« Meaning and Language, Part 1 | Main | Obama/McCain Dance-Off »

Oct 22, 2008


Dave Jones

An interesting proposal. And you're right, nobody will adopt it -- anytime soon.

A few comments:

(1) The proposal assumes that a nation-sized group of people will read a pamphlet about twenty issues. If we could actually get people to do that, the governments we already have would probably be fine.

(2) The proposal presumes that "past performance is a predictor of future results". In other fields where people attempt to predict future events (stock-picking and horse-track betting to give two examples), this tends to not be true. If someone predicts that Lucky Lady is going to win the third race at Belmont Park and he's correct, how much of it was skill and how much of it was luck? By the same token, if someone predicts a war will be over in two years and he's correct, how much of it was skill and how much of it was luck? With horse-track betting, you have a lot of knowledge of the system before the race, yet there is still plenty of luck involved. With complicated things like wars, anti-poverty programs, education systems et cetera, you're going to have much less knowledge, relative to the size of the system, going in than with horse-track betting, so there is going to be a lot of luck involved. Should people be "rewarded" for being lucky?

It reminds me of a story about Enrico Fermi. Fermi was talking to a general one day, and he asked the general, "How would you define a great general?". And the general replied, "A great general is one who can win five evenly-matched battles in a row.". So Fermi asked how many generals were great generals, and the general replied, "About 3%". The general was not very happy when Fermi pointed out that 3% was the number you would expect simply from random chance (50% ^5 = 3.125%).

(3) Suppose the voters vote "yes" on a highway project between Houston and Memphis:
-- What would you do if all of the people who were well-qualified to implement the highway voted "no"?
-- Suppose I'm selected to implement the highway, and, after doing some research, I realize that I was wrong when I stated that "it will only take $30 million". What happens then? Do we re-vote? And, if we proceed, do I get the additional funding? or does the project go bankrupt? or do we have to vote on that?
-- Suppose I'm selected to implement the highway, and, after doing some research, I decide that the highway is a bad idea. Can I call a halt to it? Can I call a re-vote?

That's my two cents worth. But it is a great idea, and I'm happy that someone is sitting around thinking about these things. Technology does make this sort of thing possible, so this seed you've planted might bear fruit sometime in the future -- probably long after we're gone, but that's all part of being human.


Who gets to decide what constitutes a valid issue worth voting on? And who gets to do the analysis to come up with valid, simple, multiple-choice answers and reasons? Those people will be the ones with the real power in such a system.

Sonny Lacey

Hey Graham, good to see you branching out. Reminds me a bit of John Muir's writings (not the conservationist, mind you, but the VW mechanic) about government in his tongue-in-cheek work "The Velvet Monkeywrench."

I do wonder, however, if weighting a person's vote based on "correct decisions" could be confused/equated with a voter who simply makes good gambles (but perhaps that is the same). What is more, we all know that reasons and outcomes can be subject to some rather heavy-handed revisionist history. Without true and objective recording of events, the system is prey to the journalists.

Keep up the work! I like to see new forms of governing brought up.

Peter Yared

Hmmm... seems like someone could "correctly" predict that for example continuing racism is a good thing for years and years, and then when the tipping point of a new generation of people hits, their vote is not equal to the older folks who have been "correct" for years and years.

Graham Glass

Hi Dave,

Thanks for your comments. Here are my responses:

1) As far as reading booklets go, that's the way it already works in California and San Francisco. Each election we are sent a booklet containing the major propositions. It seems to work well. I'm not sure whether other states work this way.

2) I think that over time, being correct wins over randomness.

3) I don't think it's likely that there would be no candidates for a particular outcome. In addition, there is no re-voting.


Graham Glass

Hi Mike,

You raise good questions. Since the state of California and the city of San Francisco already have a process for voting on issues like I suggest, I presume my system could borrow their approach for deciding on the issues.

As far as the outcomes part of the process goes, perhaps that could be decided by a vote as well? Most propositions I've read specify the reasons and outcomes for the proposition, so perhaps we'd include the reasons and outcomes from both the 'for' section and the 'against' section of the propositions.


Graham Glass

Hi Sonny,

Great to hear from you!

I think the results/outcomes questions would have be picked to be fairly easy to measure. The Iraq war ones were fairly straightforward for example. I bet similar points could be defined for healthcare, etc.


Graham Glass

Hi Peter,

I doubt whether "continuing racism" would be a proposition!

That being said, results/outcomes would normally be based on a 4-8 year time horizon.


Peter Yard

If your system was in place in the 40s, 50s, and 60s, there would have been proposition saying something like "Schools should continue to be segregated". We have a proposition like that right now about gay marriage in California. Should we weight the votes of the people who have been "correct" on past social inequality since they had majority votes over younger people who are more open to change and progress?

Graham Glass

Hi Peter,

A couple of things.

1. I removed the whole vote weighing thing from the current article because it's detracting from the core idea.

2. Weighting is based on whether you predict the *outcomes* of the issues, not whether you were in the majority. Since one of the main precepts of wanting to make gay marriage illegal is protecting traditional marriage, one outcome would presumably be something like "the number of divorces goes down because traditional marriage is saved". If you vote against gay marriage, you'd at least have to commit to one or more predicted outcomes which could then be measured. That being said, it's probably too complicated and detracts from the main idea behind my post so maybe it's just best to bury this particular variation ;-)


Tim Farage

I generally like the idea. I asked my University class today if someome could get elected President telling the truth about how they really will deal with the issues. They all thought not. So not having political parties whose goal is to get their candidated elected, and not necessarily to do what is good for the country is a good idea.

At the same time, some issues are very complicated (e.g. Social Secuity) and take a lot of work to come up with a potential solution. So how about if we elect representatives (using the Instant Runoff System) and they come up with solutions. But then their solutions go before the people who then vote for or against them?

That way the power would still be with the people but you would still have elected 'experts' who actually try to come up with solutions that people will vote for.

Tim Farage

The comments to this entry are closed.


  • Destiny is my science fiction movie about the future of humanity. It's an epic, similar in breadth and scope to 2001: A Space Odyssey.

    To see the 18 minute video, click on the graphic below.