Forgot your password?
typodupeerror
Cloud Supercomputing Science Technology

Simulating Societies At the Global Scale 64

Posted by Soulskill
from the how-hard-could-that-be dept.
An anonymous reader writes "Teams of European researchers are vying to create a distributed supercomputer of unprecedented scale to analyze the data that streams in from hundreds of devices and feeds (mobile, social data, market data, medical input, etc) and use it to 'run global-scale simulations of social systems.'"
This discussion has been archived. No new comments can be posted.

Simulating Societies At the Global Scale

Comments Filter:
  • Anybody guess the sponsors ?
  • This reminds me this excellent book: http://en.wikipedia.org/wiki/Simulacron-3 [wikipedia.org]

  • by headkase (533448) on Friday May 06, 2011 @05:32PM (#36052070)
    We just need more data to tease out the statistics in: Psychohistory [wikipedia.org]. Now, is that a good thing?
    • by jrronimo (978486)
      Maybe we could set up two separate populations for further testing... each at opposite ends of the galaxy...
      • Oh, so you mean, in the same place? ;)
        [Note to mods: if you haven't read Asimov's Foundation series, or at least Foundation and Empire, just skip to the next post.]
        All humor aside, I've often wondered about 'psychohistory.' It seems to have a decently solid basis in sci-fi theory, which in the context of "good sci-fi," means that it's at least *plausible.* However, without good models and the necessary empiricism to evaluate them, we don't really know _anything_ for sure---everything is really just specul

        • Re: (Score:2, Insightful)

          by Anonymous Coward

          [Note to mods: if you haven't read Asimov's Foundation series, or at least Foundation and Empire, just skip to the next post.]

          Good advice for anyone

          What's your native language - Fortran?

        • by Rockoon (1252108)
          You missed the part in the story about Psychohistory being constantly tweaked from the top BY A MIND READING ROBOT.

          It is much easier to take a 'thinking' sample of a population if you can walk into a crowded area, such as those on the most populated planet in the galaxy, AND READ EVERYONES THOUGHTS.
          • Can you point me to where in the series that's mentioned? I *sort of* remember something like that, but all that's coming to mind is the mind-reading human that confounded some of the best minds (outside of the monastic psychohistorians) in the Empire.

            • by Rockoon (1252108)
              R. Daneel Olivaw was the robot.
              • Oh, that's right! I forgot,thanks =)

                Since you're clearly familiar with the series, I'm curious---what did you think of Forward the Foundation?
                To me, it had a different feel (besides the fact that it was a prequel)---maybe a different pace, or a bit different writing style. Not in a bad way, but just something I noticed.

    • Re:Psychohistory. (Score:4, Interesting)

      by Tynin (634655) on Friday May 06, 2011 @06:51PM (#36052746)

      We just need more data to tease out the statistics in: Psychohistory [wikipedia.org]. Now, is that a good thing?

      Asimov on psychohistory

      "Well, I can't help but think it would be good, except that in my stories, I always have opposing views. In other words, people argue all possible... all possible... ways of looking at psychohistory and deciding whether it is good or bad. So you can't really tell. I happen to feel sort of on the optimistic side. I think if we can somehow get across some of the problems that face us now, humanity has a glorious future, and that if we could use the tenets of psychohistory to guide ourselves we might avoid a great many troubles. But on the other hand, it might create troubles. It's impossible to tell in advance."
      - http://en.wikipedia.org/wiki/Psychohistory_(fictional)#Asimov_on_psychohistory [wikipedia.org]

      I tend to agree with him. In the end, the information is out there, and someone is going to put it together. To what ends remains to be see.

  • by Ichijo (607641) on Friday May 06, 2011 @05:34PM (#36052088) Homepage Journal

    I want a large scale social simulation to be used as a test bed for proposed legislation, to give an idea whether the bill might have the desired effect and to ferret out any unintended consequences. Legislation really ought to go through the whole engineering process, not simply thrown into production without any testing.

    • Why not just turn off the computer, and systematically try out each new legislation in a different state/province/city/etc. before adopting it for the whole of the Nation -- Multiple trials can be executed simultaneously if needed.

      There's really no reason the whole of the nation (or its economic future) should be at stake due to crappy laws... Oh, that's right -- Equal rights means everyone must all have the same rights always everywhere or else --- or else -- or else state/county/township and other loca

      • by Alex Belits (437) *

        Are you a moron?

        To make any kind of meaningful experiment, you have to prevent interaction with the outside world -- otherwise all you will see is people exploiting the differences. This is also a reason why plenty of laws that would make sense, are not implemented on US States' level -- because then hordes of people will find a way to abuse the difference between that state and its neighbors. What also means that "states' rights" are a ridiculous concept, and Americans would do better by focusing on improv

    • by Anonymous Coward

      i hate to poop on slashdot's parade, but this whole thing has been tried before. and it is a disaster because it always ends in massive corruption.

      for example. the CDO market was heavily built on simulations; simulations created by legions of 'quants' (math PHDs) who sat around studying models all day long for banks and hedge funds and credit ratings agencies.

      the problem with these models were wrong. why were they wrong? it was not because of 'honest mistakes'. it was not because someone forgot to carry a 2

      • by Ichijo (607641)

        The problem then is not in creating simulations; it is figuring out how you create simulations that will not be abused and perverted by sick people who want to destroy the economy for their own ideological reasons and/or personal financial gain.

        Actually, you WANT to create a simulated world where the players are allowed to do those very things. That's how to find problems with the proposed legislation.

  • by cosm (1072588) <thecosm3@EEEgmail.com minus threevowels> on Friday May 06, 2011 @05:36PM (#36052102)
    This sounds almost akin to the Venus Project [thevenusproject.com], although a little less 'revolutionary'.

    Venus' concept is a massive global supercomputer network that monitors the worlds resources, allocating them only where they are needed and in reasonable quantities, eliminating waste and misuse, but being auditing and controlled by human-elect. A different future society (although it is debatable between dystopian and utopian) could automate everything, doctors, lawyers, manufacturing, almost absolutely everything once the infrastructure is in place, and people could live simple, happier lives and not be wage-slaves. Granted it would probably a century or two of automata innovation to make something like that happen, but it would beat having such excess waste, such as cars/drivers ratio [about.com]. It would be pretty neat to do what you love and love what you do without a lot of the extraneous worries.

    And no I am not a communist/socialist, just saying it might be a cool alternate reality.
    • Well, if and when we have automated AI at that level, it would be a true form of Communism in the way Carl Marx would approve of. It would, in the history of mankind be the first successful implementation of it once the human element is removed from the vacuum of power. Assuming that's even possible. But a couple of questions still nag me.

      1. How will the AI judge supply and demand when it's in complete control of resource allocation? How will it enough what's too little and too much?

      2. With such a system in

      • by 0123456 (636235)

        Well, if and when we have automated AI at that level, it would be a true form of Communism in the way Carl Marx would approve of. It would, in the history of mankind be the first successful implementation of it once the human element is removed from the vacuum of power.

        Communism can never be 'successfully implemented' without killing off all the productive people who refuse to be enslaved... and then it collapses because there's no-one to do any useful work. So it's a pointless exercise.

        • Re: (Score:2, Insightful)

          by Anonymous Coward

          Although I think the robo-communism idea is another useless exercise in naive fantasy, you are clearly just another political parrot who spouts off his pre-packaged, trite expositions whenever he hears the relevant key word. In this case, you heard "communism" and proceeded to explain why communism hasn't worked in the most cliched, tired way possible. However, your exposition is not really useful in the case since the GP was discussing a unique scenario in which people would not be in charge of the state

        • by cosm (1072588)
          That is another big question. In an almost completely automated world, would there be enough voluntary contributers doing it for the love of science and humanity to prop up the 90% of the population who sit at home glued to 'Ow my balls'? Who knows.
          • Re: (Score:2, Informative)

            by Anonymous Coward

            Prop them up how? If the "means of production" are completely automated it shouldn't be necessary for anyone to prop up anyone else. This is actually how Marx envisioned Communism, except he used the example of people being free to hunt and fish. So few people seem to actually read Marx or understand his analysis of history, it's like we already live in an Idiocracy...

            Quick lesson: According to Marx there is a kind of trajectory to the history of economics (or, political economy, if you prefer). This trajec

            • by Anonymous Coward

              Unfortunately, all too many that have attempted communism attempted to leap-frog the foundation of capitalism. In fact, they would pro-actively attempt to prevent capitalism. In what would be the ultimate twist of irony, the very miss-guided goals of enacting communism lead to fascism and poverty.

              Humanity. For the love of all that is holy; never ignore human nature. To do so often leads to disastrous results.

            • by cosm (1072588)
              Prop them up in the sense that the 10% that contribute to the well-being of humanity would be outnumbered 9-1 by those who literally sit at home and undertake nothing but pure self-indulgence. If comparing those who contribute to the longevity of the race to those who do nothing by fulfil their own pleasures to the concept of propping up is apropo, I duly apologize.
              • Re:Venus Project (Score:4, Insightful)

                by Alex Belits (437) * on Saturday May 07, 2011 @04:37AM (#36055174) Homepage

                The situation now in so-called developed countries is approximately this:

                1% contributes in any meaningful way,
                90% does what a machine would do better but a human has to because otherwise he will have no money and no means for survival,
                9% actively tries to steal from everyone else, 1% (out of the aforementioned 9%) succeeds and controls at least 50% of everything that people need to be productive, 8% (out of the same 9%) fails but still shits everything up.

                Letting 90% just sit on their asses and do nothing would be a great improvement.

      • by cosm (1072588)
        Those are definitely the two major arguments I think. In regards to resource allocation, theres what I like to call the 'greed' problem. Commune A puts in a machine order to X units. Commune B decides it needs X+1, a runnoff occurs and at some point the machine has to intervene and make a decision as to who gets what.

        As far as innovation goes, I think from a science standpoint, core science would still innovate, those like Curie, Einstein, Newton, Tesla, Feynman, Planck, etc, most innovated not for monet
      • by Dr Max (1696200)
        The op has a great point. Given this deus ex machina capable of reading everybody's wants, needs and capabilitys it could solve a lot of problems. If the computer was great enough to correctly calculate the absolute best allocation of resources (humans, machine, mineral, land, water, space) its possible we could all live in a world where you get everything you ever wanted and work in your dream job. However i doubt we have that level of computer power yet, and there is no way to be sure there is an allocati
      • by Dr Max (1696200)

        2. With such a system in place, won't that eliminate incentive based human goals that leads to innovation? Are we sure that we want to hand over that role/power to a machine? That's an awful amount of trust to place in what amounts to a demigod. In effect, we are trusting it to decide our future and shape of civilization. Could be a good thing, could be bad.

        I agree the trust issue would be hard to over come and it could defiantly be a bad system completely ignoring the needs of the few. But given a correctly programed demigod, incentive could be replaced with more of an honor type system if you have everything looked after for you (well cooked food, shelter, fast internet, entertainment, recreation, retirement) all you have left for life is your achievements. This doesn't solve all the problems of the less glamorous jobs (which more and more should be done by

    • Venus' concept is a massive global supercomputer network that monitors the worlds resources, allocating them only where they are needed and in reasonable quantities, eliminating waste and misuse, but being auditing and controlled by human-elect.

      Putting aside the totalitarian distopia this would imply, the idea is simply unworkable for lack of global information. The same issue applies to all central planning, really. Economically efficient allocation of resources depends not just on historical data you can study and model, but people's current preferences. At best you can study people's past choices and build a society ideally suited to them, but by the time you've done so those preferences have inevitably changed—and the first real indicati

      • by cosm (1072588)
        I concur.
      • by Alex Belits (437) *

        Translation: "When I will be rich..."

      • by kvezach (1199717)
        The distributed approach—a market economy—solves the problem by accepting that information is never perfectly complete or accurate, but tends to be more complete and accurate at a local level; in short, people look out for their own interests far better than even the most benevolent central planners (or AIs). A high degree of global efficiency is an emergent by-product of distributed local efficiency.

        Couldn't the AI act as the superstructure of that distributed approach? As it is, in the ide
        • An AI could definitely help advise the humans actually responsible for making decisions regarding allocation of their own resources. To an extent, we already do that—consider the algorithms in charge of high-speed trading. However, that is hardly the same thing as placing an AI in charge of allocation of all resources throughout society, without regard to their actual ownership.

  • It's called BOINC [berkeley.edu].
  • Reminds me of that old science fiction joke:

    The world spent untold sums of money hooking up all the computers together into a massive global supermachine.

    The day came when it was time to power it on. The most revered scientific mind flipped the switch and asked the first question:

    "Is there a god?"

    A lightning bolt came from a massive power terminal and fused the switch shut.

    Supercomputer: "There is now!"
  • After applying massive investigation and resources into studying human trends and consequences of human actions, beliefs etc., it is likely the the best conclusion will be that human activity and its consequences are completely unpredictable, irrational and largely of no value to anyone outside of the human species.

  • ...they should contact Magrathea [wikipedia.org]

    I believe they have experience with this, but they're not cheap.

  • I have read about this in various James P. Hogan novels. It always turns out bad . . . for us.

Never buy from a rich salesman. -- Goldenstern

Working...