Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Supercomputing United States Science

Military Running a Parallel Earth Simulator 470

Fantastic Lad writes "The US Department of Defense (DOD) may already be creating a copy of you in an alternate reality. Putting supercomputers to an innovative use, the military is simulating our planet in an effort to predict the outcome of different scenarios. They might run tests to see how long 'you' can go without food or water, or how 'you' will respond to televised propaganda. Billions of nodes are created in the system, intended to reflect every man, woman, and child. 'Called the Sentient World Simulation (SWS), it will be a "synthetic mirror of the real world with automated continuous calibration with respect to current real-world information", according to a concept paper for the project. Simulex is the company developing these systems, and they list pharmaceutical giant Eli Lilly and defense contractor Lockheed Martin among their private sector clients. The U.S. military is their biggest customer, apparently now running the most complex version of the system. JFCOM-9 is now capable of running real-time simulations for up to 62 nations, including Iraq, Afghanistan, and China. The simulations gobble up breaking news, census data, economic indicators, and climactic events in the real world, along with proprietary information such as military intelligence."
This discussion has been archived. No new comments can be posted.

Military Running a Parallel Earth Simulator

Comments Filter:
  • by FleaPlus ( 6935 ) on Friday June 29, 2007 @11:04PM (#19696759) Journal
    I'd argue that that using prediction markets like the ill-fated Policy Analysis Market [gmu.edu] work much better for predicting future events. It's really too bad that there was an kneejerk media firestorm in response to the Policy Analysis Market, which killed it off before it could even get started.
  • Re:What movie? (Score:3, Informative)

    by FrnkMit ( 302934 ) on Friday June 29, 2007 @11:15PM (#19696817) Homepage
    Actually, I think it was an Outer Limits episode (the old series). Could be "Wolf 359", but it's been ages since I've seen the series, and I'm only going by the plot synopsis on Wikipedia.

    Frank
  • Re:My virtual self? (Score:3, Informative)

    by dissy ( 172727 ) on Saturday June 30, 2007 @02:43AM (#19697657)

    How do you know that you aren't a simulation for some real Joe Blow out there, hmm?
    Chances are, you are.
    http://www.simulation-argument.com/simulation.html [simulation-argument.com]

    ABSTRACT
    This paper argues that at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a "posthuman" stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor-simulations is false, unless we are currently living in a simulation. A number of other consequences of this result are also discussed.
  • by Basehart ( 633304 ) on Saturday June 30, 2007 @05:22AM (#19698181)
    Fuck The Black Swan, pick up Asimov's Foundation Series!
  • by Jeremi ( 14640 ) on Saturday June 30, 2007 @09:19PM (#19702625) Homepage
    For some reason we're supposed to believe those simulations but not to believe this simulation?


    Well, yeah.... those simulations have been tested by giving them historical data and seeing how their results "predict" what would happen next... then comparing their output to what actually did happen next. If the results differ, they refine the simulation some more and try again. Not foolproof by any means, but I don't think you can dismiss them out of hand, especially when many different (independently developed) simulations show similar results.


    It would be very interesting to see what sort of tests they have made (or will make) on this simulator to test its accuracy.


    There's no way a computer can predict the behavior of each person and there's no way we can collect the parameters that "define" the behavior of each person.


    Perhaps it works like quantum mechanics... there is no way to predict the behavior of any given electron, but if you average enough together at once you find that in aggregate they follow the (mostly) rock-solid laws of classical physics. In other words, it might actually be much easier to predict the collective actions of millions of people than to predict the actions of any particular person.


    In any case, I don't that country-by-country is the best level of granularity -- countries aren't that isolated or independent from each other anymore, so playing Risk might be unrealistic. ;^)

"When the going gets tough, the tough get empirical." -- Jon Carroll

Working...