Feeling Strangely Skiffy

[akirlu]

So, today the news is full of the fact that the Pentagon is fielding the idea of using market investors to predict terrorism

You know what this is, don't you? It's John Brunner's Delphi pools, from Shockwave Rider. I re-read the book recently -- it's the Book of Honor for next year's Potlatch -- and was in general impressed at the high accuracy of a lot of the book's predictions (not, of course, that futurology is what SF is primarily about) but the concept of the Delphi pools seemed a little dotty. Shows what I know.

Though, to be perfectly accurate to the book, the new markets would have to be designed to keep bettors distracted from the covert machinations of the government. That'll never happen.

Comments

[redbird]

I find myself wondering whether bettors who did particularly well would be facing visits from the FBI. Also whether a terrorist organization could fund its attacks by betting on the sure thing of "there will be a bombing at X location on Y date."

[the-gardener.livejournal.com]

The whole concept is stark raving bonkers. Have a look at some of the comments already made about it in this LiveJournal.

On the face of it, the scheme is completely self-defeating. Place a bet on the likelihood of a particular terrorist event, and if others then do the same the security services will swoop to prevent it and the gamblers will lose their stakes -- meaning that, rationally, they wouldn't stake anything in the first place, so nobody would have any inkling that an event was about to occur and the security services would not therefore be able to act in time to prevent it.

But then the term "military intelligence" always was a bit of an oxymoron, what?

[marykaykare.livejournal.com]

You seem to be assuming that the Pentagon is not dotty. Are you sure you've thought that through?

MKK

[filkerdave.livejournal.com]

OTOH, the scheme has apparently been called off, which makes me feel slightly better.

[numbat.livejournal.com]

As was mentioned in another LJ discussion of this it really is beginning to sound like the John Brunner century. To paraphrase him from The Shockwave Rider, a most disturbing thing to find in your e-mail...

[hal-obrien.livejournal.com]

On the face of it, the scheme is completely self-defeating. Place a bet on the likelihood of a particular terrorist event, and if others then do the same the security services will swoop to prevent it and the gamblers will lose their stakes -- meaning that, rationally, they wouldn't stake anything in the first place, so nobody would have any inkling that an event was about to occur and the security services would not therefore be able to act in time to prevent it.

For various values of "self-defeating".

More than anything, this reminds me of Calvin and Hobbes up in a treehouse the sort of spooks who say, "We can't declassify that -- the adversary will know we're watching, and will call off their attack."

Let's assume such a market is accurate enough, consistently enough, that security forces do sweep in and prevent attacks before they happen once prices hit a certain threshold. How is this a bad thing? What are the tradeoffs in such a scheme vs. much more intrusive and counter-Constitutional means that could be used instead? Even if one counts in the monetary value of lost wagers/futures, would that be more or less than the monetary losses incurred by an actual terrorist event?

Re-phrase such futures as "insurance premiums", and you'll perhaps see what I'm getting at.

Mind you, such a market would be subject to the whims of quasi-random events, just as any market is -- see Nassim Nicholas Taleb's Fooled By Randomness (http://www.amazon.com/exec/obidos/tg/detail/-/1587990717/qid=1059631493/sr=8-2/ref=sr_8_2/104-0742173-7075155?v=glance&s=books&n=507846) -- but this is no different from anything else.

wow.

[leftwing.livejournal.com]

I think you're my hero....!!
//chet