"A Mathematician's View of Evolution"

[hammegk]

"We" = you and who else? A seed of given genetics, plus a given environment, makes a plant. How specific this plant is, and the extent to which environment can influence the outcome, I do not pretend to know. While perfectly willing to be convinced, I will not accept your second sentence by assertion alone.

I find your assertion strange that since output (a specific tree) differs from any other specific tree, the code to produce it changed. Of course the input changed.

Thank you for illustrating my point. This "code" that you say the machine is part of is circularly inferred from the outcome. I congratulate you on being consistent (honestly); I do not agree that the code (including the machine), independent of the environment, determined the widgets. It is perfectly consistent with your view, but I find it circular, and cannot accept it without independent evidence.

And taking your analogy where it goes, the machine we need to disrupt is at the dna level within our 'machine'; the seed itself.

This 'activation' appears to be independant of any other genetics system. In other words, somehow the environment works directly on a protein (the product of the 'code'), rather then the 'code' itself.

That code may produce output that contains pieces of added, new code that is again dependent in specifics of input is not noteworthy sfaik.

Or are you saying that all possible outcomes, due to different inputs, are already embedded in the programming in such a way that no new information ever appears?

No more so than in the sense that does the complexity exist in a specific mandelbrot set, or in the algorithm and its' parameters that produces it?

We can obviously define complexity mathematically, but what info do we think that numeric result provides other than how long the program has been running and producing output?

 

[Taffer]

That code may produce output that contains pieces of added, new code that is again dependent in specifics of input is not noteworthy sfaik.

No, because, for the purposes of this discussion, the 'code' outputs a protein. That is the sole output of the code (again, for the purposes of this discussion). But the output of the 'code' is directly affected by the environment, which goes back again to change the 'code's' output further. Thus, that result is not encoded for by the original code.

 

[hammegk]

No, because, for the purposes of this discussion, the 'code' outputs a protein. That is the sole output of the code (again, for the purposes of this discussion). But the output of the 'code' is directly affected by the environment, which goes back again to change the 'code's' output further. Thus, that result is not encoded for by the original code.

Strange, it appears to me as though it were since the information coded in a specific protein required more than input to be 'written'. You disagree that code can write added code which is also a function of input?

 

[Taffer]

Strange, it appears to me as though it were since the information coded in a specific protein required more than input to be 'written'. You disagree that code can write added code which is also a function of input?

What? The protein writes no code, it is not code, it simply starts the 'reading' of further code. I don't understand what you're on about, here.

 

[hammegk]

I do not know if a protein writes more code or not. Perhaps you can answer that question.

Using the Mercutio machine analogy, the seed (original code) builds a lathe, a milling machine, a stamping press, etc, and begins manufacture using appropriate inputs as they become available.

One output, in our case, is a protein. And I'd posit the protein is coded and may write added code in its' output as did the original seed programming.

 

[Taffer]

I do not know if a protein writes more code or not. Perhaps you can answer that question.

Using the Mercutio machine analogy, the seed (original code) builds a lathe, a milling machine, a stamping press, etc, and begins manufacture using appropriate inputs as they become available.

One output, in our case, is a protein. And I'd posit the protein is coded and may write added code in its' output as did the original seed programming.

No. The protein writes no new code at all. It appears to be a transcriptional initiation protein, in that it is DNA binding with sequence specificity. Likely, it binds to other flowering domains and initiates transcription of those genes. It does not modify the genetic code in any way.

The point I was making is that this protein is expressed constantly throughout the year, and that an environmental factor is required to cause it to have an effect. Thus, the machine is more complex then the programming.

 

[hammegk]

No. The protein writes no new code at all. It appears to be a transcriptional initiation protein, in that it is DNA binding with sequence specificity. Likely, it binds to other flowering domains and initiates transcription of those genes. It does not modify the genetic code in any way.

Nor did I specify 'new' code is restricted to dna level.

The point I was making is that this protein is expressed constantly throughout the year, and that an environmental factor is required to cause it to have an effect. Thus, the machine is more complex then the programming.

So you assert, unjustifiably imo. It is programming -- whether at dna/rna level or elsewhere in the specifics of the protein molecule -- that forces the correct response to environment. And that programming is a direct function the seed rna/dna and previous, selected, inputs thereto.

 

[Taffer]

Nor did I specify 'new' code is restricted to dna level.

My apologies, I thought you had.

So you assert, unjustifiably imo. It is programming -- whether at dna/rna level or elsewhere in the specifics of the protein molecule -- that forces the correct response to environment. And that programming is a direct function the seed rna/dna and previous, selected, inputs thereto.

Perhaps this is just my opinion, but it seemed to me you were asserting that the genetic code in a seed is just as complex as the plant it forms. Thus, I was giving an example of complexity which arrises from a source other then the genetic code. If I misunderstood your point, then I apologise.

 

[hammegk]

Perhaps this is just my opinion, but it seemed to me you were asserting that the genetic code in a seed is just as complex as the plant it forms.

Here we could discuss the meaning and measure of complexity,although that to me would generate heat rather than light concerning the actual discussion.

Thus, I was giving an example of complexity which arrises from a source other then the genetic code. If I misunderstood your point, then I apologise.

No apology needed in that I certainly am clarifying my thoughts. But the point you raise is this: what 'source'? I do not agree input -- by itself -- changes coding at any level unless pre-existing code is in place to allow that change.

I'm not referring to dna.rna changes at seed level that are caused by destructive activity; cosmic rays, inappropriate chemicals, etc.

 

[Taffer]

Here we could discuss the meaning and measure of complexity,although that to me would generate heat rather than light concerning the actual discussion.

I agree. "Complexity" is, when we think about it, a rather ambiguous term.

No apology needed in that I certainly am clarifying my thoughts. But the point you raise is this: what 'source'? I do not agree input -- by itself -- changes coding at any level unless pre-existing code is in place to allow that change.

I'm not referring to dna.rna changes at seed level that are caused by destructive activity; cosmic rays, inappropriate chemicals, etc.

Ahh, I see your point.

You are saying that the physical code already contained the 'complexity' required to react to various external stimulations, correct? In other words, all possible reactions to external stimuli are already 'encoded' in the seed?

 

[hammegk]

You are saying that the physical code already contained the 'complexity' required to react to various external stimulations, correct? In other words, all possible reactions to external stimuli are already 'encoded' in the seed?

Yes. In our machine shop analogy, each protein represents a lathe, or milling machine, etc. which presumably contains an instruction set of some sort directly linked back to seed dna via mRna.

Can we -- ala Mercutio -- discount out-of-hand this concept, requiring some 'proof'?

 

[Taffer]

Yes. In our machine shop analogy, each protein represents a lathe, or milling machine, etc. which presumably contains an instruction set of some sort directly linked back to seed dna via mRna.

Can we -- ala Mercutio -- discount out-of-hand this concept, requiring some 'proof'?

I'm not quite sure what you mean by the products of the code (ala various machines) being "linked back to seed DNA via mRNA". Can you clarify?

 

[hammegk]

If my use of "dna via mRna" is incorrect terminology -- and it well may be -- let's try "protein formation process".

Do you object to the specific protein = specific machine analogy?

 

[joobz]

If my use of "dna via mRna" is incorrect terminology -- and it well may be -- let's try "protein formation process".

Do you object to the specific protein = specific machine analogy?

If each machine performs multiple functions under various conditions, then yes I agree. I would have to mostly agree with the code arguements you have made.

But the code contained also encode the walls of the machine shop, the operators, the method of wiring the machines use to function...

I think the coding example isn't wrong. We just need to remember that there are multiple subroutines running simultaneously with each having feed back loops that affect the outcomes. It's more of a process control code.

We still don't know it completely. Think of the classic example of a catepillar and butterfly. They have the idential genetic dna what changes is the relative expresion of that code (a field called proteomics).

 

[Taffer]

If my use of "dna via mRna" is incorrect terminology -- and it well may be -- let's try "protein formation process".

Do you object to the specific protein = specific machine analogy?

No, I don't object. That analogy is fine. I was just a bit confused by what you meant by the whole thing. Not saying you're wrong, nor arguing against you. I simply don't get you.

DNA via mRNA is ... sort of correct. mRNA can be thought of as the 'instructions' by which a protein is 'constructed' (by a ribosome and tRNA).

 

[Dymanic]

Hammy, not all information is coded in the genetics of a seed.

There it is. It's an event-driven system. Think of the DNA as a library of functions, with the ordering and timing of calls to those functions being driven by events at the level of organism/environment as those events propagate downward to the level of the individual cell.

 

[hammegk]

... It's an event-driven system.

I'd say most code of any complexity (damn that word ) is event-driven.

Think of the DNA as a library of functions, with the ordering and timing of calls to those functions being driven by events at the level of organism/environment as those events propagate downward to the level of the individual cell.

Sounds correct to me, with the comment that potential useful responses to events at level of organism/environment are pre-coded, basic source being seed dna. Obviously some events may exceed coding parameters; what is disaster for $50, Alex?

Moving onwards, would someone (hi, Merc) actually propose that, say, a neural network "learns" based on anything other than underlying coding reponding to internal & external events?

 

[Paul C. Anagnostopoulos]

That strikes me as wildly improbable.

Which? That computer programs can evolve or that creationists would ever shut up?

~~ Paul

 

[Paul C. Anagnostopoulos]

Yes. In our machine shop analogy, each protein represents a lathe, or milling machine, etc. which presumably contains an instruction set of some sort directly linked back to seed dna via mRna.

Uh, no. All possible reactions to cosmic rays, chemical mutagens, excessive heat, climate change, etc., are not coded in DNA.

Moving onwards, would someone (hi, Merc) actually propose that, say, a neural network "learns" based on anything other than underlying coding reponding to internal & external events?

If I'm allowed to tweak a connection arbitrarily, then yes, I would say that.

~~ Paul

 

[Dymanic]

Sounds correct to me, with the comment that potential useful responses to events at level of organism/environment are pre-coded, basic source being seed dna.

No. A 'response' is more than what can be contained in the 'code', and looking at it that way is what leads to the bottomless recursion.

The chemical state of the cell is as much a part of the response as is anything at the level of 'code'. A cell inherits not only its DNA from its parent cell, but its chemistry as well. It is this chemistry (as much as anything) which determines which portions of the DNA will be translated into proteins, and when, and what will happen to those proteins.