Perhaps we should have sorted out the definition of "intelligence" we all can agree on first.

This is a very good idea. I would say the definition of intelligence is the ability to apply existing knowledge to new situations and to be able to acquire new knowledge.

Is just giving a definition enough though? We can obviously use that definition to show that computers as we know them are not intelligent. While they can be made to look like they are learning, this is purely as a result of clever programming, using tools such as conditionals and lookup tables. But what about us? While we may feel that we are not programmed and that we really are learning from our experiences, we don't know enough about how the brain works to definitively say that any decisions we make are not just a result of predetermined arrangements of neurons.

Or what is knowledge?

I would describe knowledge as a set of data that describes a specific scenario.

I always considered wisdom to be the tactical use of knowledge to make the best decision in a given situation. I considered knowledge to be stored information. Information that is not stored would not be a part of your "knowledge database"

I've always thought of wisdom as being an accumulation of knowledge. If you consider someone like Albert Einstein, it's very evident that he was an extremely intelligent individual. Now consider him as an infant - while he would have been just as intelligent, he would not have been wise as he did not have enough knowledge to use his intelligence. Likewise you can have people who know a lot of things but who lack the intelligence to take what they know and apply it to new situations.

Now, I am not being philosophical, but these things seem like they would be important in this conversation. Do computers forget? How much info can they retain?

I don't think the ability to recall information should be considered intelligence, though it does obviously affect the decisions an intelligence will make.

I think this topic is flawed from the very beginning, because each one of us has his own definition about what human intelligence is.

For some, it's the ability (the state), of being able to accomplish a given task (playing chess, whatever). For some others, it is the ability to learn something ... (being able given DOF, to learn to walk for instance) . A function.

Before answearing the OP question, we have to be on the same wave about what intelligence is. And this is a philosophical and still open question ...

• Understand causalities.
  • When I send out this nerve impulse, my head turns.
  • When my head moves, funny signals come from my inner ear. These are always the same for identical moves. Also, the world always moves in the opposite direction.
  • When I drop my teacup, it falls to the ground. When I drop my cell phone, it falls to the ground.
• Combine related and unrelated causalities, and make decisions based on that combined knowledge.
  • Most things I tried, including teacups and cellphones, fell to the ground when I dropped them. I expect a sausage to fall to the ground as well, without having to try.
  • Often, things break or become dirty or otherwise unusable when they fall to the ground.
  • I don't like that happening (also my back hurts from picking them up) and therefore will not voluntarily drop things.
• Perform sanity checks on decisions before and while taking action.
  • The GPS tells me to turn right. Into the river. Screw that cheap piece of junk.
  • The traffic light just switched to "Don't Walk". I don't think that I'm supposed to stop right here in the middle of the street.
• Validate decisions, as their outcome becomes apparent.
  • I really should have dropped that darn cellphone and reached for the escape ladder instead.
  • Giving that tramp a ride was a stupid idea.
• Improve future decisions based on the outcome validation.
  • Last time I got beaten up bad after pushing that guy. This time I'll pretend to give in and kick him in the balls from behind.
  • Next time someone asks you if you're a god, you say "Yes!".
• Do all of the above without someone having to tell you, and without someone telling you how.
• Adjust decision parameters by observing the outcomes of other people's decisions.

It is also possible to skip most of the philosophical aspects of the matter by realizing that compression is very closely related to intelligence. I know it sounds odd, but it is a deep and interesting idea.

It is also possible to skip most of the philosophical aspects of the matter by realizing that compression is very closely related to intelligence. I know it sounds odd, but it is a deep and interesting idea.

Although it may seem that way, I think the idea is somewhat flawed. What compressors do is reduce entropy. If they work well, that is.

Dictionary-based compressors replace parts of the input with things they have stored in their dictionary, if that takes fewer bits, according to a scheme that the programmer has decided. Maybe the programmer is intelligent, but the program doesn't do anything special.

Statistical compressors, on the other hand, determine the likelihood of the next symbol that comes in by what they have seen before. You could much more easily associate this with "intelligence" even if they still only follow a very rigid pattern that the human programmer has devised.

But if you think about it, this is a behavior which is very much akin to a gambler playing roulette, who, upon having seen 17 three times, puts his money on 17. After all, 17 seems to be the lucky number. Would you deem this intelligent? Secret tip: A horse with a name like "Lightning" cannot lose.

Maybe PPMD uses a somewhat more sophisticated algorithm, but in the end it is the exact same thing. Looking into the crystal ball.

The only difference (and the difference that decides on the outcome!) to compressing enwik8 is that the input is different. Characters in enwik8 are not random, but are highly correlated, and there is a huge amount of entropy in that text. This is why compressors are more successful than our roulette gambler. Still, they do more or less the same thing. They use some statistical model and if they are lucky, this allows them to correctly predict the coming symbols.

What would be more impressive would be compression in a sense such as you show an image to a computer and it outputs "fat guy in funny clothes making a sad face, bent over a dead girl, that's Pavarotti as Rigoletto". That is much more like the way a human would "compress" that photo.