Happy Birthday, Mr. Darwin: Growing Majority of Americans Support Teaching Both Sides of Evolution Debate

[rossum]

It does appear that the disagreement lies in the definition of information with yours being a strictly mathematical construct measuring the physical content of a message.

Yes. So far PEPCIS has been unable to produce any specifics of what he sees as information.

I assume this would explain why the Shannon value differs from the Kolmogorov value.

Yes. They are calculated by different formulae and are useful in different situations. Another type of information is Fisher information, as seen in Natural selection maximizes Fisher information, Frank, 2008. As the abstract says:In biology, information flows from the environment to the genome by the process of natural selection. However, it has not been clear precisely what sort of information metric properly describes natural selection. Here, I show that Fisher information arises as the intrinsic metric of natural selection and evolutionary dynamics. Maximizing the amount of Fisher information about the environment captured by the population leads to Fisher’s fundamental theorem of natural selection, the most profound statement about how natural selection influences evolutionary dynamics. I also show a relation between Fisher information and Shannon information (entropy) that may help to unify the correspondence between information and dynamics. Finally, I discuss possible connections between the fundamental role of Fisher information in statistics, biology and other fields of science.
Real scientists are doing real work using real measures of information, Shannon and Fisher information in the paper cited.

it seems you have shifted the real battle from one term to another, from information (which you say is measurable) to meaning (which you say is not).

It is evident that PEPCIS has no real definition of information beyond something like “I know it when I see it”. That sort of subjective definition often gets mixed up with meaning. Meaning is obviously subjective, whereas mathematical information metrics are not. Since the discussion on mathematical information was not going anywhere useful, I moved on to discussing meaning in the hope that the change of topic might create a more useful discussion.

Without stepping too deeply into this, information is useless unless it has meaning

Without a measure for meaning this is not a great deal of use. Some DNA is translated into protein, other DNA is not. Different proteins have many different functions.

and we know that DNA sequences have meaning since the information is not random and specific reactions occur as a result of its coding.

Parts of DNA are effectively random because they do not code for anything; for example introns are cut out of RNA before it is passed to a ribosome for protein assembly. DNA in introns is not coded into any protein.

Is the difference then between you and PEPCIS over the meaning contained within DNA rather than its information content?

Possibly, but as the Frank paper I referenced above shows, it is possible to handle the subject without the need for “meaning” as oposed to just information.

Did you mean to imply that meaning depends solely on the observer?

In this context, yes. An unreadable script can have no meaning, though it obviously contains information.

rossum

 

[PEPCIS]

"ricmat:

By analogy, it seems to me that we can all agree that the Enclylopedia Britannica has X amount of information.

Agreed.

ricmat:

If you randomly go in and change 1 character to some other character, I find it hard to believe that the new version would be considered an increase of information.

rossum:

That is not a correct analogy. I duplicated

the gene before altering one copy while the other copy remained unchanged. A correct analogy would be to make a new copy of the encyclopedia and then make a change in one of the two copies. All of the information in the original is still present plus the information in the changed version.

That is not the proposal. The proposal is to find out if the mutated version is an increase in information. By your reckoning with Shannon formulations, the mutated version would be an increase in information. Yet, we know that if this process were continued on and on for MILLIONS OF YEARS, that the mutated/added “information” would eventually render the encyclopedia useless as a source of information.

You attempted to change the analogy by proposing that the original stayed intact, and an exact copy is introduced into the “population,” thereby increasing the information in the population. But that doesn’t work, because the only one that is going to matter is the one that had a mutation. The original would eventually die off.

If mutations occur as quickly as you claim that they do, then EVERY ENCYCLOPEDIA SET produced is a mutated version of the original. Eventually, each of those sets would be rendered useless given enough time. And this is EXACTLY as it should be, consistent with the 2nd law of thermodynamics.

rossum:

You are confusing “information” with meaning.

Actually, you are CONFLATING meaning with information. Information has meaning, which is why it is information. If it has no meaning, it is not information.

rossum:

If I picked up a copy of the encyclopedia translated into chinese it would have no meaning at all to me because I cannot read chinese.

You’re so caught up in defending your theory that you fail to see that the meaning WAS NOT INTENDED FOR YOU. That is why you cannot read a Chinese message written to a Chinese person. DNA is the same thing. It is not intended for you to be able to understand the meaning of its code, because it was not written TO YOU. If you want to understand its meaning, you need to translate it.

rossum:

The mathematical information content would be similar, but the meaning would be greatly changed.

Meaning cannot be changed if it is properly translated, therefore the information cannot change if it is properly translated. Some MEDIUMS present a greater ability to transmit the intended meaning than other MEDIUMS, but this does not negate the fact that information has meaning, and that it is an inherent aspect of it.

rossum:

Meaning is not measurable because it depends on the observer; information is measurable and does not depend on the observer.

Meaning is not measurable because it is abstract. This is the abstract nature of information which I told you about earlier. Information is ONLY measurable via the MEDIUM, but only gives a representation of the quantity of information present. Shannon equations are unable to quantify the information content of any message without an appraisal of the meaning. That is why they limit themselves to a measurement of the MEDIUM alone.

 

[Ender]

"Ender:

information is useless unless it has meaning

Without a measure for meaning this is not a great deal of use.

Meaning exists whether or not we agree on its measure - otherwise we could have no discussion. I may not know how to measure the meaning contained in your statement but its meaning has been created by you, transmitted, and understood by me (albeit surely not perfectly).

Parts of DNA are effectively random because they do not code for anything; for example introns are cut out of RNA before it is passed to a ribosome for protein assembly. DNA in introns is not coded into any protein.

Surely the presence of randomness does not prove the absence of meaning.

Ender:

Did you mean to imply that meaning depends solely on the observer?

In this context, yes. An unreadable script can have no meaning, though it obviously contains information.

Even an unreadable script has meaning to the person who created it, it is just that although the information has been transmitted the meaning has not been understood, but the meaning is contained within the message, waiting to be deciphered. Surely you don’t believe that you create the meaning of my comments.

Ender

 

[ricmat]

What you have though is many different (adapted) encyclopedias all with slightly different characteristics. All of them are less “perfect” then when they started. They all started with the same initial language and meaning. They are as we say, corrupted.

Extinction has far exceeded survival.

Yes, Buffalo, that’s what I was trying to get to.

If you were in the market to buy an encyclopedia, would you buy “the original one” in the example above, or one that has been randomly corrupted many times (or even one time) in the hope of an extremely rare “improvement”?

 

[rossum]

What you have though is many different (adapted) encyclopedias all with slightly different characteristics.

Many of them are effectively identical to the original copies, with merely neutral mutations.

All of them are less “perfect” then when they started.

No. The word “All” is a definite error. Most of those with significant errors are less ‘perfect’, but a few are more ‘perfect’. Again using the example of pi, if the original had “pi = 3.14159” and the altered version had “pi = 3.141592” then the altered version is more ‘perfect’, i.e. more accurate, than the original. You would have been correct to say “Most that have changed significantly”, but “All of them” is a definite error.

You also need to bear in mind that the more faulty versions, “pi = 2.14159”, fail to sell and so are removed from the population, while the more accurate versions, “pi = 3.141592”, sell in larger numbers and so over time form a larger and larger proportion of the population.

They all started with the same initial language and meaning. They are as we say, corrupted.

Changed yes, but not always corrupted. As the example of pi shows, some changes are beneficial to the encyclopedia. There is also the issue of change in the outside world. The original version of the encyclopedia, written many years ago, may have said “The President of the United States is Woodrow Wilson”, which was correct at the time but the environment in which the encyclopedias exist has now changed, so what was accurate then is no longer accurate now. The correct version now would say “The President of the United States is Barack Obama”. Environments change over time so what is “correct” for the encyclopedia also changes over time.

Extinction has far exceeded survival.

Correct. There are many more extinct species than there are extant species.

rossum

 

[ricmat]

Many of them are effectively identical to the original copies, with merely neutral mutations.

It would seem to me that the vast (vast, vast, vast) majority of changes to information/meaning in the original copy cannot be neutral or beneficial.

No. The word “All” is a definite error. Most of those with significant errors are less ‘perfect’, but a few are more ‘perfect’. Again using the example of pi, if the original had “pi = 3.14159” and the altered version had “pi = 3.141592” then the altered version is more ‘perfect’, i.e. more accurate, than the original. You would have been correct to say “Most that have changed significantly”, but “All of them” is a definite error.

You are assuming that the first iteration goes directly to the improved version of pi. If it occurs anywhere downstream, you get the improved pi only in conjunction with zillions of negative changes to completely valid data/information/meaning which would have killed the sales by that time.

You also need to bear in mind that the more faulty versions, “pi = 2.14159”, fail to sell and so are removed from the population, while the more accurate versions, “pi = 3.141592”, sell in larger numbers and so over time form a larger and larger proportion of the population.

So the faulty versions don’t sell, and thus never get to the point of accidentally randomizing to an improved pi. Again, you assume that improved pi occurs on the very first iteration. Even given this extremely simple example, I wonder (from the perspective of a probability calculation) how many randomized copies of the first original would be needed for the improved pi to appear.

 

[ricmat]

BTW - Rossum, I appreciate your willingness to pursue this discussion at the level of “encyclopedia” since it’s a concept that everybody has some familiarity with

 

[buffalo]

Yes, Buffalo, that’s what I was trying to get to.

If you were in the market to buy an encyclopedia, would you buy “the original one” in the example above, or one that has been randomly corrupted many times (or even one time) in the hope of an extremely rare “improvement”?

You know I would buy God’s original copy.

 

[buffalo]

Many of them are effectively identical to the original copies, with merely neutral mutations.

No. The word “All” is a definite error. Most of those with significant errors are less ‘perfect’, but a few are more ‘perfect’. Again using the example of pi, if the original had “pi = 3.14159” and the altered version had “pi = 3.141592” then the altered version is more ‘perfect’, i.e. more accurate, than the original. You would have been correct to say “Most that have changed significantly”, but “All of them” is a definite error.

You also need to bear in mind that the more faulty versions, “pi = 2.14159”, fail to sell and so are removed from the population, while the more accurate versions, “pi = 3.141592”, sell in larger numbers and so over time form a larger and larger proportion of the population.

Changed yes, but not always corrupted. As the example of pi shows, some changes are beneficial to the encyclopedia. There is also the issue of change in the outside world. The original version of the encyclopedia, written many years ago, may have said “The President of the United States is Woodrow Wilson”, which was correct at the time but the environment in which the encyclopedias exist has now changed, so what was accurate then is no longer accurate now. The correct version now would say “The President of the United States is Barack Obama”. Environments change over time so what is “correct” for the encyclopedia also changes over time.

Correct. There are many more extinct species than there are extant species.

rossum

Explain to me again the one’s that became more perfect? Under the heading Jesus - does a new definition appear? Does the picture get more focused? Is there something new added to the picture, like a crown on his head?

 

[buffalo]

Many of them are effectively identical to the original copies, with merely neutral mutations.

No. The word “All” is a definite error. Most of those with significant errors are less ‘perfect’, but a few are more ‘perfect’. Again using the example of pi, if the original had “pi = 3.14159” and the altered version had “pi = 3.141592” then the altered version is more ‘perfect’, i.e. more accurate, than the original. You would have been correct to say “Most that have changed significantly”, but “All of them” is a definite error.

You also need to bear in mind that the more faulty versions, “pi = 2.14159”, fail to sell and so are removed from the population, while the more accurate versions, “pi = 3.141592”, sell in larger numbers and so over time form a larger and larger proportion of the population.

Changed yes, but not always corrupted. As the example of pi shows, some changes are beneficial to the encyclopedia. There is also the issue of change in the outside world. The original version of the encyclopedia, written many years ago, may have said “The President of the United States is Woodrow Wilson”, which was correct at the time but the environment in which the encyclopedias exist has now changed, so what was accurate then is no longer accurate now. The correct version now would say “The President of the United States is Barack Obama”. Environments change over time so what is “correct” for the encyclopedia also changes over time.

Correct. There are many more extinct species than there are extant species.

rossum

Before the encyclopedia had one more perfection it would have many more deleterious corruptions. In x generations it would be useless.

 

[buffalo]

The PI example - if the first deletrious mutation is ?.314159 then the number of decimals of precision won’t matter. The odds against it recovering from the ? is huge.

 

[PEPCIS]

"Ender:

It does appear that the disagreement lies in the definition of information with yours being a strictly mathematical construct measuring the physical content of a message.

Yes. So far PEPCIS has been unable to produce any specifics of what he sees as information.

Well, that’s not true at all. I’ve already given my definition for information, and I am eagerly awaiting YOUR DEFINITION. You have yet to give your definition.

I assume this would explain why the Shannon value differs from the Kolmogorov value.

rossum:

Yes. They are calculated by different formulae and are useful in different situations. Another type of information is Fisher information…

Yes, you have all kinds of information, but what you are REALLY saying is that there are different people who calculate information DIFFERENTLY than other people do. We are still no closer to understanding just what it is that you, or any evolutionist, thinks information is.

Ender:

it seems you have shifted the real battle from one term to another, from information (which you say is measurable) to meaning (which you say is not).

That is an excellent observation, Ender. As you see, the closer I get to getting rossum to admit that information has intrinsic properties of meaning and purpose, the further he will get from giving us a definition for what information is. Talking about “meaning” will only serve to widen the obfuscation.

rossum:

It is evident that PEPCIS has no real definition of information beyond something like “I know it when I see it”. That sort of subjective definition often gets mixed up with meaning.

Too bad that isn’t even close to the truth. I’ve already given my definition of information.

rossum:

Meaning is obviously subjective…

Very true. But that does not mean that it cannot be observed or otherwise investigated.

rossum:

. . . whereas mathematical information metrics are not. Since the discussion on mathematical information was not going anywhere useful, I moved on to discussing meaning in the hope that the change of topic might create a more useful discussion.

You’re not making any sense to me on this. You clearly despise it when ID’ers speak about things that are subjective, deriding us to only put things in terms of objective science. Yet here, you appeal to the subjective nature of “meaning” and tell us that you think that the subjectiveness of the “topic might create a more useful discussion”?

…and we know that DNA sequences have meaning since the information is not random and specific reactions occur as a result of its coding.

rossum:

Parts of DNA are effectively random because they do not code for anything; for example introns are cut out of RNA before it is passed to a ribosome for protein assembly. DNA in introns is not coded into any protein.

First off, DNA is anything but random. Secondly, saying that there are bits of DNA that do not code for anything does not mean that DNA has no meaning. That portion of DNA may have some other purpose other than what you can surmise.

 

[PEPCIS]

"PEPCIS:

Yes, but you still have yet to define “information.” You can’t measure what you don’t know what you’re measuring.

I have a very practical definition of information. Shannon information is that which is mesured by Shannon’s calculation; for details see Shannon’s paper.

Circularity at its finest!

PEPCIS: What is “information”?
rossum: “Information is information.”
PEPCIS: What does that mean!!!?
rossum: Information is “Shannon” information.
PEPCIS: Please elaborate.
rossum: Shannon information is that which is measured by Shannon’s calculation.
PEPCIS: That’s not telling me much.
rossum: See Shannon’s paper.

rossum:

Similarly for Kolmogorov information and similarly for Fisher information. These are all mathematically defined quantities which have been found to be useful by scientists.

Similarly, we would have the same circularity.

Information does not arise spontaneously, but is a product of an abstract mind.

rossum:

I will need to see scientific proof of that please.

Actually, you’ll need to see scientific proof that it does not. That’s called “falsification.” Take note, Leela: that satisfies your demand to make it a theory. BTW rossum, thanks for treating it like the theory that it is.

rossum:

Information can be created by evolutionary processes, as I have shown.

Actually, you haven’t shown. You’ve made claims, but I’m not seeing it. The problem with accepting your claim that information can arise spontaneously through random natural processes, is that it is impossible. You’ve given plenty of evidence that mutations occur, but mutations are not known to create anything that benefits populations.

 

[ricmat]

Does DNA have built-in facilities to prevent copying errors?

 

[buffalo]

Does DNA have built-in facilities to prevent copying errors?

It sure does. Here is how it works. Because it is a language it has syntax and meaning like the English language.

For example:

If u read this sentenc yo know what it menes.

So the DNA can reconstruct this completely.

If you read this you know what it means.

If it cannot then it will get as close as possible.

 

[rossum]

It would seem to me that the vast (vast, vast, vast) majority of changes to information/meaning in the original copy cannot be neutral or beneficial.

That may well be correct in the case of encyclopedias, but it is not correct in the case of genomes. In the human genome about 5% of our DNA codes for proteins while a further 5% codes for switches to turn genes off an on but not for actual proteins. The rest has no apparent function. Mutations in that 90% have no effect on us. Even within the areas that do have an effect there are synonyms, all four of GCA, GCC, GCG and GCT code for Alanine so such synonymous changes are also neutral. The great majority of mutations in humans are neutral. You are correct that the majority of non-neutral changes are deleterious.

You are assuming that the first iteration goes directly to the improved version of pi.

No, I am assuming the existence of a large population of encyclopedias - say a million. In the first generation 995,000 get “pi = 3.14159” while 5,000 get variants like “pi = 23.14159”, “pi = 3.1p4159” and “pi = 3.141549”. The incorrect variants disappear because of their bad reviews just leaving the original “pi = 3.14159”. Maybe we will have to wait for the tenth or the twentieth generation before the first “pi = 3.141592” appears. However, once it does appear then it immediately starts getting good reviews and sales rise. Because of this it will increase in the population of encyclopedias - everybody wants to get the most accurate and most up to date version. The differences in the reviews amplify the better variant relative to the standard “pi = 3.14159”, just as the original “pi = 3.14159” got better reviews than the errors like “pi = 3.1159”.

If it occurs anywhere downstream, you get the improved pi only in conjunction with zillions of negative changes to completely valid data/information/meaning which would have killed the sales by that time.

Which is why we neeed a population of encyclopedias. The versions with “pi = 3.@14159” swiftly disappear with no copies, or very few copies made. Remember that all of our ancestors succeeded in reproducing. Every single one for millions upon millions of generations. Any that failed to reproduce obviously have no living descendants. Errors are ruthlessly filtered out of the population both of organisms and of encyclopedias. The encyclopedia reviewers take no prisoners.

So the faulty versions don’t sell, and thus never get to the point of accidentally randomizing to an improved pi. Again, you assume that improved pi occurs on the very first iteration. Even given this extremely simple example, I wonder (from the perspective of a probability calculation) how many randomized copies of the first original would be needed for the improved pi to appear.

You seem to be thinking in terms of a single copy of the encyclopedia; you need to think in terms of many copies of the encyclopedia being present simultaneously. Where one variant does not survive a different variant, or the original, will move in to replace it.

So to the calculation. Taking the string “pi = 3.14159” and restricting ourselves to single character changes.

1 Deletions. With 12 characters there are 12 possible single character deletions. Two of these are neutral “pi= 3.14159” and “pi =3.14159”, the other ten are deleterious.

2 Insertions. Assume an alphabet of 26 letters, space and eleven punctuation marks for a total of 38 characters. A few insertions are neutral: " pi = 3.14159", “pi == 3.14159” and so forth, but the great majority are deleterious. Two insertion mutations are slightly beneficial: “pi = 3.141591” and “pi = 3.141593” while one single insertion mutation is very beneficial: “pi = 3.141592”.

If we assume that there are ten neutral insertions then we can just match them with the ten deleterious deletions and just calculate the total number of possible insertions. There are 13 possible places for an insertion and at each position there are 38 possible characters to insert. The total number is 3813 insertions = 3.44 x 1020 insertions. If we assume a population of 600,000,000 encyclopedias (one for every ten people) then using my figure above of 5,000 changes per million we have 600 x 5,000 = 3 x 106 changes per generation. We would expect on average to see the first beneficial mutant after about 3.44 x 1020 / 3 x (3 x 106) = 3.83 x 1013 generations.

The calculations for DNA are different. There is a much smaller alphabet (4 characters against 38) and usually more than three targets to hit because the code is redundant. Also for bacteria the population can be in the trillions and generation times are as short as twenty minutes. The form of the calculations is similar, but the numbers are different. This is one case where the encyclopedia example cannot be extrapolated to evolution.

BTW - Rossum, I appreciate your willingness to pursue this discussion at the level of “encyclopedia” since it’s a concept that everybody has some familiarity with

The encyclopedia analogy is not perfect, but I agree that it is a reasonable way to explore some of the concepts.

rossum

 

[rossum]

Explain to me again the one’s that became more perfect?

To say “pi = 3.14159” is an approximation. To say “pi = 3.141592” is a better approximation. Remember that none of the encyclopedias can stand still; if your encyclopedia has Woodrow Wilson listed as President then it is out of date. This is a case where the original copy may well be less good than an amended and updated copy.

Under the heading Jesus - does a new definition appear?

Does our knowledge of Jesus change over time? If it does then the encyclopedia needs to change also.

rossum

 

[rossum]

The PI example - if the first deletrious mutation is ?.314159 then the number of decimals of precision won’t matter. The odds against it recovering from the ? is huge.

There is a population of encyclopedias. That version will get bad reviews and not be copied. The other accurate copies will carry on into the next generation. There is more than one copy of the encyclopedia, most of them good copies of the original.

rossum

 

[buffalo]

There is a population of encyclopedias. That version will get bad reviews and not be copied. The other accurate copies will carry on into the next generation. There is more than one copy of the encyclopedia, most of them good copies of the original.

rossum

Only if someone knows the Pi value is wrong. Perhaps uninformed people will buy the copies with the error and these are the only ones left after a period of time. Then we will have a whole population of encyclopedia’s that have been propagated with errors. Who would know?

 

[buffalo]

To say “pi = 3.14159” is an approximation. To say “pi = 3.141592” is a better approximation. Remember that none of the encyclopedias can stand still; if your encyclopedia has Woodrow Wilson listed as President then it is out of date. This is a case where the original copy may well be less good than an amended and updated copy.

Does our knowledge of Jesus change over time? If it does then the encyclopedia needs to change also.

rossum

Oh really, by adding a horn to his head?

Or changing the word perfect to prefect. Or God is now here to God is no where?

The meaning is completely different and corrupted. After so many copies of the encyclopedia how can I know it is a faithful copy? If one chapter is devoted to motion, how can I walk if the instructions are bad?

This is the same argument that has been applied to the accuracy of the Bible. How was it preserved? It was preserved by diligent copyists and proofreading. In other words an outside observer.