There are four types of forces that affect human decisions. Genetics (nature), prior experience (nurture), body chemistry (different foods or other products put in affect responses), and external factors such as the weather, time of day, color of the room they are in, music on the radio, and on and on. If (and this is a huge if) you could know every single parameter then you could effectively predict a persons response with a reasonable degree of accuracy. But all it takes is one small variance to result in a completely different decision.

Also, @Rokakoma. I know grad schools are fairly competitive right now, but you do have the advantage of having a better education than most american applicants. Few college graduates of American colleges (even economic or business school graduates) have a thorough working understanding of mathematics. I'm preparing to move out of the states after I amass the requisite cash, and I intend to pursue post-grad in South America (Uruguay specifically, where post-grad, like undergrad, is free of charge and open to anyone who applies and can complete the coursework).

@llama projector You do realize that there are problems complex enough that exponential growth in processing power, even over a really long time, are not enough to solve them?

@llama: maybe I'm just too geek or paranoid or mathematics freak, but there were times when I had a dice with me, and I based my decisions on the outcome of rolling it. Moreover, I just adviced a cashier the other day, to use a dice for every customer, whether she enters the items or not. (She wanted to cheat taxes and I adviced she should still keep herself 'on the curve' hence the dice.

All I'm saying is a while I agree on your assesment where a male person is likely to stand on the subway, I still think there are random factor you will be never able to predict. Still, you will be right 9 out of 10 cases probably.

I also used rolling a dice technique in diplomacy, to not to be predictable, or chosing my meal, and so on. While you will be able to model that my behaviour will roll the dice in a certain situation, I'm sure you won't be able to predict my actions after.

@llama, good to hear that, and I even have a better mathematics knowledge, than most students here, as I love it, in general, and took a lot of extra courses. Still I rather expect the worst and hope for the best.

@ Tendmote: How do you know that? That seems unknowable until we see the upper-end of processing power. It also seems like we're playing favorites, and deciding that some problems are unsolvable just because unaided human brains could probably never solve them. My point is this: Any situation is reducible to it's core components (the environment, the actors, etc).

Another note: my conception of the world, and how it works, is separate from my conception of how I would rather the world worked. I would rather that human brains weren't reducible, and that "free will" actually existed, and that physics/biology worked in a way that true freedom of choice was possible. However, based on everything I know about the observable universe, this is not the case. I would be down with an afterlife, or the existence of "souls" separate from bodies that exist after our death- I miss my loved ones/Christopher Hitchens as much as the next guy. As a result, I am focused on constructing a complete enough understanding of the world for myself to keep myself safe, and maximize my lifespan (for many of the same reasons that I think predictive models will exist for human action, I think it is likely that we will see indefinite human life extension). Tendmote made a comment about betting against our conceptual models- I don't think that human beings will come up with such a model, at some point computers are going to be bright enough to improve themselves without human input, and it's likely that the resultant computers are going to be the ones that provide us with such a model.

To preempt the "omg computers will enslave us" argument, I think that conception relies on the belief that computers will act like us. Apes are competitive animals that just happen to be able to think deductively, a super-intelligent computer would have no reason to exterminate or enslave humankind, unless it was instructed to do so- I don't think human beings would have sophisticated enough knowledge to program it to do so, even if they wanted to. A perfectly rational computer wouldn't even have a survival drive, since survival drives are a mechanism of natural selection (and evolution by extent). I think it's reasonable to expect that such a machine would have no "ulterior motives".

@llama: 1st, read Asimov's Foundation, 2nd, have you ever though of turning towards writing instead of (or including) "natural sciences"?

I'm hardly a scientist, unfortunately I am one of the afore-mentioned stupid americans with a sub-standard understanding of mathematics and science. From an international perspective, I would consider myself barely scientifically literate, and only profess to mirror the opinions of experts and consensus in a particular field. Once I have the opportunity, I'd like to acquire an undergraduate certificate (or the equivalent) from la universidad de la republica in Uruguay, in edition to post-graduate studies in social sciences. As for writing- I feel like that's a career only open to people from a more comfortable starting point than me- as an American I need to contend with student debt, rising costs of living, decreasing employment opportunities and a stagnant wage scale. Since my priority is quickly amassing enough money to leave the country, I'm more worried about getting a good enough paying job for a couple of years to put that money aside. Also, I don't feel entirely comfortable expressing myself politically in public, because the apparatus exists (in the national security apparatus) for government officials to silence dissenting views- even now, the Obama administration is prosecuting journalists, and I think it's going to get worse before it gets better.

I meant writing sci-fi novels. I don't think you have to be a scientist, whatsoever, to write a good science-fiction. On the other hand, I like the way you phrase what you want to say and it seems you can think coherent enough to create an entire universe around it, regardless whether it's true or false.

It has nothing to do with politics, or journalism. Just because someone is a good journalist, it doesn't make him a writer and vica-versa, as a Shakespeare actor will never do blockbuster movies.

The point is to discover what are your skills and combine them with your interests. The rest will play out. If you are interested in social sciences, you can still combine them with writing or whatever you are good at, which doesn't necessarily have to be studying or anything related mathematics.

Nevertheless, good luck with your endeavour or continuing studying, I strongly encourage everybody who wnats to study, to do so.

Oh, fiction writing. Nah, I've never had pretensions of writing fiction. I don't think I have anything to say that somebody else can't say better, or that hasn't been said before.

it was just an idea. If you don't feel it, that's completely okay, but never be sure about the second part, that you cannot do it better, or it has been done already.

Most of the successful things around us, were already existant, and then somebody came and made a better one. From Tesla to Facebook, from Dan Brown to Rembrandt. Their "products" were all "said" by the time, yet they were able to "say it better".

You can always create new things, or better out of existing ones, never doubt that.

^Facebook is better than nothing, dude. It primarily destroys lives.

Guns don't kill people, Facebook kills people.

@llama projector Regarding computability and complexity, presumably the interconnections in the human brain would be represented by a *huge* graph. I'm not sure what exactly would have to be done with this graph to get an "answer" for behavior, but there are loads of graph problems that are intractable to compute a solution for on any conceivable timescale:

http://en.wikipedia.org/wiki/List_of_NP-complete_problems

This is the main unsolved problem with respect to computability, complexity, and intractability:

http://en.wikipedia.org/wiki/P_%3 _NP_problem

There's currently a big gap between "deterministic" and actually finding an answer for a great many seemingly straightforward problems.

(As such you can’t really say “Humans deterministic” + “computational power increasing” = “feasible predictive model for human action” within, like, the lifetime of the universe, unless you have an actual model devised that avoids all of these intractable problems)

^ P_vs._NP

it's not proven the NP != P, meaning it might happen that we will be able to calculate those problems in polinom time. Furthermore, building an actual non-deterministic turing machine might be possible, without solving NP problems mathematically.

Still, there are problems outside of NP, such as exponential ones, which still have to be solved. Or there are the double exponential ones, for which to solve you need exponential number of non-deterministic turing machines.

For me it seems rather unlikely that we will be able to model every atom of the universe anytime soon, and if we will, still there are the accidental event, like spin changes, natural fission, neutron fission, energy recombination into matter, and so on.

@rokakoma True that, true all that. I was just providing some background for llama projector, who was under the (popular) impression that problems for which there are answers are all on the cusp of being practically solvable, given enough computing power. In reality there are a vast number of problems with answers that *we will never find* given any non-infinite amount of computing power, even though we could verify the "right answer" in a moment. UNLESS, as you mention, P = NP, which is the great mystery.

The group of NP is not really a big group of problem, as it is still inside the EXP group if I recall correctly. Even if we somehow prove P=NP, we will have the exponential problems in our hands. Anyway, I agree with you, you cannot model everything by computers, additionally, to model the universe, the smallest computer capable to do that would be the exact same size of our universe. Just to start with, you cannot model multiple objects' gravitational fields, when multiple is at least 3. Gravity itself is a mystery, while a few beleives it's an interaction I tend to beleive gravity is the field itself, and no such force as gravitational exists at all. Every object is moving along a straight line with constant speed, only the field is curved around them.

Who uses facebook these days..?

Fascinating conversation.

I'd just like to mention that it is often not required to solve a mathematical problem in full. Numerical Analysis is rapidly growing along with processing power. Sophisticated approximations are often good enough to get an accurate answer in a reasonable time.

Yeah. I haven't taken a computability/complexity class in... 19 years, I think... natural selection has reduced what I remember down to what is asked in job interviews. Time to find a refresher course on Coursera! In the meantime, there's this:

http://arstechnica.com/information-technology/2013/12/exponential-algorithm-making-windows-xp-miserable-could-be-fixed/

That is *not* my fault! I remember enough to recognize O(2^n) is bad!

hey there
we already have a fb fan page for webdiplomacy.net
https://www.facebook.com/webdiplomacy
for now we are about 50
feel free to join and create a fb community, it won't hurt.

The main questions are:
i) What content should be/could be on the fb page.
ii) how can it be posted if you (hellalt) are the only admin, or who is willing to be a content admin there?

there can be numerous admins if anyone is interested.

Well we can use the page there, and if we have a group of people who are interested in helping promote it I don't think it would hurt at all.