tnf, you have a mission

[R00k]

Yea, that's kinda the point though.

[[xeno]Julios]

I've wondered before, why hasn't anybody made a simulated world in a PC -- like an FPS -- in which every natural law we've discovered is programmed? It seems like this would be a wonderful tool to test new scientific theories.

And I know that our theories and laws are tested on computers all the time already, but I'm talking about a programmed world, where every accepted scientific theory we have is applied.

It would also be useful for driving/testing computing benchmarks. Deep Blue playing chess would seem primitive in comparison, I would think.

Certain simulations would be able to elicit emergent properties that we previously were unaware of.

However, if there are emergent properties that depend on things more fundamental than the laws we program (such as the mysterious process that underlies these laws), we would not get any data. This is because we program these laws in as fundamental premises - we don't program in the "premises" that underly these laws.

Think about how we discovered the laws we have today - we did it through observation and inductive inference. The fundamental laws were found by trial and error and playing around with numbers until a pattern was discovered. We never actually explained these patterns.

[[xeno]Julios]

we can't even predict tomorrows weather, a birds wings across the world affect that. This will never be possible!

simulating chaotic systems is not inconceivable. According to some analyses, that depend on determinism, it is in theory possible to simulate such systems.

You'd need an incredible amount of digital sampling resolution, and computational power, however.

[[xeno]Julios]

In fact, chaos theory has many of its roots in Edward Lorenz's discovery of what came to be known as the "butterfly effect".

This discovery was made on a computer simulation of the weather.

http://en.wikipedia.org/wiki/Edward_Lorenz

[Keep It Real]

In fact, chaos theory has many of its roots in Edward Lorenz's discovery of what came to be known as the "butterfly effect".

This discovery was made on a computer simulation of the weather.

http://en.wikipedia.org/wiki/Edward_Lorenz

XD i felt like I was writing somebodys idea when i made that post

[R00k]

I've wondered before, why hasn't anybody made a simulated world in a PC -- like an FPS -- in which every natural law we've discovered is programmed? It seems like this would be a wonderful tool to test new scientific theories.

And I know that our theories and laws are tested on computers all the time already, but I'm talking about a programmed world, where every accepted scientific theory we have is applied.

It would also be useful for driving/testing computing benchmarks. Deep Blue playing chess would seem primitive in comparison, I would think.

Certain simulations would be able to elicit emergent properties that we previously were unaware of.

However, if there are emergent properties that depend on things more fundamental than the laws we program (such as the mysterious process that underlies these laws), we would not get any data. This is because we program these laws in as fundamental premises - we don't program in the "premises" that underly these laws.

Think about how we discovered the laws we have today - we did it through observation and inductive inference. The fundamental laws were found by trial and error and playing around with numbers until a pattern was discovered. We never actually explained these patterns.

But wouldn't it be a helpful tool in finding deeper explanations for those premeses? Seeing unexpected and inexplicable results will always lead to more research, which leads to better understanding.

[mjrpes]

I've wondered before, why hasn't anybody made a simulated world in a PC -- like an FPS -- in which every natural law we've discovered is programmed? It seems like this would be a wonderful tool to test new scientific theories.

And I know that our theories and laws are tested on computers all the time already, but I'm talking about a programmed world, where every accepted scientific theory we have is applied.

It would also be useful for driving/testing computing benchmarks. Deep Blue playing chess would seem primitive in comparison, I would think.

They do try to create programs that mimick reality. The problem is that it takes massive amounts of computation power to mimick even the physics of a molecule. 3D games like HL2 that seem to do a good job with 'physics' reduce the complexity of reality by taking an object like a car and treating it as an elemental unit, ignoring the individual molecules and atoms that make up the car. That is why when a car crashes in a game it is not realistic to life, since a ton of complex actions are going on within the car when it crashes that are too complex to mimick in the game.

About God and ID:

I feel that God is slighted when people start arguing that he is directly responsible for complexity. How much more intelligent and powerful a God must be, who is able to create a world and not have to touch it after the simulation starts. How intelligent must this God have been, to be able to define the rules of this simulation in such a way that they would naturally lead to complexity. Intelligent Design people actually dumb down God because they bring God down to the level of a human, who 'needs' to tinker with the project while it is going on in order to bring about 'complexity'.

[[xeno]Julios]

But wouldn't it be a helpful tool in finding deeper explanations for those premeses? Seeing unexpected and inexplicable results will always lead to more research, which leads to better understanding.

Perhaps. My intuition tells me that the only useful information that will be gleaned will be the type that is derived from the premises we give it.

In other words, the deepest understanding we will get is limited by the deepness of our premises.

We may increase our understanding (breadth) within the depth we give it, but I don't think we'd get clues about the deeper properties.

Might be wrong though.

[R00k]

Perhaps. My intuition tells me that the only useful information that will be gleaned will be the type that is derived from the premises we give it.

In other words, the deepest understanding we will get is limited by the deepness of our premises.

We may increase our understanding (breadth) within the depth we give it, but I don't think we'd get clues about the deeper properties.

Might be wrong though.

I tend to think designing it with our own inherent limitations, and then seeing those limitations fall short of accurately recreating what we know and expect will happen, would be a driving force that would compel us to re-examine what we consider to be basic. This is the way a lot of discoveries have been made, even if accidentally.

[R00k]

They do try to create programs that mimick reality. The problem is that it takes massive amounts of computation power to mimick even the physics of a molecule. 3D games like HL2 that seem to do a good job with 'physics' reduce the complexity of reality by taking an object like a car and treating it as an elemental unit, ignoring the individual molecules and atoms that make up the car. That is why when a car crashes in a game it is not realistic to life, since a ton of complex actions are going on within the car when it crashes that are too complex to mimick in the game.

Hence my point about driving new technologies in computing.

Surely some of the supercomputers we have now would be enough to make a start at it.

[[xeno]Julios]

Hence my point about driving new technologies in computing.

Surely some of the supercomputers we have now would be enough to make a start at it.

well IBM's blue gene is leading the way in this field, simulating the folding of proteins, and cortical activity.

[[xeno]Julios]

I tend to think designing it with our own inherent limitations, and then seeing those limitations fall short of accurately recreating what we know and expect will happen, would be a driving force that would compel us to re-examine what we consider to be basic.

excellent point. I didn't consider that way of using the simulation.

[FragaGeddon]

I have a big cock!

i think youre getting that mixed up with your new monitor.

I meant to say monitor. Not sure how cock got typed instead. Maybe I should stop stroking while typing.

[Nightshade]

A little further information on this debate: I claim that we understand things within our own framework. I keep using the analogy of the caveman and the eclipse. That is, when the first human witnessed the sun going dark, he probably shit his loincloth and thought "I don't understand what's happening, so it must be the work of some mighty god". We'll leave the proposition that cavemen had articulate internal dialogues for another thread.
My point is simply that just because we don't understand something doesn't mean that it must be attributed to some higher form of intelligence.

And BT, if you think that there's no "battle" going on between ID and evolution, I'd certainly like to hear why.

[MKJ]

A little further information on this debate: I claim that we understand things within our own framework. I keep using the analogy of the caveman and the eclipse. That is, when the first human witnessed the sun going dark, he probably shit his loincloth and thought "I don't understand what's happening, so it must be the work of some mighty god". We'll leave the proposition that cavemen had articulate internal dialogues for another thread.
My point is simply that just because we don't understand something doesn't mean that it must be attributed to some higher form of intelligence.

And BT, if you think that there's no "battle" going on between ID and evolution, I'd certainly like to hear why.

marry me

[Keep It Real]

Intelligent design (my understanding of) does not mean "no evolution", and evolution does not mean "no creator"

[Nightshade]

From what I've read, and I'm by no means an expert, there are some in both camps that agree with you. My problem lies with the extremists that not only feel that ID means accepting the existence of god, but also really want evolution removed from the classroom completely.

[Nightshade]

marry me

:lol: Sorry, my wife might object.

[MKJ]

does your wife agree with you on the matter?

if not then its clear. if God wanted you two to be together shed agree with you :@

[Nightshade]

It's not something we've really discussed at length, but I know she believes in evolution and natural selection.

By the way, has anyone read Steven Wolfram's "A New Kind of Science"? He's the guy that wrote Mathematica, and he's done extensive research into the areas of complexity and cellular automata. I'm wondering if this book is going to do much to educate me as to what complexity really is. I've perused a copy of the book and I have to say that it's a bit daunting.

[tnf]

Intelligent design (my understanding of) does not mean "no evolution", and evolution does not mean "no creator"

ID means 'no macroevolution' - they accept the idea of microevolution (for example the beaks of Darwin's finches) but they usually argue that this is inherently different than macroevolution. But macroevolution is, in a sense, simply accumulated microevolution.

ID theorists get into real gray areas when trying to explain exactly how much change microevolution can bring about. Just like they get into a gray area when trying to explain the numerous examples of bad design in nature.

And the theory of irreducible complexity - the basic foundation upon which their entire argument rests - is flawed in so many ways it is amazing it can still be perpetuated upon the public without them laughing and pointing.

So, go out and laugh and point.

More detail later NS. I have to go learn powerpoint.

[rgoer]

btw this goes back to something from the first page but I got to this thread late: it is fundamentally impossible to represent the entire universe with absolute accuracy using any mechanism smaller or less complex than the entire universe itself

[[xeno]Julios]

btw this goes back to something from the first page but I got to this thread late: it is fundamentally impossible to represent the entire universe with absolute accuracy using any mechanism smaller or less complex than the entire universe itself

dunno about that - people like wolfram believe that you can model some very complex and dynamic processes using very simple rules.

Look around for flocking behaviour demos for an example.

So there may be grounds for a "compression" between software and output.

[rgoer]

yes that is familiar to me and is all well and good, but even the cleverest of compressed models wouldn't provide absolute accuracy

of course, this is silly to debate, because I doubt humans would ever be able to do anything with an absolutely accurate simulation of the entire universe, anyway

[[xeno]Julios]

yes that is familiar to me and is all well and good, but even the cleverest of compressed models wouldn't provide absolute accuracy

one way to get around it would be to have only portions of the entire universe simulated at once (if you were to do the whole thing simultaneously, it would seem that the representation of the output would require the amount of particles in the universe).

This is sorta like Chomsky's competence/performance distinction.

We may have the ability to create an infinite variety of sentences, but that doesn't require an infinite brain capacity.