic voting in a decentralized, open network (i.e a blockchain):
for quadratic voting to work, IMHO you need to enforce delegation through
this mechanism. otherwise, people will coordinate outside of the DE framework
(through smart contracts similar to bitcoin tumblers or multisig)
and cast individual votes, achieving a bigger impact than people playing
the quadratic delegation game. Please note I am not talking about a sybil attack
here. Any thoughts on how this could be enforced?

@santisiri
Copy link
Member Author

Hi, a thought on quadratic voting in a decentralized, open network (i.e a blockchain):
for quadratic voting to work, IMHO you need to enforce delegation through
this mechanism. otherwise, people will coordinate outside of the DE framework
(through smart contracts similar to bitcoin tumblers or multisig)
and cast individual votes, achieving a bigger impact than people playing
the quadratic delegation game. Please note I am not talking about a sybil attack
here. Any thoughts on how this could be enforced?

Thanks for raising this issue up. I had this concern as well: it's the kind of idea that needs to be enforced at the core since the beginning.

@marc-wagner
Copy link

Rather than trying to block such outside collusion,
an easier way may well be to support it

through an 'everyone' contract that impersonates anonymous masses at ballots, with a weight = sum of individual votes on both sides.

@doctor-gonzo
Copy link

doctor-gonzo commented Jun 11, 2018

I think DEF uses a different definition of quadratic voting than most other people/groups that use the term
DemocracyEarth/paper#369

copy / pasted:

The way DEF seems to usually speak about QV involves the number of votes you can delegate to another person (it has to a number that comes from 2^x, so 1, 2, 4, 8, 16, 32, 64, 128 etc)

In every other instance that I have read about QV, it has to do with balancing the "tyranny by majority" effects of 1-person-1-vote and creating a mechanism by which passionate minority groups can use more than one of their votes on any one particular issue, but at progressively greater cost with each vote (for example, 1 counted vote would cost 1 vote, 2 counted votes would cost 4 votes, 3 counted votes would cost 9 votes and so on). This is the meaning employed by Vitalik Buterin and Glen Weyl in his book Radical Markets

This seems especially important if votes are able to be purchased on the open market, because it allows a dedicated minority to make themselves heard (a group of 1000 dedicated people could fairly easily amplify their voting influence by a factor of 10 if they were willing to sacrifice votes on other issues) , but does not allow tyranny by majority (a single rich person buying 1,000,000 votes on the open market could be counterbalanced by 1000 individual people casting only one vote each on a particular issue)

If DEF wants to stick with their version, I think it might be wise to change the name to something else so as not to cause confusion

@santisiri
Copy link
Member Author

hi @doctor-gonzo, yes, we are aware of that. our view is a simplification of the original QV model that deals with the accounting of votes and their price separately.. our aim was to simply make a synthesis of both.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
Delegate v2
  
To do
Propose v2
  
To do
Development

No branches or pull requests

6 participants