Found this in the news today.

[pakman]

This has got to be one of the most ridiculous ideas that I have ever seen. Let's trade on the probability that Yassar Arafat will be assassinated on September 12, 2003 at 3:34 GMT.

Here is an article describing the site and the futures and the program. This is the FutureMAP site, and this is the DARPA page.

[TaiDaishar]

Oh my god... what a bunch of dimwits...

[Skandranon]

Why's it ridiculous? Seriously.

Economically, it's known that the best way to organize expertise is in markets. This is a market of ideas, where all the analysts are told to put their money where their mouth is. The result is that, quite naturally, the best ideas which have the most solid support get the most money behind them. The analysts that are consistently right win money. The analysts that are consistently wrong lose money. All resulting in an extremely efficient way of collating intelligence information. As it stands now, people in power have to read through a number of intelligence reports organized by a number of teams, and then pick one report to believe on the basis of absolutely nothing more than "it feels more true". This way, strong propositions get strong backing.

Let's trade on the probability that Yassar Arafat will be assassinated on September 12, 2003 at 3:34 GMT.

This is a ridiculous straw-man slander of this project. The project works more like - assuming that Yasser Arafat is assassinated at some time in the future, what is the United States' optimal response? Analysts will suggest several plans, which go up on the board as "idea futures". Other analysts will suggest more plans, or will review the plans currently posted. The best plans will receive the most backing.

Read. And learn, before you mock.

I, myself, was exposed to the idea through Marc Stiegler's novel Earthweb. While it's definitely science fiction, the idea futures/information markets concept is sound.

[DarkCrown]

I have no idea whether a futures market in ideas would give decent information on future terrorist activities. I think it would depend strongly on the traders chosen and the amount of money involved.


But there is a difference between logically sound and morally repugnant. Swifts' solution to the Irish problem was completely logical.
Unfortunately that was satire and this is our government at work.

-DarkCrown

[pakman]

Then why do they have that meaningless graph? Let's assume it is correct for a moment. The expert (analysts) poll is way off the mark. This will do nothing more than just make another analysts poll. Sure, some ideas will have more backing than others, but ultimately, there will still be too many ideas to even start taking into account.. Maybe it should be an intra-government thing, not open to the public. Perhaps if it was structured in that sense, it might work.

People use futures to hedge other positions against future events and transfer known risks to speculators. So what would be getting hedged here?

The analysts that are consistently right win money. The analysts that are consistently wrong lose money

Then you better be a damn sure your speculation is good. Your're betting on ideas, intangibles, I'd rather go to a psychic to get information. Take a look at the example on this page. In most futures the variable are controllable and limited to a minimum. In this, you are dealing with the entire world. Political situation change daily, tomorrow Saddam might die or England will decide to pull out of the conflict (unlikely), but it could happen. Going back to Pete's signature a while ago, opinions are now counted by how much money people throw at it. The government is conducting a commo internet poll, but with real money. That's the only thing that makes it different. Also, have a look at what Warren Buffet has to say about derivatives, page 12. It's a risky business. The example about betting on the number of twins born in Nebraska in 2020 is how I view the PAM.

And what about if those very people we are talking about in the news decide to play the game? Take Osama BinHidin'. He sets up a third party to place bets, does something nasty, and then collects the money?

[Skandranon]

Then you better be a damn sure your speculation is good. Your're betting on ideas, intangibles, I'd rather go to a psychic to get information.

You must have a pretty low opinion of political analysts and professors if you think psychics are just as reliable. Using the example you supplied, the two people interacting are a specialist in Jordanian domestic affairs and a specialist in US military planning and operational capabilities.

What the idea market is doing is drawing together their expertise without actually having to get the two people together. In effect, it creates a "think tank" of every specialist that has access to the site. You think psychics are this reliable?

Given the example again, setting up futures on whether the Jordanian monarchy is overthrown and if the Iraqi regime persists: these are not intangibles. These are not just ideas. Take into account the stability of the Jordanian people; the strength and number of revolutionary movements within Jordan; the ability of the monarchy to maintain order; the loyalty base among the Jordanian people toward the monarchy. It is not that these things cannot be known. Just because they are to you, pakman, intangibles because you, pakman, do not know anything about them does not mean that no one does.

Political situation change daily, tomorrow Saddam might die or England will decide to pull out of the conflict (unlikely), but it could happen.

And of course someone's free to put a bet on that. And if they're wrong, they lose money. It's not, "nothing happens". There's a clear and very direct penalty for a wrong prediction. Don't disregard it.

Going back to Pete's signature a while ago, opinions are now counted by how much money people throw at it.

And what's wrong with that? You get a small return for being right and a huge loss for being wrong. The way the system works, people only bet on what they have expertise in. The system is based around everyone betting having expertise in something, because the people with expertise in nothing will lose all their money fairly quickly.

Also, have a look at what Warren Buffet has to say about derivatives, page 12. It's a risky business. The example about betting on the number of twins born in Nebraska in 2020 is how I view the PAM.

The example you refer to is just a throwaway line on the lengths to which derivatives can go. Otherwise, his comments deal mainly with the accounting issues related to the derivatives market and are therefore not particularly germane to this discussion.

This criticism of yours also implies that you believe that with an idea futures market, people will just go around betting on whatever they want, throwing their money away. I'm not sure where that belief comes from. I certainly don't like losing money and a lot of other people don't like it either, and I'm sure that my dislike of losing money won't magically go away if I'm offered the opportunity to lose money on an idea futures market.

And what about if those very people we are talking about in the news decide to play the game? Take Osama BinHidin'. He sets up a third party to place bets, does something nasty, and then collects the money?

So Osama puts up a prediction that, say, the Empire State Building will be struck by two airplanes the next morning. Puts money on that, probably a million or more dollars. The act of putting so much money on that raises the percentage assigned to that prediction. The US Government, which is watching the board (after all, that's what it's for) notices the prediction, notices the bet, and immediately alerts authorities all over the country...preventing the attack. And if Osama puts only a dollar or two on it, keeping the percentage low? I'd call the benefits of the system far better than the possibility of handing Osama one whole dollar.

The idea, which I'm still not sure you get, is that somewhere in the world or even just in the United States, someone knows part of what is going to happen for exactly the right reasons. And someone else knows another part of what is going to happen for exactly the right reasons. The information market lets them get that out where a community of potentially tens of millions can read over it.

All of this, of course, is a sideshow, even the winning and losing of money. The point is that the US government, by watching the board and the predictions that surface and become either more strongly or weakly supported, can virtually predict the future...and in a far more accurate manner than psychics. What this is, is a potentially incredible tool. Naturally, the current form as suggested by DARPA has some flaws. What doesn't, when it first starts out? But you have to start somewhere.

[Bone]

First of all, the program has been canceled.

Secondly, isn't it a bit of an erroneous comparison to draw, that of the predictive value of the current commodities futures market, which is based on a multitude of factors that affect prices over time and are available to everyone to interpret, and that of the predictive value of a futures market on one time events? Where the price of orange futures is heavily dependent on weather conditions in Florida, the value of a futures contract on the assassination of a political leader predictive factors that only few may have access too. The wide range of useful information that is available in the commodities market would not be available in the Pentagon version. On top of that, much of the information may be of low reliability, so as to only cloud the predictive information.

Thirdly, it is bad politics. If other countries began sponsoring wagers on when George W. Bush is going to be assassinated, or when members of congress will be assassinated, or other catastrophic type events, I'm sure there would be public outrage, as they should be. Profiteering at the expense of others should be distasteful. Not only that, the plan creates a financial incentive to encourage certain criminal acts.

Those are all things I don't want my government engaging in.

[TaiDaishar]

I'm even going to point the obvious flaws of such a market...

I'm going to put the moral weight on this topic, do you think that having people collecting money by some tragedy is right? Imagine people bet that a war erupts in Korea (just an example) and thousands or even milions of people die, and the people who bet correctly got money from other people's suffering.

Bone is also right about politics, it's bad politics, very bad.

[Skandranon]

I understand that the program has been cancelled. I am simply making sure it is not dismissed as "ridiculous", when it is absolutely not.

Where the price of orange futures is heavily dependent on weather conditions in Florida, the value of a futures contract on the assassination of a political leader predictive factors that only few may have access too.

That's the point. The Pentagon themselves wrote that, if you'll read their site. The idea futures market is to make sure that the few people who do know what's going on get their information out to where it can be noticed. The key concept is that someone, somewhere, does know what's going to happen. As for information of low reliability, the market system itself weeds that out. If it isn't reliable, it won't be strongly supported, and thus won't be clouding anything.

Not only that, the plan creates a financial incentive to encourage certain criminal acts.

Quotes like this make me wonder if I'm the only one who understands the idea futures system.

If someone wanted to encourage a terrorist act to earn money, the act itself would drive up the probabilities, meaning that said act would be predicted and therefore, stopped. You can't abuse it. Plus, analysts can also bet against the proposition, so it's not as if money can only be earned if bad things happen. For that matter, one could run idea futures on the best way to democratize Iraq. Would this be met with moral outrage?

I don't understand why people are focusing so heavily on the deadpool aspect. Can idea futures be used to run wagers on assassinations? Yes. Is that their sole use, or their most common use, or even a use of any frequency? No. Just about anything can provoke moral outrage; if we were to ignore things just because of that possibility we'd never get anything done.

[Skandranon]

I'm even going to point the obvious flaws of such a market...

Oh? Then point them out. Show me obvious flaws that can never be corrected.

Or are you missing a "not" there?

Imagine people bet that a war erupts in Korea (just an example) and thousands or even milions of people die, and the people who bet correctly got money from other people's suffering.

I'm actually going to respond to this, even though it's clearly a flawed example and illustrates just how little you've read or comprehended of the material relating to idea futures.

The people who bet that a war in Korea would begin did not cause the war. They did not contribute to the war and have not increased the suffering of the Korean people. They did not get money from anyone else's suffering by any means. However, what they have done is alerted the US State or Defense department that tensions are rising in Korea and that something should be done to stop it. And it might just prevent a war in Korea or contain its scale. And for that, they receive a small monetary reward as incentive. Prediction is what matters, not the money, and especially not people dying when the majority of ideas placed on the market will not involve death in any way, shape or form.

[Bone]

Not only that, the plan creates a financial incentive to encourage certain criminal acts.

Quotes like this make me wonder if I'm the only one who understands the idea futures system.

If someone wanted to encourage a terrorist act to earn money, the act itself would drive up the probabilities, meaning that said act would be predicted and therefore, stopped. You can't abuse it….

Is it your contention that the plan does not create incentives to encourage certain criminal acts? Are you further implying that I or other readers do not have the knowledge of the futures market that you may possess?

Just because placing a "wager", or a market order if you will, on a specific event would effect the given probability of that event happening, does not mean that that specific event would be prevented or stopped as you contend. I don't know how you came to that conclusion, but I doubt you'd be able to support it.

I don't understand why people are focusing so heavily on the deadpool aspect. Can idea futures be used to run wagers on assassinations? Yes. Is that their sole use, or their most common use, or even a use of any frequency? No. Just about anything can provoke moral outrage; if we were to ignore things just because of that possibility we'd never get anything done.

The "deadpool" aspect is significant. No one has contended that the sole use of this project would be to predict assassinations. However, it is always necessary to weigh the negative aspects against the positive ones. Saying that it is not practicle to dismiss something based on a single overriding moral repugnancy, is untrue. There can always be elements of a specific thing that no matter how small a percent of the total project, can be so negatively significant as to cast a shadow over the whole project. I recognize there could be positive results to this proposed project. However, I believe the negative aspects outweigh the positive ones.

[Skandranon]

Is it your contention that the plan does not create incentives to encourage certain criminal acts?

It is. Because, a fully formed idea futures market isn't about problems - it's about solutions. The bet wouldn't be "The Empire State Building will be hit by two airplanes tomorrow morning"...it would be "A terrorist event is planned on the Empire State Building...and here's how to stop it." It wouldn't pay off in the event of the terrorist event or criminal act succeeding - it would pay off only if the proposed plan succeeded in halting it.

Could that be abused? If you could find people willing to stage a mock terrorist act and then be imprisoned for many years, all for the sake of someone else's profit, yes. I doubt you'll find too many of those people.

I recognize there could be positive results to this proposed project.  However, I believe the negative aspects outweigh the positive ones.

Bone, that's all I was asking for, and I can respect that opinion. What I was arguing against was the contention of pakman and TaiDaishar that the idea was ridiculous and could be summarily dismissed without discussion.

[Bone]

The bet wouldn't be "The Empire State Building will be hit by two airplanes tomorrow morning"...it would be "A terrorist event is planned on the Empire State Building...and here's how to stop it." It wouldn't pay off in the event of the terrorist event or criminal act succeeding – it would pay off only if the proposed plan succeeded in halting it.

(my emphasis added)

Can you support this? Because it was my belief that the system paid off on events, not the successful prevention of events. How could we gauge the success of a negative event, that of something not happening?

I only have information based on reading what has been currently available. I found nothing that led me to believe that the program only paid on successful preventions. Even further, this contradicts what you said earlier, " Plus, analysts can also bet against the proposition, so it's not as if money can only be earned if bad things happen.". So if you could bet against things, how does that reconcile with your contention that the plan would only pay off for successful prevention?

[phylogen]

Hi,

I agree with any moral qualms about using idea futures to try to predict terrorist
attacks. However, I first became aware of idea futures in the context of the
relationship between politics and the financial markets. Instead of trying to
predict the timing of some vague future event, you can use futures to predict
binary (or almost binary) outcomes. For example, whether there will be a
Democrat or a Republican in the White House after the next election. Two
binary events can be combined also (e.g., Dem/Rep control in the House vs.
the Senate). These types of futures markets provide tools with which to
hedge political risk in financial markets (these were actual futures markets
I followed in the runup to the 2000 election). For more information, see:

http://www.biz.uiowa.edu/iem/

Phylogen.

Edits: typos

[Skandranon]

Because it was my belief that the system paid off on events, not the successful prevention of events. How could we gauge the success of a negative event, that of something not happening?

Note the careful distinction in my post between the current system and a "fully-formed" idea futures market. What the Pentagon proposes is far from fully-formed, because the user base is relatively small - it is not large enough to propose comprehensive plans (in which case you would not be gauging the success of a negative event, but rather the success or failure of the preventative plan). With a small user base, all you can do is act as "early-warning", and there are some flaws with an early-warning system, though, I was arguing, not enough to kill it outright. However, with an expansion of the user base, the system transforms from early-warning to a dynamic database of solutions, a virtual consulting resource involving thousands if not hundreds of thousands of experts.

What PAM was, was an embryonic idea-futures market - with significant drawbacks, but with a much higher potential payoff, especially once it transforms into a full solution market. Which is why I emphasized - one has to start somewhere.

With a fully formed market any one problem would be responded to with a number of proposed solutions. Experts, certainly not universal experts, but who contain expertise in various areas, would then comment. For example, a plan to democratize Iraq could receive input from an advertising specialist, an Islamic cleric, a sociologist, and a political scientist, then modified by engineers or military experts or people with experience in desert regions for feasibility. None of them could propose the plan on their own, but what the market does is organize their expertise in such a way that what areas they do know well are utilized to their full potential. With thousands more experts working on alternative plans or laying money on plans they support (or laying money against plans that they believe will fail), the solutions that emerge as strongly supported are the most likely to succeed, having been designed by experts and supported strongly by experts in one of the most concrete ways possible - with money.

[Bone]

Skan,

Yes, I see that you have segregated your proposal from that of what existed under the Pentagon plan. When we enter the realm of the hypothetical, you definitely have an advantage, since it's your world and all :P

Either way, I don’t see how something could pay off only in the prevention of an act. Furthermore, I don’t see how if that is possible, that you could place bets on both sides of the event, both happening and not happening, akin to the "no pass" line in craps.

If I bet event X will happen on 1.1.04, and nothing happens, do I win because something was prevented? Or was there no threat in the first place? And if I'm able to bet that event X will not happen on 1.1.04, and nothing happens, do I win because I was correct?

In any event, this now becomes a theoretical discussion, much different than the original thread intent.