Revolution No. 10

I have had an idea. Hopefully, someone will make a useful comment as to why it can't work, or provide a suggestion of how to improve it, or use it in a different context or situation. Whoever has watched The Social Network (and is geeky enough to understand the computer science mumbo jumbo that is mentioned throughout) might have noticed that Facemash was based on a pretty fundamental mathematical algorithm called the Elo Rating System. The Elo Rating System (ERS) is a method for calculating the relative skill levels of players in two-sided games such as chess. Performance can't be measured absolutely; it can only be inferred from wins, losses, and draws against other players. A player's rating depends on the ratings of his or her opponents, and the results scored against them. It just struck me that we could solve a lot of the disagreements in the world - in both micro and macropolitics - if we used a similar sort of system. No matter what any government/leader/person decides to do within a community, some member of that community will ALWAYS disagree with some aspect of the decision. No one is ever happy. The common perception in politics seems to be that it's also a two-sided game. Right vs Left. Rich vs Poor. Pro-abortion vs Anti-abortion. Pro-anything vs Anti-anything. Love vs Hate/Fear. Those of you who have watched Donnie Darko will argue (if you agree with Donnie) that emotions/opinions cannot be separated into two distinct categories, and there are millions of other emotions/opinions that don't fit into one of the two categories.  If we were to create a version of the ERS that looked at every possible opinion of the right way to do something in current politics (as a starting point) and asked people to look at pairs of these opinions and decide on the favourable one, then once everybody in a particular community had gone through all sets of pairs (1 vs 2, 1 vs 3, 1 vs 4, 2 vs 3, 2 vs 4, 3 vs 4) we would have a absolute perfect representation of the opinions of that community, and a winner would come out on top, with a sorted list of runners up. I think someone needs to build a web application that enables this to take place. It would be a social decision-making app.  Users of the system would select a topic (such as whether to privatise the NHS or not), and then throw into the melting pot every available opinion they can find/think of about what the government should do. On the surface the two contrasting opinions will be 'They should privatise the whole NHS' or 'They should keep the whole NHS in public hands'. If you were then to go down a level you will see that those that want to privatise will disagree on how they do it, when they do it, all manner of things. And there will be even more fundamental disagreements if you venture down further (it's a bit like dream-levels in Inception).  The app would then throw back at the users a pair of opinions (They should privatise the whole NHS vs They should privatise x part of it and leave the rest public), and the users would decide which one of the pair is preferable. It would then give them another pair until all possible pairs have been seen and voted upon. Once every user has made a decision on every pair, the most popular opinion of the 'right thing to do' will appear.  There might well be a viable reason behind why it wouldn't be feasible on a macropolitical scale (I don't know what it is yet), but within smaller communities (30 children in a school classroom, or a planning committee trying to decide the best thing to do in regards to a planning application) it would do wonders for making a fair decision quickly.  So, I need a mathematician/statistician, a political visionary, and a computer programmer. Any thoughts? FROM FACEBOOK JUST SO THEY DON'T GET LOST FOREVER

Fiona Cheung Which voting system would you use? That affects how many votes each person gets, and also the result. Also, people are blasé enough about referendums and we have them hardly ever. People are apathetic. Fiona Cheung I get what you're trying to do, its kind of like a decision bush, with branches that get smaller and more diverse along with the increasing complexity or detail of an idea. Nice idea, difficult to implement, probably would be more feasible in smaller communities, but then again, is voting mandatory? Because then the results will be skewed if not everyone votes. Also, is it any more efficient/effective than a ballot box?

Josh Alidina "There might well be a viable reason behind why it wouldn't be feasible on a macropolitical scale (I don't know what it is yet), but within smaller communities (30 children in a school classroom, or a planning committee trying to decide the best thing to do in regards to a planning application) it would do wonders for making a fair decision quickly." - What about asking everyone to raise their hands and count the number of people with their hands up? I know my answer sounds a bit crude, have I missed the point here?

Nick Miller Which voting system would it use? It is a voting system. You each get one vote per pair, until all possible combinations of pairs have been used up. If there are 30 children, and 10 put their hand up for one thing, 12 put their hand up for a second thing, and 8 put their hand up for a third thing, then the thing with 12 votes would win (although 18 children didn't want it). It's the same problem that you get with FPTP. So it's not fair. The majority have voted against a motion, but the motion still gets chosen because 12 is a bigger number than 10 and 8 separately. Same goes for the ballot box. Choice between Labour, LibDems, and Tories. 30 people voting: 8, 10, and 12. More people have voted for an alternative to the Tories, but the Tories still win. Difficult to implement yes, but with a web app pretty easy to do. If you choose not to vote then you can't affect the final outcome, but that's always going to be the case if you don't vote. I'm not at first talking about on a national scale, but for the smaller disagreements where a decision is needed quickly. By a process of elimination, everyone would decide what is most preferable action to take. And the best outcome would always come out on top. It seems so simple to me. Ben Miller Just got round to reading this properly and it's very interesting. As you propose using this system on a more local level first, it would be useful to see a local example where it might work. I think there are some confusions though - are you equating the 'right thing to do' or 'the best outcome' with the one that gets the most votes/'wins'? These aren't the same things, although you could of course argue for a democracy that worked like this. You do say that the outcome would provide a great representation of opinions, and this is certainly true - a system like this could be used for some detailed market research or polling. I'm just not sure that you can make a jump from opinions to decisions. Please correct me if I have misunderstood you. I totally agree that too many debates in this country are presented in black or white and that they need to catch up with reality, which is more complex than ever!