Science Forums: Mr Skeptic - Viewing Profile - Science Forums

I have a degree in biology, but I don't know what life is either. Each definition has its flaws, especially if we aren't restricting it to biological Earth life. At the end of the day, it is just a definition we make up and things fit it or they don't. But it is very educational to try. To start off, decide which things you want to define as alive or whether you want to restrict your definition. Earth life, viruses (which can't reproduce without a living cell), alien organic life, alien inorganic life, mechanical sentient robots, which of these are alive and which of these are not, or which of these do you want to ignore for the sake of simplicity?

These concepts might help: life must at some point be able to reproduce (else it would not exist), there needs to be a boundary between the living thing and its environment (for non-virus Earth life, that is a phospholipid bilayer, much the same thing as a soap bubble but more stable), the ability to maintain homeostasis (though again viruses are an exception).

URAIN, on 14 November 2011 - 12:00 PM, said:

Wrong on both counts. Also, how come you were even born, aren't you more organized than a fertilized egg cell and a bunch of food? Once you understand the second law of thermodynamics, you will understand that it is possible for you to have been born.

I think the agreed-upon definition of intelligence was "The ability to solve mental problems." or "The ability to construct a model of a problem and solve it.". The disagreement stems from the fact that there are many kinds of problems with many solutions, and the definition of intelligence does not give any way to measure intelligence. Also, many people are more interested in certain kinds of problems, say concerning abstract concepts rather than things like being able to recognize objects' identities, sizes, and distances given a visual image of them.

For example, shooting a basketball through a hoop requires all sorts of calculations, first recognizing what the hoop is, then measuring the distance to it using size comparisons or trigonometry, then if bouncing off the backboard measuring the angle the board is at, then calculating the velocity and direction to throw the ball at given a uniform downward acceleration of g and perhaps also air resistance and the curvature caused by spin, then finally sending signals to hundreds of different muscles to contract and relax in such a way as to throw the ball at the calculated angle and velocity. If also playing basketball, these calculations need to be done in a fraction of a second, and should also consider the odds of making the shot and the actions of nearby players to potentially block your shot or that you might pass the ball to a teammate with better odds. Yet many people would balk at having this be considered intelligence. Similarly, computers can perform millions of calculations in the time we might do one, but they are limited in that their abilities only apply to very specific problems and are completely unable to solve novel problems (that they have not been programed to solve or learn to solve). So a computer's intelligence beats ours by far when it comes to arithmetic, but they can be as dumb as a rock for things we might find simple.

While we're on the topic of computers, a measureable definition for intelligence might be the number of calculations per second the entity is capable of. This is a good definition for comparing computers, but for neural networks it seems what people are interested in is the number of synapses (the more synapses the more difficult it would be to construct a computer simulation thereof). Of course calculations per second doesn't actually say much about whether all those calculations help solve a problem, so that isn't really a measure of intelligence but perhaps of complexity or calculating power.

Some people also place great value on the ability to learn as an aspect of intelligence. Of course people with amnesia might have a problem with that.

More problems occur when we allow the use of the environment in problem-solving. What about people who use pencil and paper/written notes to aid their mental capabilities, or books, or computers, or asking other people? These things might seem like cheating on a test but in reality are how things are done. But then our intelligence would depend not only on our internal state but on what we have access to, which would make it even harder to understand and measure.

THoR, on 23 November 2011 - 08:17 PM, said:

So if I'm understanding correctly, you propose that there is a machine that can produce work by taking gravitationally condensed clumps of matter and dispersing them through space? This seems rather odd to me; I can easily think of a machine that does the opposite but not one that does what you seem to be proposing. If you don't propose that such a machine can be made, then are we agreed that distributed matter has more useable energy than gravitationally clumped matter, and that gravitational clumping reduces the amount of useable energy?

THoR, on 23 November 2011 - 12:38 AM, said:

I'm not sure how that follows. Also, you and cosmologists might not be talking about the same sort of creation -- certainly at the time of the BB any information from any previous universe would have been largely destroyed due to the environment, so we could say our universe was created then then even if the matter and energy were there forever. On the other hand, I have heard suggestions that the total energy of the universe is zero due to some sort of negative energy from the expansion of space

Quote

That's the First Law of Thermodynamics, and in no way conflicts with the second law.

Quote

Exactly, and in an insulated system the entropy increases -- so says the Second Law of Thermodynamics. Entropy is not destruction of energy, it just means you won't be able to use that energy.

Edit: Just to clarify, an example of increasing entropy would be what happens when you mix 1 gram of water with 300 calories of energy with 1 gram of water with 350 calories of energy -- you get 2 grams of water with a total of 650 calories of energy (325 calories each). See how the entropy increased there?