How do you know that you can freely think?
Does the computer know that it''s just a pile of metal?
Do we know what exactly we are?

Are we not just lots of little ifs and switchs, albeit a lot?

Awareness is nothing more than what we know we should and shouldn''t do, given any situation, based on what we know of the world around us.

The computer doesn''t have eyes. It has a camera. It can have hands, provided we teach it how to use them (this is instinct, btw). It can have smell, chemical analysis of the air. It has hearing, a microphone.

We think of ourselves as ''analog''. Are we really? Don''t our nerve endings simply convert ''the feel of something'' to pulses our brain can work with? Doesn''t our brain simply send pulses to our body? Is this not what the computer does?

The only difference here, is that we are alive, and that the computer can never be. Hence the term artificial.
Because we, ourselves, can really ''feel''. We have consiousness. The ''life''. Sight impulses are sent there. Where would one send sight in a computer program? And would anything really ''see'' it? Or just evaluate it and generate a response?

Yes, we CAN create an artificial world, with artificial plants and artificial trees, aritificial physics, creatures, humans. But would those human ever really ''see'' anything? Of course not.
We can create an artificial world, but no one can ever ''live'' in it.

Please note that I am not stating any of the above to be the absolute truth, more likely than not I''ll be proven wrong within the next few posts.

Anyway, this is my opinion, and I''d love to know what you think.

The_Minister
1C3-D3M0N Interactive

The_Minister, INHO, virtual humans in an artificial world would not ''see'' more or less than we do. Perhaps the whole universe is a computer simulation *g*

Visit our homepage: www.rarebyte.de.st

GA

Hi,

I think that one actually can develop quite complex creatures with a computer but
the question is whether it will take thousands of years or less.

Yes, electron, I believe that your idea might work but it would take considerable
cpu-time to make the creatures smart, and I mean years!!!
The reason is that you create them with next to nothing pre-programmed. They must learn
everything by themselves. What''s more, they don''t have any "idea" of their goals (yes
I know that''s the point) but they might evolve to be a lazy creature only dragging
themselves to food when necessary, and "sleep" there until next food-wandering. And I
assume that''s not that we want, or?

Do you agree?

Subject put aside, I think this is a great way of brain-storming ideas.
It doesn''t matter whether the ideas are wrong or right in my opinion, as long as
they carry the discussion to a point further ahead.

/If God is cloned, is he still God?

/If God is cloned, is he still God?

Oboy, i never thought so many would answer on my post!

First i''d like to answer ddnewbie about my thoughts about the program: You''re absolutely right. I want to create a virtual world (Where the creature can live) with, for example, food.
Next, i want to create the "structure" of the creature,
simply a class like:

class Creature
{
private:
unsigned char matrix[1000000];
public:
void startme();
};

That''s all!
And, for startme(), it''s only a "decoder" for the matrix, that execute the right command, and head on.
NOTE!!
The commands is NOT advanced, like eatfood();
NO! Every command is only a tiny bit of code, some sort of assembly, but for the creature and it''s surroundings.
Now THIS is the hard part: creating theese tiny bit so that the creature can have as huge sortiments as possible.
Anyone got tips about this one? =)


...And for Houdini about your thoughts of emotions and the talk about "The creature cannot actually live in an artificial world", i''d like to quote Morpheus in "The Matrix":

" What is real?
How do you define real?
If you''re talking about what you can feel, what you can smell, what you can taste and see, then REAL is simply electrical signals interpreted by your brain. "




Electron

"Who need more than 640kb of RAM?" -Bill gates -89

A lot of this is philosophy, so there''s little point going much deeper here. I will say a couple of things however. Too awkward to quote, sorry: the original posters know who they are.

"Your abilities to learn, to think, to think about thoughts or thinking about whatever you think about is "coded" in the genetic code, which was developed by evolution. Electron''s idea was to emulate this evolution. "

Not gonna happen. Your neural net can analyse the nodes of the neural net, but it''s not gonna be able to read the code that forms those data structures and therefore can never see where the ''rules'' are coming from.

"Are we not just lots of little ifs and switchs, albeit a lot?" and "We think of ourselves as ''analog''. Are we really? Don''t our nerve endings simply convert ''the feel of something'' to pulses our brain can work with? Doesn''t our brain simply send pulses to our body? Is this not what the computer does?"

The mind is not just electrical impulses, it is also largely based on chemicals (such as neurotransmitters) too. The amount of chemicals flowing around makes a lot of difference. In fact, it has been shown that (in animals, at least) reducing the electrical activity to nil in the brain does not wipe the memory or kill the creature. The electrical side of it is perhaps the processing side of it, but the data is (at least largely) chemical, and thus analog. (This is where someone shouts ''fuzzy logic''! )

"It can have hands, provided we teach it how to use them (this is instinct, btw). "

Instinct had to be learned somewhere too. That is assuming we all evolved from single-cell organisms at some point. If you teach instinct to the machine, you are bypassing evolution. If you are willing to do that much, then you may as well just program an expert system, which certainly is nothing like life, but appears ''intelligent''. If you are looking for something ''life-like'', the only way you could really justify it is by showing the path of evolution from single cell to intelligent actor.

Personally I think these efforts are doomed to fail as there are no realistic and yet worthwhile ways of simulating natural selection. We have no idea how many trillions of ''species'' of single-cell organisms have died out in the past, indicating that nearly all ''realistic'' attempts will end in failure. And let''s not forget that the majority of lifeforms that actually made it this far would not really exhibit any kind of easily measurable intelligence. Also, most of these systems make no allowance for communicating information between lifeforms, or (if you are looking to human-level intelligence) ''cultural transmission'' (the ability to pass down info from generation to generation without genetics, ie. language).

And, of course, we are assuming evolution accounts for everything. Which is the most logical argument, but we don''t know for sure.

I think it might be fun to play with for a while, but I don''t think anything truly ''intelligent'' or ''alive'' is possible with any combination of current AI methodologies and current technologies.

quote:

---

Not gonna happen. Your neural net can analyse the nodes of the neural net, but it''s not gonna be able to read the code that forms those data structures and therefore can never see where the ''rules'' are coming from.

---

Kylotan, there''s also a way to simulate this aspect of intelligent life:
Don''t start with emulating living beings but a universe only with its physics. If there are the right conditions, life will appear and if it becomes intelligent it''s going to examine itself. These living creatures won''t know that they''re just a computer simulation - like we don''t know if our universe is a computer simulation or not. They will never see were the physical rules of their world come from - like we can''t in our world.

Visit our homepage: www.rarebyte.de.st

GA

I know that instinct must evolve as well, sorry I musta left that out.

And yep I had completely forgotten about chemicals and fuzzy-logic etc, but they can all be simulated.

ga: You misunderstood my meaning, when we look around we actually see, we don''t just evaluate. Those signals go from the outside, to our eyes, to out brains, to our consiensness (sp?). Computers can not have consiensness, as this is what makes us alive. Which computers aren''t.

Quick plug here: Some updates to my site, check em out

The_Minister
1C3-D3M0N Interactive

Ok, i''m gonna ask a question that will lead to an even MORE complicated issue on this post! (NOOO! a lot of people now shout =)

Do we really have to emulate our own world? Why don''t we give a creature an x,y,xspeed and yspeed - based world to live in?
That''s what evolving is all about! to let a creature adept to it''s environment. A sea creature don''t look and act like a land creature, do they?

It''s all about where we will draw the line to "life". From a falling rock to the "intelligent" human.

I''d say "life" is a series of procedures and methods, EVOLVED BY ITSELF!!

quote:

---

ga: You misunderstood my meaning, when we look around we actually see, we don''t just evaluate. Those signals go from the outside, to our eyes, to out brains, to our consiensness (sp?). Computers can not have consiensness, as this is what makes us alive. Which computers aren''t.

---

The_Minister, what is consciousness?? There are just physical reactions in our brains, and we call it consciousness. But they are deterministic and nothing which couldn''t be calculated by a computer.

Visit our homepage: www.rarebyte.de.st

GA

quote:

---

Original post by ga

There are just physical reactions in our brains, and we call it consciousness. But they are deterministic and nothing which couldn''t be calculated by a computer.

---

Can you back this statement with any evidence or hypothesis?What you have said is certainly not ''accepted'' scientific knowledge.