I asked a closely related question here: How would our democracy change if we had quick, reliable, accurate means to instantaneously vote on issues online However, I feel I left it too open ended and should focus on more specifics.

The premise is simple, imagine we were able to use the internet to allow everyday citizens to cast effective instantaneous votes from home which were secured and trustworthy (i.e., it stopped voter fraud and voting as someone else). Also assume that the poor have roughly the same ability to file a vote as the rich or at the least lack of internet access/computer does not restrict their ability to vote.

My question previously was how would this affect democracy. The consensus was that a purely direct democracy was not possible and honestly that was my belief with the question as well. However, if we had the above and confidence in it, I feel like the increased ease and speed of voting (and weren't prevented from doing so out of simply inertia causing us to stick with whatever currently is in place) I feel like we would find some way to better utilize the ability compared to our current approach.

Thus, my question is how could you make a representative democracy which utilized safe instantaneous votes to better allow voters to express their desires without forcing voters to vote on every little (or even most) decision.

The other major issue with the above question that people had was that an average voter can not be expected to fully read and understand a bill due to the complexity of the legalese, elected representatives usually have numerous legal aids to translate these things for them which common folk won't possess. Thus any solution should also minimize the risk of encouraging/forcing people to make decisions they are incapable of understanding (at least not to a larger degree then current democracies already do).

This question is about creating any democracy, be it a modification of our current democracies in the near future, or a new democracy from scratch which was built with the presumption that instant-votes are possible.

How would we correctly adopt instantaneous voting to increase voter voice without overwhelming voters or degrading our system? Would we stick purely to our current systems, except that votes are easier during elections? Would we stick purely to representatives but vote them in/out more often? Or can we create a system that allows more direct feedback on occasions?

Perhaps representatives may choose to open certain votes up to public; for example the countries laws require direct vote for a declaration of war. Maybe individuals have the option to directly vote on any congressional decision, but representatives effectively control the 'block' of votes that are not explicitly made by direct votes for a particular decision. Maybe some other more interesting ideas I haven't come up with yet etc?

I'm interested in knowing why an approach does or does not work. Why would option x be chosen over pure representative or pure direct democracy, or better yet what tradeoffs are made with option Y (ie, in some ways Y hurts the democracy, in some ways Y helps, I think Y will be chosen as ultimately better option then X despite the potential harm).

'Votes' do not need to be limited to a specific Yay or Nay on a bill either. Anything that could be used to express a legal voice online may count, including petitions, Impeaching unpopular representatives, call for votes on certain issues or vetoes of existing votes, or even the ability to write a bill or propose an amendment for an existing bill, if it's deemed the option could be made to work in an effective manner; though all of the above do not need to be possible if some are deemed too disruptive. In any case the key difference from today is that the 'voter' is uniquely identified in such a way that he can not misrepresent who he is, what district he is from, or get more then one 'vote' on any given issue when serving a legal capacity as a voter.

edit: let me just say all of my comments below, now and in the future, are mostly about elaborating on ideas. I keep being afraid people will see them as dismissive of good ideas rather then constructive attempts to elaborate to handle potential complications. So I'll put a blanket comment for all here, my asking about potential draw backs is not a disagreement with the ideas I'm commenting on at all. I really love the feedback so far and look forward to more :)

  • 1
    $\begingroup$ We could use a system like stackoverflow or fixtomorrow.org so that people can bring up issues and then politicians could vote on them. If politicians actions on these issues were tracked we could create a system for removing them if they aren't following the people's will closely enough. $\endgroup$
    – Varrick
    Commented Aug 28, 2015 at 14:52
  • 1
    $\begingroup$ See “delegative democracy” at Wikipedia or elsewhere, e.g. realized with Liquid Feedback. $\endgroup$
    – Crissov
    Commented Nov 3, 2015 at 20:00

5 Answers 5


To start with, let's define the 'most efficient form of government' as being ' the one which most closely represents what a direct democracy would look like if every citizen always voted on everything after researching and becoming aware of the intricacies of that particular piece of legislation.'

So how do we go about doing this? The easiest way is to use a form of representative democracy in which every voter can freely choose his or her representative and change that choice at any time, and for any issue.

"But wait," you might say, "Our current government is a representative democracy, and it's barely functional. How is your solution going to fix this issue?"

*How will we structure our government?**

We'll try to fix this by making the following tweaks to our democracy:

  • Each voter can choose anyone as their representative, even themselves.
  • There is no limit to the number of representatives.
  • Each representative is given a portion of the vote proportional to the number of people who have voted for them.
  • Each voter can choose to change their representative at any time for any specific vote, any specific issue, or permanently.

In addition, we'll make some changes to the nature of how our pool of representatives, which we'll call congress, votes on issues. These are necessary because we'll being doing away with the partisan machinations that currently decide on things like which bill gets voted on.

  • Each representative can cast a vote every day on which bills should be voted on, and can split their vote in any way they wish. Any representative can propose a bill at the 'what to vote on' stage, though they are required to vote on it in order to do so.
  • Any bill which receives more than 20% of the vote on any given day will be voted on in one month, during which time alternative versions of the bill may be drafted. These alternate versions of the bill must address the same topic.
  • Each vote shall consist of a preliminary and a final round. In the preliminary round, in which members of congress will vote using a preferential voting system to choose which variants of a given bill are most popular. The final round shall use a preferential voting system to choose the best among these variants and a null variant. (Voting for the null variant is equivalent to voting against the bill.)

One thing to note is that none of these votes takes place in a single physical space. Every delegate will use our secure system of instantaneous effective voting to vote remotely, which is good, since there will likely be thousands of delegates.

There are additional details to work out, of course, such as whether there exists some form of fillibuster to protect the interests of the minority, and how this fillibuster would work (see comments), but this is the general structure I would use.

Why is this a good structure?

Everyone votes through their first choice representative on every issue. This is a huge change from our current system, in which a little over half the population votes through a representative they like marginally better than one other person. There is no need to compromise because a representative isn't electable. There is no gerrymandering. Furthermore, on any particularly important issue, each voter can vote exactly as they wish, regardless of whether or not they agree with their representative.

Everyone's vote carries the same weight in this system. The only thing that being more politically active will get you is a more precise application of that voice. This is an improvement over direct democracy, in which people only have a voice if they vote.

The voting process is structured so that the bills the populace views as most important will be voted on. If a particular issue is being stalled by the representatives, only 20% of the populace needs to raise their voices in order to bring it to the floor.

In addition, it's simple. Anyone can understand exactly how their vote influences politics, which should lead to less discontent about how everything works.

  • $\begingroup$ I think this is a pretty good idea mostly. My only catch is that so many voters involved in every step of the process seems likely to bog down the system, look at our current tendency to filibuster, now imagine if you have 5 times the representatives. Also who will introduce bills, and will too many variants get introduced. I think the idea is still good, but they may have to have a slightly more complex system to allow efficiency with such numbers. Maybe some reps are elected as having higher privileges then just voting (like bill proposal), how to avoid stalling by filibuster though... $\endgroup$
    – dsollen
    Commented Aug 28, 2015 at 12:59
  • $\begingroup$ oh sorry, to clarify I want to repeat that I do like this idea. My elaborating on pros and cons is not a dismissal of the idea :) oh, and I would also say that at any time a voter may vote on a given bill personally, separate from any representative, if they so choose. Or allocate their vote between reps. "this guy is for social issues" and "this guy is for financial ones" and "this is a special interest guy to represent my particular interest in topic Y" sort of thing. Each gets the person vote for specific types of votes, but voter only picks them once and specific which topics they $\endgroup$
    – dsollen
    Commented Aug 28, 2015 at 13:01
  • 1
    $\begingroup$ I'd thought about allocating votes for topics, but I'm not sure its a good idea. The main issue is that, in many cases, bills are not clear cut as to exactly which topic they fall under. Does a bill covering the financing of an EPA project count as an environmental or a financial bill? The only solution seems to be to divide up a persons vote to every representative in a topic covered by the bill, but that introduces a potential for political gamesmanship, in which each representative is incentivized to get as many bills as possible declared as belong partially to their field. $\endgroup$
    – ckersch
    Commented Aug 28, 2015 at 14:37
  • $\begingroup$ For fillibustering, I don't think it would work to have a single representative be able to do so, but I could see a system in which everyone is given a 'fillibuster' vote each day, and if some threshold (say, 35%) of the representatives choose to fillibuster a certain bill, the vote on it is deferred for a week. There's other issues with it, though, such as how many times can the same bill be fillibustered, that seemed more technical than I wanted to go with my answer. I'll add some more detail on the bill proposal bit in my answer. $\endgroup$
    – ckersch
    Commented Aug 28, 2015 at 14:43
  • 2
    $\begingroup$ For old fashioned fillibustering, one solution would be to let representatives vote as soon as the floor opens on a bill. They can choose to wait and let someone talk on the issue, and then vote, but once enough votes are in that the remaining votes won't change the winner, voting closes. This process can happen in parallel for all issues being voted on during a given day. Voting on all bills that have been selected to be voted on in a given day will also be closed at the end of the day, with uncast votes not counting towards the result. $\endgroup$
    – ckersch
    Commented Aug 28, 2015 at 23:55

The way I would handle this is to have voters set policy direction and priority - not vote on individual bills.

In other words voters would say "we want to ban smoking in public places Y/N". Depending on how that vote goes appropriate legislation would then be drawn up to enact the law, and the strength of the law would be influenced by the strength of feeling.

Having said that you should look at some of the direct democracy experiments in Iceland and other countries recently though. For example Iceland cloud-sourced their new constitution.

See some related articles: http://www.truth-out.org/news/item/19200-direct-democracy-in-reykjavik-the-wisdom-of-the-icelandic-crowd-sourcers


  • $\begingroup$ okay, but what forces do you think the voters would have to enforce their vote? for example lets say that the voters all voted that they think congress should make less money per year, but none of the representatives chosen really want to lower their salary. What prevents the representatives from stalling/boycotting the direct democracy policy indefinitely by never writing up the law? (this isn't to say I dislike the proposal at all, just trying to reason through implications and requirements to make it a consistent world. I honestly am interested in suggestions for addressing my concern) $\endgroup$
    – dsollen
    Commented Aug 27, 2015 at 20:23
  • $\begingroup$ Because the record would be very clear and trackable to show what has and has not been enacted from the proposals. Politicians with a track record of not doing what they are supposed to lose their jobs. $\endgroup$
    – Tim B
    Commented Aug 27, 2015 at 20:41
  • $\begingroup$ @dsollen Do you need to force your votes, or can you just encourage the representatives to go along with it? If you need to force it, then you have to be very careful that it actually knows enough to do what is needed, because all safeguards have been removed. At least if the representatives find a way to do something exploit that is bad, the people can rebel. $\endgroup$
    – Cort Ammon
    Commented Aug 27, 2015 at 21:38
  • $\begingroup$ @CortAmmon I hear what you and tim are saying, and I don't entirely disagree. My real catch is honestly my not trusting the research of average voters. If a rep doesn't stick to the opinion voted on will voters realize it and change their votes rep accordingly? Sure someone may try to calculate an "average times he voted against public opinion" numbers, but since bills don't directly correlate to public opinion that will just lead to bickering over rather or not vote X was or wasn't in support of opinion Y. I fear lack of voter follow through, and lack of choice for voter to register ... $\endgroup$
    – dsollen
    Commented Aug 28, 2015 at 13:15
  • $\begingroup$ displeasure by picking an alternative (right now after primaries you get only two choices with Hugely different values, and even before primaries you have limited options) will lead to a system where some, or many, representatives see opinion poles as...well today's opinion polls. Somewhat useful for gauging what to do when they are undecided, but not all that binding if the rep feels strongly about an issue....or is encouraged to feel strongly with sufficient pork. $\endgroup$
    – dsollen
    Commented Aug 28, 2015 at 13:17

Use the power of the market. Vote on values, bet on beliefs.

The key point here is that individuals don't take much responsibility when they vote, especially if the opportunites to vote are both plentiful and involve little effort.

However, when you put your money where your mouth is then one is likely to do more due diligence and is more likely to feel sufficiently well informed to suggest the right policy. Its all about having skin in the game.

Votes are used to bubble a subset of many policy proposals to the top.

The top policies are put to a commercial betting market in which investors/speculators sell and purchase contracts that pay out based on measurable outcomes of the policy at some defined future date.

The policy that actually gets written into law is that one which the betting market selects.

When the day of judgement for the policy arrives the people who made the right call in the betting market get paid by those who made the wrong call.

See: http://mason.gmu.edu/~rhanson/futarchy.html

One can imagine various ways to arrange this but the details are quite involved so I refer you to this article which does address issues such as the zero sum nature of the prediction market:


Basically whether this idea works for your setting depends entirely on you - YMMV. However a society equipped with advanced interconnected IT, artificial intelligence and so on deployed to make this scheme of governance work does IMO have legs as an SF background.

  • $\begingroup$ Umm, maybe I'm missing something, but I don't see how investors are being incentivized or moving a policy forward. How is 'betting' on something giving any source of income, beyond what you may win from the other betters (which means worse then a zero sum game, since the net value of the market can never go up, but the expense of maintaining systems to record and handle bets means a slow drain of the betting pool money towards logistical costs). In addition, what good does an investor purchasing a contract due towards making that 'contract' actually become a policy? $\endgroup$
    – dsollen
    Commented Aug 27, 2015 at 20:19
  • $\begingroup$ I updated the answer to address your points... $\endgroup$
    – rumguff
    Commented Aug 27, 2015 at 21:33
  • $\begingroup$ Who defines what the 'societal goals' are? This system also seems to heavily favor the wealthy. Monsanto can afford to drop ten billion dollars into buying favorable policy, while the urban poor cannot. $\endgroup$
    – ckersch
    Commented Aug 27, 2015 at 21:51
  • 1
    $\begingroup$ I think my main issue with this solution is still that anyone with sufficient money can still buy legislation. I can buy some 'remove all worker safety laws' legislation for ten billion dollars, fully expecting it to harm society, if I plan to save more than that amount in workers comp claims over the lifetime of the bill. Add to that the fact that someone with lots of money is more likely to be able to hire a good set of lawyers to argue that this bill actually does benefit society, based on some other metric (such as industrial profits), and you're left with a plutocracy. $\endgroup$
    – ckersch
    Commented Aug 28, 2015 at 14:50
  • 1
    $\begingroup$ Why not? Those same rich organizations have enough money to mount a propaganda campaign and mobilize a network of supporters. Especially in a country like the US, with strong conservative/libertarian movements, it would be easy to mobilize people to 'get the government out of our businesses.' Even if they're outnumbered 2:1 in actual voting, they don't need to win, just to get enough votes to put the issue into the betting market. $\endgroup$
    – ckersch
    Commented Aug 28, 2015 at 15:06

First of all I will use the definition of the best democracy given by ckersch, the best democracy being 'the one which most closely represents what a direct democracy would look like if every citizen always voted on everything after researching and becoming aware of the intricacies of that particular piece of legislation'.

What can help you to achieve such a democracy ?


If you select truly randomly (i.e. uniformly) a sample of the population (say, 1 000 people out of 1 000 000), with enough people in it, they will most likely be truly represetative of the overall population. This is a statistical principle.

So what will you do ?

  1. Select a random sample of the population of 0.1 % of the total.
  2. During several days, force them (as a civil duty) to adress and solve a set of problems (by political means). Since they have a reliable way to vote and (I assume) discuss online, there is no practical problems.
  3. During the last day, make them select a set of problems to solve for the following random sample.
  4. Go back to 1.

What is it a good way to make a democracy ?

  • Most likely, the people will accurately represents the whole population.
  • The 'government' change very rapidly and its composition impredictable, making corruption impossible.
  • Being choose for 'government duty' is rare enough to really envolve chosen people, and to make them work full time on the problem without crippling their life.
  • People can be chosen in small enough number (or randomly split in subgroups) to allow fruitfull discussion, and ensure occasion to exprim themselves for everyone.
  • Process can be totally automatised.
  • $\begingroup$ Hmm, an interesting idea. Though I would worry that there would be too many political ploys to pick 'random' samples that suit your needs (look at gerymandering for examples of playing with supposedly representative values). I also wonder if picking people who don't usually vote or have an interest in it would be a good thing. Would they be prone to picking at random, or based off of what their 'party' tells them, rather then taking time to understand a bill? In short, how do you make sure the random sample is taking time to make a truly informed decision? $\endgroup$
    – dsollen
    Commented Aug 31, 2015 at 14:21

Any change is ultimately a function of the magnitude of the driver of change, and the time in which the driver can meaningfully act. The ability of the masses of a nation to case a secure, fair vote at any time from an Internet-connected device is a fairly significant driver.

What might immediately change is that opinion polls, gathered real-time, might be used by representatives in government as decision-making tools. A U.S. House Representative might poll his district's residents for their opinion on a measure. Within five minutes, he gets enough responses for a representative sample that will indicate the general level of support or rebuke of the measure, which will likely influence his vote especially if it's wildly to one side or the other. This of course is predicated on universal participation; in reality you tend only to hear from people who have a strong opinion on the topic at hand, the guys in the middle typically don't weigh in.

If and when this system becomes binding, perhaps as a voting method, probably the first thing you'll see is a dramatic increase in participation in elections. If your phone gave you a notification that it was election day for some level of government, and all you had to do to cast your vote is tap on the notification and then on your choices, you'll see turnout double at least, especially at lower levels of government and earlier stages in the electoral process.

I agree you're probably not going to achieve a direct democracy, because of the previous weakness; even if the ability to participate is right there on your phone, people will typically only weigh in if they feel strongly about the subject. That means you'll get a binding vote on the matter that is skewed in favor of the voting bloc that's more mobilized for or against the measure. While that's not all that different from the voting systems we have today, the ability to instantly vote basically turns each vote into a caucus; the more outspoken side wins, not the true majority.

There really isn't a fix for that (and some would argue it's not a problem). The only real solution is to make voting an obligation, not a right. There's a real difference there; anything you have a right to do, you usually have a right not to do, inherent in a "right" being something you can do without permission from anyone else (including the government). If you required people to vote on every legislative action, it would bring productivity to a halt and many votes would be indistinguishable from random activity.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .