Quadratic voting is when you give voters 'x' voting points (e.g. 100) and voters score their options. Each increment on the score scale costs (increment^(2)) points (i.e. '+1' = 1 pt, '+2' = 4pts, '+3' = 9pts, etc)
Use-case: groups which are small enough to have mutual trust but big/diverse enough to have competing interests.
Ping me if you have a lead.
@douginamug Why the non-linearity?
@amiloradovsky Forces participants to 'economise' the intensity of their opinions: disallows participants to vote with extreme opinions the whole time. This is a known issue with score voting.
I'm interested in using it for cases where a group of people need to share heterogenous resources.
@douginamug I know a bit about consensus protocols and the Arrow's Impossibility Theorem, but never heard that voting for one option, despite being able to distribute the vote, is an issue. So far the non-linearity doesn't make any sense to me.
@amiloradovsky Hmm. Let me try to explain my experiences to see if you can relate.
I do a lot of stuff in voluntary/self-organized contexts. In these, there is an emphasis on sharing responsibilty/authority. This often means it's not clear who feels responsible for/cares about stuff. However, when topics are brought to discussion, it can be the case that many have an opinion and the person who took the initiative feels overwhelmed!
@amiloradovsky ... by forcing people to economze their strength of opinion, you may get a less biased impression of things than a group discussion could.
Perhaps that's confusing: that my main interest is not using quadratic voting to actually vote on things, but as an adjunct to collective deliberation/coordination.
@douginamug I'm not sure this is the best approach, but still this is an interesting one.
What it reminds me of is the law of diminishing returns, in this case, for the efforts spent on the probability of the desired outcome.
I don't see what's so special about the quadraticity here, and you probably want to use a more general power function, x ↦x^α, for a fixed α > 1.
I doubt there is any ready to use software for that, but making it shouldn't be too difficult.
@douginamug Either way, IME, despite all the non-linear deviations looking very cool, it's very difficult to make sense of the dynamics they describe. Even the math itself gets unfeasible very quickly.
Here you get waist deep into the non-linear thermo-dynamics and stuff — exciting, but not very practical ATM, IMHO.
@amiloradovsky Hmm, thanks for the feedback! Sadly my math skills are lacking.
I have a strong intuition that some non linear 'vote-buying' method would produce useful results for communities, but I can't prove it, just want to try.
Would you suggest something like 'dot-voting' instead? (i.e. given 100 dot-stickers, place as many on options as you want)
@douginamug Most obvious approach to me is giving one unit of vote to each voter, and the ability to distribute it among the available options however they want.
Thus we're getting a preference distribution for each voter. Then we aggregate the individual preferences by adding them up and dividing it by the number of voters.
The resources should be distributed according to the aggregated preference.
@douginamug Looking for the most and least popular option is a bad idea. Although, if you must, you can drop the least popular options, or retain only the most popular ones, and repeat.
Introducing a non-linearity is going to be a mess, and will likely lead to numerous unintended consequences. But someday it may indeed become viable, when we research the area better.
@douginamug Seems to be that. Although I haven't researched voting systems in any depth.
@imattau ahhh! That looks really great! I had a similar idea of 'jacking technology into politics' by having human reps guided by a democracy app, but they're doing it!
@imattau Do you know much about them?
I'm flicking through their white paper...
... and read
"The task of redefining democracy therefore requires dealing with this problem of authority. Until now, no system of democracy has achieved that. ... Arguably #quadraticVoting came close, but has a few problems"
I unfortunately can't contribute any leads, but I'm curious about quadratic voting. It seems to have all of the issues of first-past-the-post, except that it's not only more complicated, but also increases the need for strategic voting. Maybe I'm not understanding it correctly?
@jaxter184 My interest in quadratic voting is in a completely different context to electing representatives!
My interest is commoners/collectives sharing heterogenous resources.
I live in an experimental common house where we attempt to maximize the sharing off resources. However there are many cases where it doesn't make sense to have common rules, because of diversity of interest. In these cases it would be interesting to get snapshots of economized opinion...
For example, we have a lot of rooms https://kanthaus.gitlab.io/expfloorer/ none of which are privatized—which works surprisingly well!
But it's quite hard for people to make changes, because some people have opinions, but it's not clear who... often people just give up. But if we knew who *really* cared about the rooms, the consultation process would be easier.
I think quadratic voting is quite interesting for allocation/distribution of finite resources.
Oh that makes sense, I can see how quadratic voting improves the tradeoff between priority and vote economy over FPTP. I guess in a situation where you're not combining your vote count with other peoples', the mechanics of voting are pretty different.
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!