Intelligent Evolution?

[cyborg]

The premise of the OP was to convince a theoretical future person of the validity of evolution through mechanical development, a non-starter.

In your opinion.

You've been elaborating on incorrect usages of "evolution," generalizing terms into uselessness.

Nope.

 

[ImaginalDisc]

In your opinion.

Not quite. Given the constraints of Evolution, and the lack of similar contrainsts in mechanical design, there's no similarity to be found.

Nope.

You most certainly have. You've turned "evolution" into "change over time by pretty much any means."

 

[cyborg]

Not quite. Given the constraints of Evolution, and the lack of similar contrainsts in mechanical design, there's no similarity to be found.

Yes. I get this. You don't understand abstraction. You can't appreciate the similarities - hell, because you don't understand abstraction you don't even understand how formulating an opinion about a simularity works. You don't understand the computational mathematics. You refuse to acknowledge the realities of engineering.

You most certainly have. You've turned "evolution" into "change over time by pretty much any means."

ID, remember when I told you the term evolution existed before it was applied to biology? Remember how Charles Darwin never used the term? Ready to look foolish and start flustering about equivocation? Okay?

1. any process of formation or growth; development: the evolution of a language; the evolution of the airplane.
2. a product of such development; something evolved: The exploration of space is the evolution of decades of research.
3. Biology. change in the gene pool of a population from generation to generation by such processes as mutation, natural selection, and genetic drift.
4. a process of gradual, peaceful, progressive change or development, as in social or economic structure or institutions.
5. a motion incomplete in itself, but combining with coordinated motions to produce a single action, as in a machine.
6. a pattern formed by or as if by a series of movements: the evolutions of a figure skater.
7. an evolving or giving off of gas, heat, etc.
8. Mathematics. the extraction of a root from a quantity. Compare involution (def. 8).
9. a movement or one of a series of movements of troops, ships, etc., as for disposition in order of battle or in line on parade.
10. any similar movement, esp. in close order drill.

So uh, like I said: YOU HAVEN'T HAD A CLUE WHAT THE DISCUSSION HAS BEEN ABOUT FROM THE BEGINNING. Not only is my usage valid it is the number 1 definition according to dictionary.com.

Now stop being stupid.

 

[ImaginalDisc]

Yes. I get this. You don't understand abstraction. You can't appreciate the similarities - hell, because you don't understand abstraction you don't even understand how formulating an opinion about a simularity works. You don't understand the computational mathematics. You refuse to acknowledge the realities of engineering.



ID, remember when I told you the term evolution existed before it was applied to biology? Remember how Charles Darwin never used the term? Ready to look foolish and start flustering about equivocation? Okay?



So uh, like I said: YOU HAVEN'T HAD A CLUE WHAT THE DISCUSSION HAS BEEN ABOUT FROM THE BEGINNING. Not only is my usage valid it is the number 1 definition according to dictionary.com.

Now stop being stupid.

Congratulations on admitting to using the most vauge and useless definition of a word, directly contrary to the purposes of the OP. Now, can you stop prattling on about your groundless analogy?

 

[quixotecoyote]

Yes. I get this. You don't understand abstraction. You can't appreciate the similarities - hell, because you don't understand abstraction you don't even understand how formulating an opinion about a simularity works. You don't understand the computational mathematics. You refuse to acknowledge the realities of engineering.



ID, remember when I told you the term evolution existed before it was applied to biology? Remember how Charles Darwin never used the term? Ready to look foolish and start flustering about equivocation? Okay?



So uh, like I said: YOU HAVEN'T HAD A CLUE WHAT THE DISCUSSION HAS BEEN ABOUT FROM THE BEGINNING. Not only is my usage valid it is the number 1 definition according to dictionary.com.

Now stop being stupid.

But those are the same semantic games the creationists play, setting up versions of evolution having nothing to do with biological evolution and then trying to tie them together.

No doubt you found a definition in the dictionary that you like. But if you look back in the beginning, this was about the legitimacy of an analogy to explain biological evolution. So if your argument is that you aren't talking about biological evolution, then

YOU HAVEN'T HAD A CLUE WHAT THE DISCUSSION HAS BEEN ABOUT FROM THE BEGINNING.

Which, considering you started the thread, is fairly impressive in its own way.

 

[cyborg]

Congratulations on admitting to using the most vauge and useless definition of a word, directly contrary to the purposes of the OP.

Yet again you do not understand.

Class - Instance.

Now, can you stop prattling on about your groundless analogy?

You do not understand the grounding.

Mathematics.

But those are the same semantic games the creationists play, setting up versions of evolution having nothing to do with biological evolution and then trying to tie them together.

No.

But if you look back in the beginning, this was about the legitimacy of an analogy to explain biological evolution.

Which has devolved into an argument about the validity of terminology. I didn't want to have to pull out a dictionary definition but I am getting ****ing tired of this nonsense.

I wish you would respond to the computational aspects I outlined instead but because you won't I have no ****ing choice do I?

 

[quixotecoyote]

Yet again you do not understand.

Class - Instance.



You do not understand the grounding.

Mathematics.



No.



Which has devolved into an argument about the validity of terminology. I didn't want to have to pull out a dictionary definition but I am getting ****ing tired of this nonsense.

I wish you would respond to the computational aspects I outlined instead but because you won't I have no ****ing choice do I?

If you are basing your other aspects on an invalid definition, there's not much point to arguing over the other aspects.

 

[ImaginalDisc]

Yet again you do not understand.

Class - Instance.



You do not understand the grounding.

Mathematics.



No.



Which has devolved into an argument about the validity of terminology. I didn't want to have to pull out a dictionary definition but I am getting ****ing tired of this nonsense.

I wish you would respond to the computational aspects I outlined instead but because you won't I have no ****ing choice do I?

You could choose to use the correct, scientific, definition rather than whichever vauge and useless definition you choose. All your analogies fail because none of them adhere to the constraints the Theory of Evolution.

 

[cyborg]

If you are basing your other aspects on an invalid definition, there's not much point to arguing over the other aspects.

Class - Instance.

I understand the biological instance. Will you come with me and step back a bit or not and consider the next class up?

All your analogies fail because none of them adhere to the constraints the Theory of Evolution.

IT'S CLASS ****ING INSTANCE DISTINCTION ID. You are denying analogies on the basis that they are different instances NOT because they have different properties.

Biology is biology is it? Thanks for that tautology.

 

[ImaginalDisc]

Class - Instance.

I understand the biological instance. Will you come with me and step back a bit or not and consider the next class up?



IT'S CLASS ****ING INSTANCE DISTINCTION ID. You are denying analogies on the basis that they are different instances NOT because they have different properties.

Biology is biology is it? Thanks for that tautology.

You might remember that I specifically pointed out that imperfectly self-replicating machines would certainly Evolve, but machines are not made that way, and do not owe their origin to Evolution (except, of course, that their desginers evolved.)

The annealing process of metal is an excellent analogy for natural selection, but not for Evolution. There are numerous good analogies for various elements of Evolution, and many theoretical analogies for Evolution using machines. Sadly, none of them are yours.

 

[quixotecoyote]

Class - Instance.

I understand the biological instance. Will you come with me and step back a bit or not and consider the next class up?

They whole point of an analogy is to understand a specific unfamiliar concept through a relation to a familiar concept. Generalizing the concepts into sameness defeats the point.

 

[mijopaalmc]

You know, I find it interesting that the people who claim to be the most outspoken skeptics on this forum are also the same people who insist on playing the most semantic games.

It is true that "evolution" does mean "any process of formation or growth; development", but in the context of biology it means "change in the gene pool of a population from generation to generation by such processes as mutation, natural selection, and genetic drift". The insistence on using nontechnical definitions in technical situations seems to be the main problem with the analogy of technological development with biological evolution. In fact, it was the source of the problem when we were discussing the relationship between evolution and randomness; those who insisted that evolution was non-random insisted on using every definition of "random" except the mathematical one, which is the only definition that those who argued that evolution is random were. So we have the same here as we did there: we are essentially talking circles around one another because we can't agree on a definition of "evolution" to use.

 

[ImaginalDisc]

You know, I find it interesting that the people who claim to be the most outspoken skeptics on this forum are also the same people who insist on playing the most semantic games.

It is true that "evolution" does mean "any process of formation or growth; development", but in the context of biology it means "change in the gene pool of a population from generation to generation by such processes as mutation, natural selection, and genetic drift". The insistence on using nontechnical definitions in technical situations seems to be the main problem with the analogy of technological development with biological evolution. In fact, it was the source of the problem when we were discussing the relationship between evolution and randomness; those who insisted that evolution was non-random insisted on using every definition of "random" except the mathematical one, which is the only definition that those who argued that evolution is random were. So we have the same here as we did there: we are essentially talking circles around one another because we can't agree on a definition of "evolution" to use.

It's not about agreement. In the context of the of The Theory of Evolution, it means only one thing. All other definition are, in that context, completely wrong.

 

[Southwind17]

You know, I find it interesting that the people who claim to be the most outspoken skeptics on this forum are also the same people who insist on playing the most semantic games.

Correct me if I'm wrong (and I might well be), but isn't that exactly what you were doing here?:

Who you say that a coffee filter "chooses" not to let the coffee ground through?

Would you say that density gradient centrifugation "chooses" to let the densest particles collect at the bottom of the test tube?

My interpretation of what you were suggesting by asking those questions is that 'choosing', in the context of intelligent design development, is completely different from 'selecting', in the context of natural selection. If that's correct then I see 'choose' as a perfect similie for 'select'. Both infer two possible outcomes; survival or extinction. In the case of natural selection the environment 'chooses' or 'selects' through survival of the fittest which reproductions subsist and which don't. In the case of mechanics the environment equally 'chooses' or 'selects', again, through survival of the fittest which developments subsist and which don't. And please don't repeat the argument: No, the designer 'decides' which designs to retain and which to dispense with. Yes, he does, but his decision is wholly derived from the environmental pressures under which the machine will be required to perform. If it isn't the 'best' in its class, then it's unlikely to survive. Competition and market pressures will see to that. In principle natural selection and market competition are no different as regards the drivers that determine whether one particular design will prevail over another.

 

[ImaginalDisc]

Correct me if I'm wrong (and I might well be), but isn't that exactly what you were doing here?:



My interpretation of what you were suggesting by asking those questions is that 'choosing', in the context of intelligent design development, is completely different from 'selecting', in the context of natural selection. If that's correct then I see 'choose' as a perfect similie for 'select'. Both infer two possible outcomes; survival or extinction. In the case of natural selection the environment 'chooses' or 'selects' through survival of the fittest which reproductions subsist and which don't. In the case of mechanics the environment equally 'chooses' or 'selects', again, through survival of the fittest which developments subsist and which don't. And please don't repeat the argument: No, the designer 'decides' which designs to retain and which to dispense with. Yes, he does, but his decision is wholly derived from the environmental pressures under which the machine will be required to perform. If it isn't the 'best' in its class, then it's unlikely to survive. Competition and market pressures will see to that. In principle natural selection and market competition are no different as regards the drivers that determine whether one particular design will prevail over another.

Firstly, there is no such thing as "a perfectly simile" save for an identity. You're overemphasizing, likely because of how tenuous the simile really is.

Secondly, "choose" necessarily implies an entity capable of making a choice. Evolution does not require any such intelligent actor.

Thirdly, the "best" organism is the one which happens to survive and reproduce, a matter that could easily be the result of serendipity.

A disaster could wipe aware stronger, faster organisms based purely on the geography of an event, making their 'plans' lost forever. ('Plans' go in scare quotes because DNA is hardly a plan.) Even if a machine is lost, the plans can easily remain. Machines are divorced from the information needed to make them; organisms are not.

 

[Southwind17]

This is whre I disagree with you.

Assume that one starts with two copies of the same design.

One performes an evolutionary algorithm, (not evolution) and imperfectly copies the prototype many times, removing those which fail.

Another performs classcal engineering.

Both initial prototypes fail. The classical approach would be to fix just the problem, the evolutionary algorithm is to randomly alter the design parameters and select the prototype with the best fitness score for the next iteration.

Eventually both approaches produce successful designs.

The product of the classical engineering approach will look far more similar to the initial design than the evolutionary algorithm because the only changes were to the parts causing the problem.

Yes it is possible that a succession of random changes would have produced a similar product, but (unlike discussions of ID, the odds really are agianst this). There is a strong probability that the evolutionary approach would produce a better result (in a lot more iterations). But it would not be so similar to the "parent". One would expect to see a multitude of differences between "generations", some beneficial, and others neutral. With a classical approach the alterations would be with the express intent to solve specific problems.

I follow and understand your logic here, but nonetheless, what's wrong with considering the 'evolutionary approach', as you put it. What's the big deal about the end result not being so similar to the parent? We're talking about a very large number of iterations here. How similar would you say we are, as humans, to our distant 'parents'? I maintain that mechanical design could follow an evolutionary approach if we allowed it to, and we'd end up, as you rightly say, with something different from what classical design might produce, but it could still be a machine that appears to be irreducibly complex.

 

[ImaginalDisc]

I follow and understand your logic here, but nonetheless, what's wrong with considering the 'evolutionary approach', as you put it. What's the big deal about the end result not being so similar to the parent? We're talking about a very large number of iterations here. How similar would you say we are, as humans, to our distant 'parents'?

Morphologically? That's incredibly difficult to quantify. Genetically, even Neanderthals are 99.5% similar to us, and they might not even be our ancestors.

Mind you, that's largely because the majority of DNA is self-replicating, self-regulating material, or codes for intricate microbological cellular traits that you'd never notice.

maintain that mechanical design could follow an evolutionary approach if we allowed it to, and we'd end up, as you rightly say, with something different from what classical design might produce, but it could still be a machine that appears to be irreducibly complex.

"Irreducibly complex" isn't a valid term in either engineering or biology. Why introduce Intelligent Design nonsense?

 

[jimbob]

Of course, if someone discovered a mouse with a human humoral immune system, for example, then I would argue that this was evidence of some intelligent agency altering the mouse, because that couldn't happen by chance.

Something like that would not be possible with evolution alone.

Oh look

(sign up for nature e-alerts, most of the stuff is behind a paywall, but a lot is interesting)

The evoltionary approach is very good, look at crop yields, or dog breeds, these are altered for specific goals, not just their survival.

Evolution is powerful too, but subtly different from artificial evolutionary approaches. (Of course with selective breeding the breeds are also subject to natural selection as well as artificial selection).

The classical engineering approach is completely different, as you can make investigitive experiments that would fail in total, yet still provide useful information. You can also learn from mistakes, and make deliberate changes, backed up by calculations. Even making random changes only on a failure site is not analogous to evolution.

The OP doesn't work to deal with "irriducible complexity" because the argument is that these mutations changes couldn't have happended by chance so needed an intelligent agency to instigate them. This is exactly what happens in the engineering examples that you list.

Your idea is an exact analogue of intelligent design as it involves intelligent designers. I am not aware of many "perfect creationists", which your point provides counterexamples for.

Cyborg.

What are the similarities.

Both processes are iterative, and thus (by definition) there are changes between "generations" one towards a "better" system, as defined by intellignece, and the other towards replicators that would have more replicating offspring.

Evolution requires that the template is carried by the replicator, as that is the only way that evolution can act on the template.

 

[jimbob]

Cyborg, what have I posted that shows that I (or ImaginalDisc, mijopalmic or quixotecoyote)

"refuse to acknowledge the realities of engineering"?

Evolutionary algorithms are very powerful problem solvers; they are partial analogues of evolution, but with subtle and important differences that seem to mislead many people.

 

[jimbob]

Of course we have had this discussion somewhere before:

If you can see why 'artificial' selection by some agent which might be said to be 'purposefully' moving a population towards some allele frequency then it should be clear that one can make just as clear an argument that it is just as 'purposeful' for other such occurrences.
<snip>
Either that or you should be happy to think of other similar examples in 'nature' as being 'artificial' selection.

In case you are wondering, I was arguing that natural selection is "probabilistic" and not "purposfull".

I would go further and say that as the environment contains other organisms, also being selected probabilistically, and in themselves altering the selection criteria for other orgainsms, the "direction" of evolution is also probabilistic.

This thread is a good example as to why I feel this is important.