Irreducible Complexity Debunked?

[lucaspa]

What a strange, strange definition. It looks like he couldn't understand Shannon information and simplified it.

"If I am sitting at a radio transmitter, and can only transmit zeroes and ones" is not a sufficient description for any information measure.

The equation shows increase in information in terms of "bits". It is what is chosen over the possible choices. There is no "probability" involved. But yes, if there are 10 possibilities and you choose only 1 (10% of the possibles), then the information is going to be more than 1 bit.

His information measure can't say squat about the evolution of a particular genomic string, too, when that is the exact thing which the ID exponents need to explain away.

Doesn't matter. What Dembski's equation shows -- ironically since Dembski is an IDer -- is that natural selection will always increase information. We don't have to go to the base sequence on DNA. Dembski has undercut all of ID by showing that natural selection is a means of increasing information -- exactly what ID claims evolution can't do. That's the beauty of the situation. Dembski, and IDer, refutes ID.

 

[shernren]

Doesn't matter. What Dembski's equation shows -- ironically since Dembski is an IDer -- is that natural selection will always increase information. We don't have to go to the base sequence on DNA. Dembski has undercut all of ID by showing that natural selection is a means of increasing information -- exactly what ID claims evolution can't do. That's the beauty of the situation. Dembski, and IDer, refutes ID.

Dembski's equation shows that NS will always increase "information" ...

... information measured according to an ill-thought-out information metric that, to me, says squat about actual evolution.

So should we attack the propriety of the metric itself, or do we run with the silly metric and show that even with it ID is an epic fail?

You gotta give them credit. At least they make entertaining targets.

 

[Buho]

It has one saving grace over other usages; it actually works.

Come on, Barbarian. You know what else works? The internal combustion engine. But just like Shanon's theorems, it has nothing to do with information. Shannon is a red herring.

Shannon's understanding of information is what allows us to pack more of it into a smaller pipe.

And that has nothing to do with the common understanding of information. I believe Shanon himself later wished he used a different word than "information," because it's an inaccurate descriptor.

If that were true, the evolution of a nylonase would not have happened. It was by a frameshift, which would make all the "letters" into nonsense if it were a language.

A frameshift does not necessarily produce randomness. As with the written language, DNA has built-in redundancy. In the case of the frameshift, it was in a repetitive section.

 

[shernren]

Come on, Barbarian. You know what else works? The internal combustion engine. But just like Shanon's theorems, it has nothing to do with information. Shannon is a red herring.

And that has nothing to do with the common understanding of information. I believe Shanon himself later wished he used a different word than "information," because it's an inaccurate descriptor.

I humbly disagree, as would most communications engineers. Shannon information, for example, is one of the key ideas behind how you would allocate different pipelines to different Internet transmissions - and it doesn't get more "informational" than that.

Shannon information says that the more choices available to the "source", the more information will be contained in the "signal". That's actually fairly self-evident. It explains, for example, why we use English (and other languages) to communicate, and not binary.

Let's say you're reading an English book. Treat it as a signal from a source: an author pulling out sequences of English letters that s/he wishes to communicate to you. You know beforehand that (disregarding spaces and punctuation) the contents of this signal will be a series of English letters, of which there are 26.

Okay, enough mathnerding. Start reading. The first letter is "A". Wow! Before you read that, there were 26 possible choices that letter could have been, each with its varying probabilities. But your reading it narrowed the possibilities down from 26 to 1. The second letter is "n". Again, that letter cuts down 26 possibilities to 1. And so on.

Now let's say you're, for some obscure reason, reading binary code. Brace yourself. The first character ... is 1! But that's not too exciting. It was only going to be 0 or 1, anyhow. You haven't really learned very much, as opposed to when you found that the first letter of the book was "A". The second character is also 1. So is the third. The fourth is a zero, but the fifth is a 1. Yet each of those symbols doesn't reveal a lot.

Why English, not binary? Because there are a lot more possible permutations of a larger symbol set, and so any one word carries a lot of information. If I wanted to construct a language in binary, I would be hard-pressed: there are only 255 possible binary "words" of eight "letters" or fewer. In contrast, how many "words" can you form out of combinations of eight or less English letters? 217 trillion.

The wider the distribution of characters available to the source, the more information it can convey in a given signal.

(Though yes, Shannon information is quite a red herring to the evolution discussion. But not for the reasons you make out.)

A frameshift does not necessarily produce randomness. As with the written language, DNA has built-in redundancy. In the case of the frameshift, it was in a repetitive section.

What mechanisms, exactly, exist in DNA that allow it to retain function in the event of a frameshift? I want to have a clearer answer before I respond to this claim.

 

[The Barbarian]

Barbarian on Shannon's equation:
It has one saving grace over other usages; it actually works.
​

Come on, Barbarian. You know what else works? The internal combustion engine.

But it isn't about information. Shannon's equation is what makes it possible to move huge quantities of information in very restricted pipes. It's what makes it possible to communicate with spacecraft millions of kilometers away.

But just like Shanon's theorems, it has nothing to do with information.

Engineers who work with information would find that amusing.

Barbarian observes:

If that were true, the evolution of a nylonase would not have happened. It was by a frameshift, which would make all the "letters" into nonsense if it were a language.
​

A frameshift does not necessarily produce randomness.

Let's take a look.

Sdenk t udi tee nbegrijpelijk ezi ni.

Filtok vo spifigleek orn bekaa bor.

For each of these, ask yourself: Do you think this is a comprehensible sentence?

Would a frameshift do this? Is one or both frameshifted?

As with the written language, DNA has built-in redundancy. In the case of the frameshift, it was in a repetitive section.

That's how most new genes appear. Redundancy permits one copy to be mutated without harming the organism.

But what do you think about the strings of letters I generated?

 

[Chesterton]

I linked it embedded in a youtube link on page 5 or 6:

Thanks for that, I missed it before.

At :50, it says "How could this have evolved?". You know I want the video which says "How this evolved." It's like the difference between "Here's how we think the spontaneous generation of life might occur" versus "Here's how we know the spontaneous generation of life occurs". I don't hold it against science for not being able to do something difficult (prove evolution) - I just want to reserve judgment until it's done. By the same token, don't hold it against me (by considering me ignorant and backwards) because I want to know if something is true before I consider it true.

Also, I can think of a more appropriate Boston song to accompany a video about a bacteria: "A Man I'll Never Be". Har har.

That's the Argument from Personal Incredulity. It's not a valid argument.

See post #52.

Uh, not quite. What is happening is that we are testing evolution to see if it is true.

Right; let me know the result when the testing is done.

If we couldn't come up with a way to incrementally build the flagellum, then we would have a real problem with evolution by natural selection.

And being able to imagine how a historical event may have occurred is not a solution to a problem, it's a hypothesis.

That's not how it works. Your "evidence" is simply lack of imagination. That's not evidence.

No, the evidence of its improbability is the lack of evidence (as of now) that it could have occurred. I'm not sure about the science lab, but no matter how imaginative one is, one's imagined ideas are never allowed as evidence in a courtroom.

And "already believe the theory" has no weight, because our job as scientists is to disprove theories. It doesn't matter what we "believe", if the data refutes the theory, then the data refutes the theory.

I wish that were true. This forum wouldn't be here if it were. There are no forums here containing debate about the shape of the earth. But there are forums here containing debate about world history; the reason being that, unlike the shape of the earth, no one living can observe or measure historical events, such as the Crusades or the creation/formation of flagellum.

That's how ID and creationism got falsified in the first place. Remember, they used to be the accepted theories in science. All scientists "believed" them. But the data showed they were wrong.

They've been falsified? Cite?

LOL! Sorry, but that is nonsense. At least 99.99% of all proposed scientific theories/hypotheses have been shown to be wrong. If what you say is true, then that couldn't have happened. Remember, people once thought the earth was flat. They don't accept that anymore. According to you, we must. Once upon a time, everyone thought that the earth was the center of the solar system. We don't think that anymore, either. Again, if what you said was true, then we would still think that. And before 1800, all scientists thought the earth was indeed less than 10,000 years old.

I was referring to the theory of evolution only, not all scientific theories, or any other science.

So, you think that there are created "kinds" instead of a common ancestor, right? Then there is evolution of new species within the "kind". Well, Darwin thought the same! You've just stated evolution! Congrats. You do "believe" in evolution after all.

I said "even if". I'm agnostic here. Unlike you, I have no firm "belief" regarding the matter.

 

[Drekkan85]

Thanks for that, I missed it before.

At :50, it says "How could this have evolved?". You know I want the video which says "How this evolved." It's like the difference between "Here's how we think the spontaneous generation of life might occur" versus "Here's how we know the spontaneous generation of life occurs". I don't hold it against science for not being able to do something difficult (prove evolution) - I just want to reserve judgment until it's done. By the same token, don't hold it against me (by considering me ignorant and backwards) because I want to know if something is true before I consider it true.

But it's not about considering truth. ID made a claim - that the bacterial flagellum is irreducibly complex. Without any of its parts it wouldn't function therefore it must have been designed not evolved. Thus, if you can show how an evolutionary process could have led to the flagellum, you've shown the system is not IC. Without EC there is no ID - and so the burden becomes showing how it could have evolved.

That's where the video comes in. It demonstrates a step by step process, one that does not require any huge leaps, and each step gives an evolutionary advantage to the step before it. Now, that may not be how it happened, but it shows that the system isn't IC.