Skip to content

Max from Flux has a detailed response to Nick's analysis "Evaluating Democracy Reform Proposals"

Max's Response

Just some comments

Issue-Based Liquid Democracy with tweaks

-- Nick_Merange

Haha, I wouldn't put it like that (you could just as easily say direct democracy with tweaks). Given how strongly some ppl feel about IBDD I think putting it in the delegative democracy bucket isn't great, except maybe as a hierarchical thing. Issue based specialization and trade isn't something that anything else offers, and is pretty core.

(In future, it is likely to be possible for people to create their own bills, with votes available for each of those)

-- Nick_Merange

This is going to be a very early or launch feature. The votes available for member-suggested bills are not pre-parliament, but rather because those bills are seen by parliament.

Can choose to give your vote to someone else who votes on your behalf, and you can take back your delegation at any time

-- Nick_Merange

This also delegates political capital. I'd say "Voters can choose one or more delegates (so long as the total doesn't add up to more than 100%) and delegates receive the voter's resources (votes and political capital) to the proportion selected, with any "change" being returned to the voter. Additionally voters can optionally select to withhold an individual vote from being transferred to delegates if they'd like to vote personally."

Can choose to not vote, and instead gain political capital that can be spent later on future votes (How much political capital can be saved, whether it becomes less valuable, and if so over what period is something unresolved right now as far as I can tell)

-- Nick_Merange

We have fairly solid ideas, but open to changes / improvements / experiments. This goes to the "political capital" market category of questions. Voters bid on unused votes in a market / auction system using political capital. The proceeds are then proportionally distributed to the users who opted out of that ballot.

There are 3 ways to acquire political capital: * signup bonus, everyone gets, say, 1000 political capital tokens when joining * basic income - there's a ~30% inflation rate on political capital tokens, and all voters receive more political capital on a frequent basis (maybe daily) with 100% of newly created PC tokens being distributed in this way * opting out of issues - votes go into an auction market with the PC/vote "price" being decided by the market

Plan to implement a system for bidding for political capital, where supply/demand of political capital on a particular vote determines the value of each political capital - the more people are bidding for tokens, the lower the value of each token used, and the higher value given for saving tokens

-- Nick_Merange

This is a core requirement, so "plan" is only true in the sense that we're planning all of the above. Also I think you have PC and votes backwards: supply/demand for each vote decides the value of each vote in PC tokens. The more ppl bidding for a vote the higher the "price" of that vote. etc.

Evaluating Democracy Reform Proposals

-- Nick_Merange

Have a few problems with your methodology in this section. Will respond to your reasoning first:

Generally I think direct democracy produces worse decisions because it is vulnerable to mob mentality leading to policies created in the heat of a moment, that, on reflection, should be dealt with in a more appropriate way

-- Nick_Merange

This is, I think, a good crit of DD. But IBDD is anything but DD. A lot of consideration at an epistemic level has gone into thinking about how IBDD can vastly outperform existing systems. Main mechanics here are specialization and trade, opportunity cost, feedback loops, and error correction. IBDD works by having short cycle times and allowing groups to dynamically form around issues. The trade components mean making worse decisions is a sub-optimal strategy, and the ability for anyone to suggest an edit means we can facilitate fast error correction (with short cycle times). I've written a lot about this here: http://xk.io/2017/05/27/ibdd-and-poppers-criterion/

Balancing this direct democracy approach in Flux is that it enables delegation of votes and gives political capital to people who elect to abstain from voting, meaning that specialisation of interest can occur

-- Nick_Merange

This is required for better decisions.

This said, maybe that will make the system even more vulnerable to mob mentality decisions

-- Nick_Merange

Error correction + trade means even if this does happen it's quick to undo and mobs that form around this will waste political capital, so it's suboptimal.

not necessarily those who are most informed or think it is a difficult issue to resolve and would prefer a more nuanced way to resolve a problem

-- Nick_Merange

voters are free to organise however they want, so this is out of scope for IBDD. E.g. a group of scientists can get together and use deliberative democracy to design some policy.

I might be underrating direct democracy’s effectiveness and increased the score of Flux on this metric

-- Nick_Merange

If you're rating DD and not IBDD then your reasoning will be flawed. IBDD is a world of difference. For example, I expect very few issues will have more than single % point participation since most issues aren't that relevant to the majority of ppl.

I’m going to settle with 4 for the moment, because I’m worried about whether the optional voting system combined with the ability to gain political capital will merely incentivise emotional responses being fanned by media for people to spend this capital on their preferred response

-- Nick_Merange

As I explain in the article you link, this is not only suboptimal but actually a very bad strategy. Additionally, the media thing isn't comparable to situations today because right now media argue vote for party X over party Y because we need solution A not solution B; in IBDD-land the media how have to say vote for solution A over all other solutions you think might work in all other areas of politics**. It becomes personal, which is a much harder sell.

Also, this part isn't good reasoning - you're basically saying "because this feels right" instead of giving a more solid argument.

Example: can you come up with situations where self-allocation of (real world) capital is generally worse than centralized allocation of capital? This is important bc much of the economic reasoning behind why IBDD can work well is similar to explanations behind capitalist progress (which is considerable).

I think it's funny that you give ODD a better rating than Flux considering your reasoning :P

Also, I think it's misleading you sort of "average" out results based on uncertainty - if you have uncertainty you should record that. I.E. giving IBDD a 0-10 range for "better decisions" is more honest than a 4 since the reality is we don't know yet (though have good reasons for believing it).

Responsive to Crises?

-- Nick_Merange

Ironically I think this is a much bigger challenge for Flux than other systems because we have a more sophisticated voting system that has more requirements around the cycle time. We've had ideas on this in the past but not settled on anything that feels "true" to IBDD itself.

There is a risk for Flux’s approach that stores political capital, as accessing stored cryptocurrency has been able to be exploited by hackers.

-- Nick_Merange

This is identical risk to any online voting system since it's about holding private keys.

MiVote does not store capital, so is much more secure.

-- Nick_Merange

This doesn't relate to security in a technical sense. You could argue PC means IBDD systems are more likely to be targeted, but that is independent of the security of the platform.

TBH I don't think you know enough about this topic to make any useful comments. You also don't consider the IP in voter anonymization SecureVote holds and that Flux will leverage.

Might get to replying to more later, but need to do some work now.

Nick's Reply

Hi Max,

Thanks for reading and the comprehensive response and rebuttal. It was also good to hear more of the detail of what Flux wants to do, as I wasn't sure on all the details.

My concerns about whether Flux's approach will lead to better decisions share some characteristics with concerns I have about Direct Democracy, but they do take into account the elements your talking about. I didn't spend time addressing those in detail in my original post, but I will try and do so more now.

Basically I think issues that are likely to have feedback loops (awareness that a policy is bad) that are weak or slow are likely to be a source of weakness for better policy under Flux's proposal, which I argue happens quite frequently.

First, let's look closely at the idea of 'error correction'. I will define it as a process in which - a group of people judge that a policy is bad for some reason(s) - That group proposals an alternative policy that addresses their concerns and the concerns of the original policy proponents

I agree with your logic behind flux being more effective, but: - I don't think the feedback loops will be as strong as you think they will be - error correction will take a longer time than you're estimating.

I also think that there are certain technical areas such as tax policy or budget allocation that will suffer until people realise that solutions that satisfy both parties by effectively paying the losing side out will have negative externalities on other areas in which people want to spend money. Let's say live cattle exports. Maybe all sides will be happy with the live cattle exporters getting compensated in exchange for not exporting live cattle. This would not be a very fiscally responsible approach, but there are few people who will care about that in the context of resolving this issue. But then more and more issues come up that have similar trade-offs, if our answer is always 'buy the losing side off', then the aggregate decisions made destroys the budget if we want to spend that money on other things. It's inequitable also if simply because the live cattle exports issue got raised first, say cattle exports get banned but they get paid out and another situation later doesn't because people identify the problem with paying out policy 'losers' by that point.

Proposals that have small, tight feedback loops with fewer confounding factors will be the best kinds of problems to deal with under IBDD, others might take quite a bit longer. I think also you've got higher confidence than me that people will care about the policy externalities.

I definitely think you have a strong argument and may be right that the elements and processes of political capital allocation you describe. The more I think about it the more it would make sense that people would be less antagonistic in their policy-making. I guess what I was distracted by was the fact that this won't remove antagonism from political campaigning - you'll still have adversarial politicking by two sides that want different things (e.g. Live cattle exports), but that won't influence the policy in the long run, as long as someone can actually be creative enough to come up with a solution that satisfies most of what all sides want.

The major points of disagreement I think we have so far is about: - The ability of people to identify 'bad policy'. You seem to be equating 'bad policy' with 'policy that is created that people don't like'. This is preference-based policy-making, but that will not necessarily create better policy, just more agreeable policy. I expect things such as mandatory sentencing laws, more police and a restriction on skilled migration will become the new policy norms, as they are already. This is aligned with people's preferences and beliefs about what dangers they perceive to themselves and how they believe they can be protected from those dangers. The only problem is that our perception of fear does not necessarily reflect the reality of how likely it is we are attacked or whether there is even any effective action that can be taken to reduce the chance of this further than is already the case. In this instance, you'll have policy that is based on placating people's feelings, rather than effective policy (maybe the best way to reduce crime is reduce inequality).

I also do not have the same confidence in the market mechanism that you do, especially when it comes to emotional issues. Maybe it will encourage more rationality by encouraging people to instead fight the right battles for themselves. I think you mentioned that politics would become more local, not sure that's going to work great either, because sometimes we need to see the broader context - e.g. we could ban all houses above two storeys in our neighbourhood, but then if everyone does that, it will have a negative externality on utility, transportation and public service costs because density will be so low that the city will spread rapidly.

These kinds of things are happening already in the current system, but I believe I have given plausible examples of issues that would not be resolved any better under IBDD. There will be more experimental programs that will flourish under IBDD, but I want to highlight that there may be some issues that resolve better than others under this system.

I am worried about the aggregate effect of low percentage engagement decisions too, and I think if implemented, it will be critically important that the media analyses these proposals closely, so it avoids corruption and reduces hidden externalities.

I will end my response there for the moment, hope that clarifies some of my thinking on this.

Max final reply

Basically I think issues that are likely to have feedback loops (awareness that a policy is bad) that are weak or slow are likely to be a source of weakness for better policy under Flux’s proposal, which I argue happens quite frequently.

-- Nick_Merange

You have to contextualise this within the competition.

Is this more powerful than direct democracy? Yes (because it's not a static set of voters for each issue)

Is this more powerful than parliamentary rep democ? Yes (because we don't have to wait 3 years)

Is this more powerful than MiVote's model? Yes (because it's permissionless and doesn't require long negotiations and cycles times that MiVote's model does)

So to argue that Flux is less effective than the above is logically incongruent unless you can refute any of the above points.

error correction will take a longer time than you’re estimating.

-- Nick_Merange

I don't think I've ever estimated a time - EC is continuous and ongoing...

EC is weak in DD.

EC is a bit better in MiVote and Rep Democ but not much.

EC is continuous and ongoing in IBDD.

I don’t think the feedback loops will be as strong as you think they will be

-- Nick_Merange

They're stronger than in alternates.

I also think that there are certain technical areas such as tax policy or budget allocation that will suffer until people realise that solutions that satisfy both parties by effectively paying the losing side out will have negative externalities on other areas in which people want to spend money.

-- Nick_Merange

EC and the fact that all policy is treated as one ecosystem fix this - the only stable equilibrium is sensible tax legislation. See my essay re IBDD and popper's criterion.

Also, your argument is a straw man. It's "theoretically possible" (e.g. the system allows it bc to be compatible with progress systems have to be general and non-biased towards the nature of solutions) but there's no good economic (within the IBDD economy) reason this would be done.

I think also you’ve got higher confidence than me that people will care about the policy externalities.

-- Nick_Merange

That's because I understand the ibdd-economic implications of externalities - they bring more participants into the issue which makes it harder to pass or keep that sort of policy.

[...] as long as someone can actually be creative enough to come up with a solution that satisfies most of what all sides want.

-- Nick_Merange

This is generally true, just not instant.

The ability of people to identify ‘bad policy’. You seem to be equating ‘bad policy’ with ‘policy that is created that people don’t like’.

-- Nick_Merange

That's one of the reasons people would care about it, but it's not how I define bad policy and it's not what I think the emergent behaviour will be. For an overview of the epistemology and system see https://www.youtube.com/watch?v=Zq25UXc_ONg ("The Case for Flux")

I expect things such as mandatory sentencing laws, more police and a restriction on skilled migration will become the new policy norms, as they are already.

-- Nick_Merange

Really? You think, given the chance to specialise, there wouldn't be significant opposition to this? Plenty of ppl have been affected by bad immigration law (including businesses) and there are public groups dedicated to trying to improve our legislation on these fronts.

I also do not have the same confidence in the market mechanism that you do, especially when it comes to emotional issues.

-- Nick_Merange

Emotional issues will be problematic on day 1, and day 10, but not day 100 (i.e. EC over time means acting emotionally is suboptimal, which reduces your ability across the board to be effective)

I am worried about the aggregate effect of low percentage engagement decisions too, and I think if implemented, it will be critically important that the media analyses these proposals closely, so it avoids corruption and reduces hidden externalities.

-- Nick_Merange

Haha, you want more media in politics?

Low engagement means a lot of things, low % of ppl voting personally does not mean low engagement in IBDD, and we actually expect this due to specialisation on a per-issue basis. Analogy: as a % of population ppl working in agriculture has massively declined over the last 100 years but our ability to produce food massively increased.

Anyway, I think you need to look a bit at your explanations and make sure the "I'm scared of what I perceive as worst case" theme is removed. Epistemically this is called the "precautionary principle" (see The Beginning of Infinity for a deep explanation to why this is a very bad philosophy). Rather, I'm claiming we're basing all this on economic principles, so the best criticisms you can offer are "this economic arrangement leads to some bad result X", and particularly for the killing blow you'd need to show that the ability of IBDD to self-modify with low resistance (when the idea is good, esp in cases where there are multiple IBDD instances) somehow is incompatible with fixing these problems.

IBDD has never been put forward as "the answer" or a panacea. Rather, it's the seed that can flower into a system that continuously improves without need for massive social organisation.