Intelligent Evolution?

[cyborg]

You haven't considered a single thing presented here - you are simply content to be dogmatic.

YOU cannot quantify the differences other than by assertion. I, however, can, AND HAVE, by not refusing to abstract the concept of design to include everything.

 

[ImaginalDisc]

You haven't considered a single thing presented here - you are simply content to be dogmatic.

I have read and considered each point you have made. What frustrates you, I venture to guess, is that I do not agree with any of them.

YOU cannot quantify the differences other than by assertion.

Design can overhaul, evolution cannot.

Design can plan for long-term development, evolution cannot.

Design can lift elements from one type of thing and apply them to another, evolution cannot.

Design can retain the plans of a form indefinately, evolution cannot.

Living things are produced by autonumous reproduction, machines are not.

Living things have heritable traits, machines do not.

Living things mutate, and those mutations are passed on. Machines neither mutate, nor pass on mutations.

I could go on for some time. At each point, you have dismissed these differences as irrelevant, or generalized the terms into useless vaugeness, whereas the peculiar contraints of living things, the struggle to survive, and the pressures of impersonal, unguided selection, are the very essence of Evolution. Without these elements, living things would not exist, whereas machines and design do not need any of these traits.

I, however, can, AND HAVE, by not refusing to abstract the concept of design to include everything.

Unfortunately living things and the proccess of Evolution are not at all absract. Living things are not patterns and ideas scribbled onto blueprints and patterns to be fabricated by intelligent actors with an understanding of the intended results; living things are breathing, eating, killing entities continuously reproducing themselves and giving rise to new forms entirely without any abstraction. The Theory of Evolution is a description of the origin and properties of the very visceral living things in the natural world and it is, to date, the most compelling, most accurate, and most parsimonious explanation for these diverse and contradictory organisms.

As machines lack all the elements Evolution seeks to explain, I fail to see any reason for using machines an analogy to teach it.

 

[cyborg]

Design can overhaul, evolution cannot.

No - it is merely a matter of distance to another point in the design landscape. There is NOTHING that prevents a purely random change of design achieving this mathematically.

Show otherwise.

Design can plan for long-term development, evolution cannot.

Elaborate.

Design can lift elements from one type of thing and apply them to another, evolution cannot.

Really?

http://www.sciencenews.org/articles/20070901/fob1.asp

There is no physical mechanism preventing such a thing it is therefore possible for it to evolve: and it has. It's probably responsible for the evolution of multicellular organisms in the first place.

Design can retain the plans of a form indefinately, evolution cannot.

Elaborate.

Living things are produced by autonumous reproduction, machines are not.

UGH. AGAIN, IT IS IRRELEVANT. Autonomous replication doesn't confer any ability to replicate anything non-autonomous replication cannot achieve.

And I note that you did not say, "design cannot produce machines with autonomous replication," because that would obviously invalidate your point.

Living things have heritable traits, machines do not.

Again, quite, quite irrelevant. Where the specifications for the traits are does not confer any special abilities at all.

And again I note you speak not of design: you merely assert something about living things.

Living things mutate, and those mutations are passed on. Machines neither mutate, nor pass on mutations.

Living things DO NOT mutate - unless you are going to call DNA a living thing. articullett has already talked about this fallacy earlier.

INDIVIDUALS DO NOT EVOLVE.

And what, I ask, is to prevent a machine from achieving such a thing by design? Would you like to insert such a design is impossible? No? I wonder why not...

I could go on for some time.

I'm sure you could. But you would be missing the point at each and every stage.

At each point, you have dismissed these differences as irrelevant,

Because they are for the abstraction. You clearly do not understand what an abstraction is.

whereas the peculiar contraints of living things,

...which are irrelevant when considering the wider abstraction. (Something you are having a singular inability to grasp).

the struggle to survive,

Things struggle to survive do they? That sounds dangerously anthropomorphic. Better check yourself otherwise you'll give ammo to the IDers

and the pressures of impersonal, unguided selection are the very essence of Evolution.

Guided/unguided is irrelevant - unless you would care to explain to me exactly how an impersonal process could comprehend such a difference?

Without these elements, living things would not exist,

No - without appreciating how evolution works one cannot understand how living things can exist without appeals to 'design has a designer' - the fallacy you are continually engaging in. One simply goes around saying, "but without a priori knowledge of how the universe works nothing could be made!"

whereas machines do not need any of these traits.

Which is, again, irrelevant.

Unfortunately living things and the proccess of Evolution are not at all absract.

I didn't realise Evolution was a concept written in stone. Did you get it from a mountain after conversing with a deity?

Here was me thinking it was a human invention used to describe the world. Now I know it simply 'exists'.

Again I do not think you know what an abstraction is otherwise you would not say such silly things.

Living things are not patterns and ideas scribbled onto blueprints and patterns to be fabricated by intelligent actors with an understanding of the intended results; living things are breathing, eating, killing entities continuously reproducing themselves and giving rise to new forms entirely without any abstraction.

You do not get it.

The Theory of Evolution is a description

Earlier you said it wasn't abstract. Descriptions are abstract. Make up your mind.

As machines lack all the elements Evolution seeks to explain, I fail to see any reason for using machines an analogy to teach it.

1) I didn't say anything about teaching it.
2) Until you understand how machines are designed you can't understand why the Intelligent Design movement is so fundamentally wrong.
3) There are more fundamental concepts at play here. Your refusal to discuss them is irrational.

 

[ImaginalDisc]

I have previously addressed your points. I see no reason to go over old ground.

 

[cyborg]

Well, let's just take one shall we?

Show me how it is mathematically impossible for a change in a genome to allow a transition from one arbitrary place in the design landscape to another.

 

[ImaginalDisc]

Well, let's just take one shall we?

Show me how it is mathematically impossible for a change in a genome to allow a transition from one arbitrary place in the design landscape to another.

You're equivocating the word "design." You used it in one place to mean, "intention of intelligent actors," and now you're using it to simply mean "form."

See where inconsistency gets you?

 

[cyborg]

You're equivocating the word "design."

Noun. Verb.

Learn 'em, love 'em.

 

[jimbob]

Would the mamalian eye have been designed?

Even an incompetent designer would put the blood supply behind the light-sensitive cells.

The results of evolution and design often differ

 

[ImaginalDisc]

Noun. Verb.

Learn 'em, love 'em.

You're only, deliberately, confusing the issue. No one is aruging that a small genetic change can't have a large change in the form and function of an organism. However, the form and function of organisms tend towards small changes (neonatal trait retention and atavism excepted) rather than large ones, lead to dead-ends, and do not require intelligent actors, which design does.

If you're going to persist in using "design" arbitrarily, don't.

 

[jimbob]

The process are fundamentally different, the results differ, and creationists like the analogy; anything else in its favour?

 

[cyborg]

Even an incompetent designer would put the blood supply behind the light-sensitive cells.

Again I wonder if you really have an experience with the world of design. There's a whole butt-load of half-working **** out there.

And interestingly enough it doesn't always get fixed to the optimal design.

I wonder why... maybe Mr Betamax and Mr VHS can give us another lesson in real life human design that isn't based on some idealistic nonsense about how people think intelligent design works?

You're only, deliberately, confusing the issue.

No. That's your irrational reaction to the words again. I blame ID and ID.

No one is aruging that a small genetic change can't have a large change in the form and function of an organism.

You've missed what I was saying again. This will continue to happen whilst you restrict yourself to only considering the restricted set of possibilities afforded Earth-bound biological evolutionary systems and NOT considering the consequences of varying the parameters of the evolutionary algorithm in play including extending the concept beyond that of simply the biological domain.

Can you or can you not hypothesise about alternative worlds?

And before you point it out, yes, we only live in one world. Biology has only happened in one way on this planet. The Theory Of Evolution specifically relates to, and was developed by observing, biology. I know this. Stop wasting your time trying to persuade me to accept things I already accept.

What you nor the IDers seem to comprehend is that the specific variables DO matter in defining the overall shape of what we see. This is why, as I said earlier, it IS a continuum. This is the consequence of the mathematics. Pretending it is not so simply makes you wrong.

If you're going to persist in using "design" arbitrarily, don't.

My usage of the term is not arbitrary. You simply don't like it.

The process are fundamentally different, the results differ, and creationists like the analogy; anything else in its favour?

You refuse to understand. Again I ask:

Function X outputs A into function Y.

Function Y outputs B into function X.

Function X is an evolutionary algorithm.

Function Y is arbitrary.

When will I better be able to predict the results of the algorithm based on the properties of function Y? When it is simpler or more complex? When it is a specified and definite goal or when it is a broad simulation of some real-life biological system? How do you stop me from making it progressively more and more complex? What do you think will happen as the complexity of the system tends to infinity?

 

[jimbob]

The "design-fault" in the mamalian eye is obvious.

I'd also have designed the human birth method differently.

I am not saying that humans can't make poor designs, but do you disagree that these types of instance are "signatures" of evolution. Or evolutionary algorithms.

An analysis of such a design as implimented, would spot the weakness, and correct it.

Why bother with an appendix? Yes there might be some activities that it performs, which are useful, but they could be performed better without risk of rupture.

You refuse to understand. Again I ask:

Function X outputs A into function Y.

Function Y outputs B into function X.

Function X is an evolutionary algorithm.

Function Y is arbitrary.

I am not refusing to understand what you say, but I do not understand what you are getting at, or indeed what relevance this has to "intelligent design of engineered systems".

 

[jimbob]

Evolutionary algorithms are not evolution. Evolution requires that imperfect copies are made, these would replicate, and compete for resourses so would be subject to environmental selection pressures.

The copier has to be th ecopy and has to be what the selection pressures act on otherwise it is not evolution, but an analog

 

[cyborg]

The "design-fault" in the mamalian eye is obvious.

I see "obvious" design faults in human designs all the time.

I'd also have designed the human birth method differently.

Design and implementation are worlds apart - as are the consequences, costs and effectiveness of any particular design.

I am not saying that humans can't make poor designs, but do you disagree that these types of instance are "signatures" of evolution. Or evolutionary algorithms.

Yes. It is entirely dependent on the particular variations in the system. There is nothing which cannot be costed for if you can quantify it.

For example, as I mentioned earlier, evolutionary algorithms tend to construct functions where variables are entangled so that extracting and varying design parameters is hard. What is stopping me from making a selective pressure that selects against designs with such entanglements?

I could avoid the 'poor' eye design by forcing a solution where 'good' eye design is given a beneficial cost.

In nature the cost of the 'poor' design is not enough to force a solution. You would have to agree though that if it were we would see them, yes?

An analysis of such a design as implimented, would spot the weakness, and correct it.

The point you do not get though, and I am trying to get across, is that evolutionary algorithms are fully general enough such that there is no algorithm that produces results which cannot be replicated by it.

The only advantage you can get is time. That's it. Of course time may be very important - it sure is if we want results for particularly tricky problems before we die.

Dumb can do whatever smart can do in all cases.

I am not refusing to understand what you say, but I do not understand what you are getting at, or indeed what relevance this has to "intelligent design of engineered systems".

If you understand the computational aspects you understand why an evolutionary algorithm represents a general problem solving system that will reach a solution in NP time.

The only difference attained by using other algorithms is the time complexity class - P. (And the space complexity too - i.e. we don't need as much 'memory'). It cannot solve any problems an evolutionary algorithm cannot solve. It cannot produce any solutions an evolutionary algorithm cannot produce.

This is it in a nut shell. It's why I find the hard distinction arbitrary. It is a continuum.

 

[cyborg]

The copier has to be th ecopy and has to be what the selection pressures act on otherwise it is not evolution, but an analog

*Sigh* Then make up a new name, call it that, go away happy that biology has it's own little impenetrable world impervious to more abstract reasoning where the term Evolution is exclusively its.

But then you are left only being able to argue against ID by saying, "You can't talk about that here." That looks dogmatic.

 

[mijopaalmc]

cyborg-

The problem with your analogy has always been that when there is a mistake or flaw in an engineered design, an engineer (not necessarily the same one) will go back a correct the design, thereby removing the mistake or flaw from the design. This simply does not happen in evolution. A mutant allele is removed from or fixed in the population at large by the differences in its reproduction relative to the wild type. Very rarely, if ever, does the allele mutate back to the wild type and even then the reverse mutation does not happen because the mutant allele is disadvantageous; it happens without purpose of reason, just like the the mutation to the mutant allele happened.

 

[quixotecoyote]

*Sigh* Then make up a new name, call it that, go away happy that biology has it's own little impenetrable world impervious to more abstract reasoning where the term Evolution is exclusively its.

But then you are left only being able to argue against ID by saying, "You can't talk about that here." That looks dogmatic.

I agree. I get so upset when people walk into my office and try to insist that my communication research should only deal with communication between people or groups thereof. Obviously comets can communicate information between solar systems, so my research on the moments of comets should be published in the Communication journals.

It just pisses me off that they can't make up a new name for what they want communication to be and go away happy with their own little impenetrable world impervious to my more abstract reasoning.

Let my comets go!

 

[cyborg]

You're not paying attention mijo.

PLEASE PEOPLE: does anyone GET the difference between talking about what actually happens in the biological world, the concept of evolution as applied to biology and the more general mathematical abstraction and the consequences thereof of that abstraction?

Anyone?

Let my comets go!

Cute quixotecoyote but this just belies the underlying problem here - you people all want to talk in the highest level chunks possible. I am reducing the problem before restructuring it.

There seems to be a real lack of grasping the class/instance distinction here.

And since there *are* fundamental properties as far as the transmission of information from one bit of reality goes to another maybe it is worthwhile applying them to the world at large? I dunno, maybe I'll call it "physics" and from there on out I'll start chunking bits of it out to explore instances of the general class.

How does that sound? Can we stop building strawmen now?

 

[quixotecoyote]

You're not paying attention mijo.

PLEASE PEOPLE: does anyone GET the difference between talking about what actually happens in the biological world, the concept of evolution as applied to biology and the more general mathematical abstraction and the consequences thereof of that abstraction?

Anyone?



Cute quixotecoyote but this just belies the underlying problem here - you people all want to talk in the highest level chunks possible. I am reducing the problem before restructuring it.

There seems to be a real lack of grasping the class/instance distinction here.

And since there *are* fundamental properties as far as the transmission of information from one bit of reality goes to another maybe it is worthwhile applying them to the world at large? I dunno, maybe I'll call it "physics" and from there on out I'll start chunking bits of it out to explore instances of the general class.

How does that sound? Can we stop building strawmen now?

Your abstractions are misleading. Sure, you can talk about evolution in terms of mechanics, but it's misleading and you would be better off discussing it in terms of biology. This is true just as I can talk about comets in terms of communication, but it's misleading and I would be better off discussing it in terms of astrophysics.

 

[cyborg]

Your abstractions are misleading.

No; they are not.

Sure, you can talk about evolution in terms of mechanics, but it's misleading and you would be better off discussing it in terms of biology.

*Ugh.* Nobody gets it.

This is true just as I can talk about comets in terms of communication, but it's misleading and I would be better off discussing it in terms of astrophysics.

In a SPECIFIC instance.

class - instance

THIS IS WHAT ABSTRACTIONS ARE ALL ABOUT! You fix an abstraction to an instance!