What Is Information?


(Tim) #1

It would seem to me that no new information can be addded into the genetic code for each species. I suppose we could sice dna today and see where that leads, but that is the closest thing to natural selection for the creation of a new species if it remains viable.

If information gets lost, is there enough redundancy to bring it back after several generations? Evolution is basically a corruption in the copying process that has a change in outcome. Although that is probably a gross definition. There are a lot more components involved than the simple copy process.

From what I would say is the simplest reason why there is commonality of species, is because we all live in the same environment even if that environment is as different as water and air. It would be plausible, but not very effecient for each species to have dna that was so different that we would have to learn something new for every single species. The miracle would be how each kind would have to adapt to the environment due to their genetic makeup.

The same basic building blocks even for inanimate objects is just a common set of atomic particles “bonded” together in different combinations.

When God made everything good and perfect in Genesis one it did not neccessarily mean kind and loving. It meant mature and complete. There was no evolutionary process from nothing to a known species. There does not even have to be an evolutionary process in matter from nothing to billions+ of stars either. It could have been perfect as in effecient and going full blast with an age of 14 billion years.

Even the ANE accounts seem to portray Matter existing before God “manipulated” it. Some would even say it existed before God. So creating matter in one instant, and then “animating” it in another allows for the removal of a prolonged period of time. The ANE accounts can also be interpreted as taking enormous amounts of time for matter to do what it needs to do to form all the stars, which the ancients may have missed the reason why it needed so much time was because of the size of the universe.

I think that the enuma elish though was only the record of a change in the solar system which humans themselves experienced. It had little to do with the rest of the universe. They experienced it if you rule out the cataclysmic event known as the Flood. But I assume their account is about as believable to modern man as the Bible is. The Bible says it destroyed the majority of life on earth. The ANE accounts seem to have an even more blurry view of what happened in a prior living environment. Probably because history started over at that point, and the Bible is the only reliable source of information.


Ok, what am I missing here?
(James McKay) #2

The problem with this statement is that it is simply factually untrue, and it represents a fundamental misunderstanding of what constitutes information.

Information is measured in bits and bytes. Ones and zeros. Or, in genetic terms, As, Cs, Gs and Ts.

Information in the genetic code can be changed through mutation. Mutations come in several forms, such as insertion, deletion, substitution, duplication and so on. These have all been observed happening in the laboratory.

Insertion and duplication, by definition, add new information for starters.

People who think that information can’t be added to the genetic code also tend to think that “password”, “qwerty”, “123456” and their date of birth are strong passwords. And then they wonder why their Facebook account gets hacked by someone in Tauri-Hessia.


(Chris) #3

Don’t you mean
Inserection and duplicplicplicplication, by definition, add new disinformation for starters.


(James McKay) #4

No, I mean information. In the sense that every scientist, technologist, information theorist, and IT security professional understands the word to mean.

Information is measured in bits and bytes. More strictly, information is measured by the number of bits and bytes required to tell a computer to make an exact copy of it. Or, to put it another way, the number of bits and bytes that it occupies when it is compressed.

Sure, duplicating something may only add one or two bits to the size of a compressed file, but once you introduce additions, deletions and substitutions into one of the duplicated regions but not the other, all bets are off.

And yes, you may argue about the semantic meaning of information, but what constitutes semantic meaning is highly context dependent. Something like “CHrH5wuha0fXlQPRZeGagHSWD” might seem like gibberish, but it would be very, very meaningful if it were your Facebook password.

Seriously, out of all the anti-evolution arguments that I’ve seen, this is the one that annoys me the most. It annoys me because it’s not just wrong; it introduces wrong thinking about information theory in particular. And wrong thinking about information theory is an IT security risk.


(Tim) #5

So if two different species tried to splice their information on to each other it would be viable?

How does a species during this evolutionary period figure out how to “add” information via cloning?

The evolutionary process is during the incubation period of the newly conjoined egg and sperm regardless of the type of species. Yes, it does work in theory. It also works in the lab. It also can be seen in the fossil record. It just has not worked out in practice enough to change from theory to law, with the constant change of two species or even one species becoming more advanced. I would not rule out that an outside intelligent agent like humans could make it work in practice, thus we have theistic evolution. Since the absence of proof from God is not proof to the contrary it is still a judgment call to accept or refuse the abilty of information to have been so manipulated. If we claim that it happened millions of years before there was any inteligent physical means, i think we are kidding ourselves. We are also back to the argument that God deliberately did it to deceive us, if even in the fossil record. You would have to admit that the sons of god or humankind would have been around since at least the sixth day. And the fossil record took place after the sixth day as well. Now we may make fun of how the record portrays this manipulation in Genesis, but the principle of the theory remains in tack in the Genesis account.

We are attempting ourselves to make information in our own image. Hoping that information will be able to manipulate itself. I suppose when that happens, we as information ourselves will have succeeded.

@ history needed to accomplish reality and the false fact that God is out of the business of controlling the universe.

If humans can speed up and slow down the rate of the process, why would God not be able to do so as well? That is not to say God did it. Pehaps not even in a specific way. God would have known the outcome even if the reason of said outcome changes. Or the opposite if God actually intervened via the binding and loosing principle given to humans.

Water would not have been necessary, but since that was the means mentioned why plead for a miracle. The excuse since not mentioned is that one tectonic plate rose up on its own volition and buried another plate catching all vegetation and plant life on the surface of the former plate as it was “absorbed” back into the earth. I mean it has been alledged (not even sure how to credit this) to have occurred over 10,000 years ago by Native Americans who passed that knowledge down over the generations. Sure lava also pushes up whole mountain ranges, but not even that is one plate trapping another plate fast enough. Nor does this negate the need to find the most reliable energy source via dating methods. Age does not change. The argument is about the “history” needed to produce results used as interpretive proof of ones belief system.

A footnote on Adam’s narrative and the creation of humankind reffered to as the “sons of god” created in the image of God 6 days after the universe: Adam is one as agreed by all. What is not agreed is the notion that Adam was the only one and created twice. Lets attempt to erase from human understanding that Adam was the only person created on day six. Even if Adam is just the place holder name, and the Garden is only figurative, why use that to claim it negates reality. The reality that there were multiple males and females created on day six. That is what it says. Saying that Adam was one or all is not a contradiction. If Genesis 2 is figurative it does not negate a literal sixth day creation of humans, so Genesis 1 still stands.

Now the claim is that the Garden was not a local event. Why argue that? That is still the figurative argument stated in a different term. If it was not literal it was not local either. So within an unspecified passing of historical time, a part of the earth became barren and needed to be replanted and a human was singled out, who evidently did not mate or lost a mate premature.

This man in keeping with ANE bragging rites declared that his new mate was the mother of all mankind. God seemed to agree; started the process that would later wipe out the ability of the other sixth day created beings to enjoy earth, and only allowed the offspring of Adam to survive. Of course Adam’s offspring now deny that it all happened and instead replaced the past with imaginative tales of super humans and brushes with ET, who are more than likely the other sixth day humans who no longer enjoy the physical earth. Do I believe what i just typed? Not sure, but the claim that the evidence is there is just as real as those who declare the evidence of EC. Or even more than those who claim there is no evidence of God either. The truth does not live in our belief systems. It is sad though that we accept the ability for all to retain their own truth. But such is the nature of free will.

I do not even think that claiming Genesis 2 is both literal and figurative is a contradiction. For one no one is denying that something happened. If it was potrayed in a figurative manner cannot change reality. As pointed out reality does not need to be changed either to fit the figurative account. If Genesis 2 is an actual reality it does not contradict without special pleading by creating a history that is assumed and never in the record. The verses that can be in either narrative act as a segue that seems to be convenient as a literal objection. Why create a figurative objection? We use the passage of time to argue that Genesis 2 is figurative, and even for the argument of EC. But when the passage of time is implied, we tend to disregard it because it literally was not mentioned.


(James McKay) #6

Hi Tim,

I’m having a little trouble in trying to formulate a reply to you here. I think this is mainly because you’re bringing up a whole lot of different concepts at once.

Here’s a bit of advice for you. When you’re trying to tackle the subject of evolution, break it down into different components and discuss them separately. The biggest mistake that a lot of people make when discussing evolution is to treat it as a single, take-it-or-leave-it monolith, and I think this causes more confusion than anything else. Just because I’m saying that mutations can produce new information doesn’t mean that I’m also saying that “God is out of the business of controlling the universe”: I’m not saying anything of the sort.

I also get the impression that you have a few misunderstandings and misconceptions about what scientists actually teach about evolution, and about the terminology that’s used. For example:

How do you actually think of evolution (macroevolution in particular) as happening? It seems to me that you’re thinking in terms of large, dramatic changes happening in single individuals to the extent that you end up with a lone straggler so far ahead of the others that it can’t find a mate. That’s not how evolution works: it happens by a large number of small, inconsequential changes spreading through whole populations and accumulating over the generations. In other words, macroevolution is a whole series of consecutive microevolutions: nothing more, nothing less.

You also need to understand correctly what is meant by “theory” and “law” in science. A theory is not a wild guess; it is an explanatory framework that is supported by evidence. Likewise, a law is not a rule that can never be violated; it is an explanatory framework that can be represented as a mathematical equation. The idea that “theories” graduate to “laws” simply doesn’t reflect the reality of what these words actually mean.


(Tim) #7

@ jammycakes

Species is the most widely known and used description of the taxonomy break down of biological life. It used to be understood that species generally do not inter-breed probably for this reason. While it is true that some species never will, trying to fit “kind” from Genesis 1to common usage of species no longer fits into the frame work of feasible conversation. There are in fact large genus of species that do just fine breeding together.

If we really are going to burry our heads in the sand and say kind is not species, we should at least set the record straight. Kind was never strictly species, it was just the simplest way to communicate what was undestood. Wether or not the argument continues just merely for the sake of the argument or not, perhaps we can find a better term in taxonomy that properly portrays what is being communicated instead of just strawmanning the topic. I do not think genus is proper. Perhaps family would be better as Genesis 2 and 3 deal with human families why not just use kind as family? We even use the proper family names like in the breakdown of tree species even though each family name also breaks down into hundteds of species. I do not see the names “beast of the field” “birds of the air” being that off to what ANE may have undestood as the seperarion of species after their kind or family description.

What we fail to see in the forest of species, is that Adam was said to be the keeper of taxonomy. A taxonomy that may have been totally irrelevant to even those living 2000 years after the Flood. What is not irrelevant is no one recorded drastic changes in what these animals remained to be since they were created and named by a human who by name represented the whole of the human race before he was disgraced from the rest of humankind.

Now, can animals adapt over time in order to survive in changing conditions? YES

Can they change over time and become something they were not supposed to be?


(Mitchell W McKain) #8

Of course not because I don’t believe in your control freak god who has everybody’s life planned out ahead of time. I believe in a God who chose love and freedom over power and control, thus all living things can and do change to become what they choose to be because that is whole point for which God made life in the first place.

Another one of your additions to the text of the Bible?

I am an advocate of Sola Scriptura and that means people shouldn’t be rewriting the Bible to fit what they want to believe.


(Tim) #9

Do you know the definition of taxonomy? I do not view God as a control freak. More a scientist with a bent towards efficiency. If a person is asked to name all the animals, why not classify them as well?

I think the issue is you see the time of Adam as a Disney movie of prehistoric fabrications. There is little indication of what life was like. There was agriculture of both plants and animals. There is a strong possibility they were more advanced than we are. They did have the best of the Spiritual and Physical aspects of reality. Not to be confused with history after the Flood when Humans had to start over in the stone age.

For the record I do not “have” a God. I just converse about God as recorded in the Bible whether that experience was good or bad. Leave the god making experts to other humans with more of an imagination. Reality is not what we make. Reality is what we experience. Free will is the choices we are allowed to make, regardless of the consequences. Consequences are the result of the universe running in accordance to the rules the universe was created with. If you think that rules are too deterministic, imagine how fun life would be without them. At least imagine for a split second because that may be all the time you get without rules.


(James McKay) #10

You mean the idea that Noah had machine tools and advanced technology? Why not just go the whole way and claim that he was assisted by ancient aliens?

In actual fact, now that I mention it, Giorgio A Tsoukalos’s proposal that Noah’s Ark was a DNA bank is far, far, far more plausible than Answers in Genesis’s hyper-evolution…


(Mitchell W McKain) #11

Google: the branch of science concerned with classification, especially of organisms.

No the Bible says nothing about Adam being in charge of the classification of living organisms. All it says is that God brought animals to Adam and Adam named them and the way it says “that was their name” could be taken to mean that there was only one language at the time, especially in conjunction with Genesis 11 where it has God changing this to many languages.

That is another big difference between us then. A scientist investigates nature to find out how things work. But God already knows such things so calling Him a scientist would be nonsensical. I see God as a parent, who created the universe and life not as some kind of experiment but for an eternal parent child love relationship.

No. I see the way some people trivialize the story with excessive literalism as turning the story of Adam and Eve into a Disney movie for children.

There is no possibility they had anything like modern technology. They had no need of anything like that. And there is no indication that there was any loss of technology in the the story of Noah and the flood either. This is sounding more and more like you are rewriting the Bible as a science fiction fantasy novel. Nothing wrong with that really. The sci-fi Noah movie was cool. But that is just entertainment and I am not going to confuse either with reality and consistency with the objective evidence goes a long way towards making the story real.

I think the rules are a necessity for the existence of the physical phenomenon of life. I do imagine an existence without such rules after we die. And I think this explains part of the reason why we confront a divergence in ultimate human destiny known as heaven and hell. The other part being that only God has what is needed to make an eternal existence worthwhile.


(Chris) #12

I take it then that you are referring to “Shannon Information” in which H(S) measures the optimal compression of the source messages. However this is actually not information but the entropy of the source message, S. As such, it is information ABOUT the message but it is not a measure of the semantic information IN the message. In fact the message need not have any semantic meaning.
Ref

“Frequently the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem. The significant aspect is that the actual message is one selected from a set of possible messages.” (Shannon 1948) [ref p6]

Shannon’s theory, taken by itself, is purely quantitative: it ignores any issue related to informational content.

When I was doing database programming we were asked what is the difference between data and information. Briefly “While data is an unsystematic fact or detail about something, information is a systematic and filtered form of data”. [More]

So the number of bits and bytes a file occupies when it is compressed is a measure of the size of the file. The file can contain data of variable quality, right down to completely random noise. From this data information can be extracted. However semantic items, such as meaning, reference or representation, are not amenable of quantification. [Ref p7] (Although in some cases it can be quantified.)

So your statement that " Insertion and duplication, by definition, add new information for starters." actually means that it increases Shannon Entropy but it will only rarely increase information. Mostly, as in my example, using only insertions and duplications, " Inserection and duplicplicplicplication, by definition, add new disinformation for starters." the entropy may have increased but the semantic information has decreased.

You are right that semantic information is highly context dependent, or highly specified, as in the case of your password, which would be complex specified information, and any insertion or duplication would reduce its information content to zero.


(James McKay) #13

Yes Chris, it is Shannon information that I’m talking about when I talk about “information.” Good to see you’re on the ball there.

There was a thread here a couple of years ago (exactly!) that had a much more rigorous treatment of this subject. You may want to take a look at it:

You’ll need to read through it (at least the original post) to grasp it properly, but the basic gist of it is that the claim that mutations can not add new information – in whatever sense you choose to define information – is either unproven, unprovable, or wrong.


(Chris) #14

It will take a while. Just the first post is 5 pages when I copy it into my word processor and I’m not going to read the remaining 94 posts. From my first quick read I’m going to stick my neck out and say that @Swamidass is wrong. More after I’ve had time to digest it.


(Chris) #15

@jammycakes, @Swamidass

Swamidass has made a basic confusion about the meaning of information, and I say this knowing that his “PhD is specifically in “Information and Computer Science”, with emphasis (in my case) on information.”

In the second paragraph of Shannon’s paper of 1948 he says “The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point. Frequently the messages have meaning; that is they refer to or are correlated according to some system with certain physical or conceptual entities. These semantic aspects of communication are irrelevant to the engineering problem.”

The information Shannon is referring to is the information required to transmit a message but not the information in the message; as Shannon says that is irrelevant to the engineering problem.

While the amount of information required to transmit a message is often called Shannon Information it is actually a measure of the Entropy, so it should be called Shannon Entropy. Unfortunately this misnomer encourages some people to conflate the calculated entropy of the message with the semantic information within the message. This is similar to conflating the weight of a box with the contents of the box. For shipping purposes the weight is the relevant measure and it doesn’t matter if the box contains diamonds or dirt.

But once you conflate the entropy OF the message with the information IN the message then it is an easy step to conclude that maximising the entropy maximises the information contained in the message; and hence conclude that information is maximised when the message is completely random.

Consider Abraham Lincoln’s famous Gettysburg Address (1473 characters)

Four score and seven years ago … and that government of the people, by the people, for the people, shall not perish from the earth.

Replace it with the following and you have reduced the Shannon Entropy drastically but lost the information contained in the message.

111111111111111111111111111111 … 11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111

Replace it with the following and the Shannon Entropy will be high but again the information in the message is lost

1rdn74mao.28spawjjk16wpstedyw0 … 12bd03payentwpd674ba;m81$%(dgepsn52plFQ2mduw06%dkwatpeyW710FtD&1brghepWQL$v%1hsdeQyw72 qrT2H84s-


(James McKay) #16

No Chris, @Swamidass does not conflate Shannon information (entropy) with semantic information. In fact he clearly differentiates between the two when he talks about information content and mutual or shared information:

NOTE TO THE CONFUSED: This discussion can get confusing because of the many definitions of “information” in common speech, and also that there are two main types of information that work in different ways: (1) entropy or information content and (2) mutual information or the shared information. Information content is the amount of information in a single entity (and is measured as the entropy). Mutual information is the amount of information shared by multiple entities (and is a measure of commonality, and is equal to the difference between two entropies). When communicating with the public, it is hard to keep these two different types of information straight without devolving into dense technical language. But if you detect a contradiction in what I wrote, this is probably the reason. Because these two types of information behave similarly in some cases, and exactly opposite in others.

The important point that he makes is that semantic information is (a) poorly defined, and (b) impossible to quantify (since it is impossible to rule out the fact that there may be a meaning in a seemingly random string of characters that we had not managed to identify and decode). It is this specific point that you need to address, because it is this specific point that tells us that it is simply not possible to claim, on information theoretic grounds alone, that mutations can not produce new semantic information.


(Chris) #17

I think I actually covered this In my previous post but I will risk repeating myself.

Claude Shannon in 1948 was addressing a technical problem in communications and was using the word information in that context and within the conventions of that time. He was not trying to measure information itself but only the size of the message that needed to be sent via the communications system. He said specifically that "These semantic aspects of communication are irrelevant to the engineering problem.

In his paper Shannon referred specifically to “7. THE ENTROPY OF AN INFORMATION SOURCE” but unfortunately over time this has come to be called Shannon Information, a misnomer.

Consider the analogy of sending a parcel by a courier company that charges by volume, the size of the parcel. They measure the length, width, and depth and calculate the volume. The calculated volume provides information about the size of the parcel you are sending, but volume is not the parcel nor the contents of the parcel. Neither is volume in general information about the parcel but only the specific calculated value.

Similarly Shannon Entropy is a measure of the size of the message. It provides information about the message but it is not the message nor is it the information within the message. Hence it is invalid to equate Shannon Entropy with Information as Swamidass does in his post.

Now if you want to talk about what information actually is that is a whole other subject and beyond my pay grade. I did find Werner Gitt’s book “In the Beginning Was Information” helpful in in this regard.

Unfortunately @Swamidass has advised me that he is not able to respond in this forum.

[edit]
The message itself is not information either. The information is coded in the message. I might start with the thought about what we will have for dinner tonight. That thought is coded into English words which are represented by strings of characters which make the message, “Bangers and mash for dinner tonight”. My wife gets the message and decodes the letters to words to information. She might then reply “What, again?”

The encoding and decoding steps can go wrong if the sender and receiver use different coding systems; like when I downloaded some product information and found it was in German. The information had been encoded into the message and transmitted successfully but I didn’t have the correct system to decode it back into information. I had this problem in Germany last year when trying to order a gluten free meal in a restaurant. I could encode the request into English, transmit it successfully, the hearer received (heard) the message but could not decode it. In the reverse direction I could receive but not decode their messages. (We ended up with a meal of spargel.)

A message can be encrypted and it will then appear as gibberish or random data because we don’t have the correct decoding system. This is not the case with random noise which does not have information encoded by any system. This is another reason why Swamidass is wrong when he says " Another surprising result is that the highest information content entity is noise , exactly the opposite of our intuition of what “information” actually is."


(James McKay) #18

Yes Chris, you are repeating yourself. You are also repeating points that @swamidass made in his post. And the points you are repeating do not address the question at hand.

There are two different meanings of the word “information” under discussion here.

  1. How @swamidass uses it (to refer to Shannon entropy)
  2. How you insist it should be used.

The point I made is that @swamidass acknowledged your definition. He referred to it in his post as “shared information.” He also said that it does not have a rigorous mathematical definition, and can not be properly quantified. And however you insist that the word “information” should be used, a rigorous definition is essential before you can insist that mutations can not introduce more of it.

Nothing you have said addresses this specific point.

I’m not ruling out the possibility that there may be an information theoretic argument against undirected evolution, but if there is one, it will be complex, difficult to understand, and even harder to fact-check. And even if such a proof is discovered, it won’t change the fact that life in the Cretaceous was very different from what it is today; it won’t change the fact that humans and animals have the appearance of being related; and it certainly won’t reduce the age of the earth to six thousand years.


(Mervin Bitikofer) #19

At the risk of jumping into this without having read all leading up to it … I think I know what Swamidass (and others before him) mean when they refer to noise as having the highest of information content. Think of it this way. You can probably gs wt I’m sag here evn thgh mny of the Let***s are msnisg or scrbelamd. And the reason you could probably read most of that sentence is that it was predictable enough for your brain to fill in or correct the missing parts. In other words, letter-for-letter, due to redundancy, the characters are not each carrying a lot of information weight and the message can still come through with a good many of them missing or out of order. Contrast that with me sharing a phone number with you: 271-555-8?42. In this case each and every character is critical and there is no redundancy or any way for your brain to fill in or correct even just one digit since each digit is independent of all others around it. You have no basis for guessing which of the possible 10 digits the ‘?’ should be. This latter string of information is “noise” in the sense that every bit of information is independent of the bits that surround it. I believe that is what is meant when ‘noise’ is referred to as carrying, letter-for-letter the most ‘information’, even if it is “truly random” (whatever that is - and if it even exists). All it would mean is that it is, as you say, indecipherable to everybody as there would be nothing there to decipher. So your point is well-taken that it certainly isn’t information in the common sense way we think of it. But distinguishing those situations would in practice often be impossible for us. 59265358979323846 will be noise to one person, but a segment of the digits of pi for those who can recognize it as such. And if such a sequence was turned into an electrical signal sent to speakers you would hear something indistinguishable from white noise (static) - precisely because of its high information content density, whereas if you send the digitized version of the decimal for 1/3 (33333333…) then you will hear … nothing. Because there is almost no information added in the sequence by any particular ‘3’. Remove or rearrange as many of them as you want and the sequence suffers no loss whatsoever.


(Tim) #20

A nice reference to pi on pi day.