Evolving New Information

You are a veritable font of misinformation.--Percy

Here's something Shannon actually did say, it's from his paper A Mathematical Theory of Communication: Semantics are not part of information theory.--Percy

You're going to argue about the interpretation of plain English?Shannon's information theory was proposed as a solution to the engineering problem of communicating information in the presence of noise. There are no semantic aspects to information theory. The only people who think they've come up with theories of information that include semantics are creationists like Dembski, Gitt and Spetner, and they've never bothered to connect their ideas to reality.But though these people think it is possible to include semantics in a theory of information, even they understand that Shannon information does not include semantics. I couldn't say it any better than creationist Perry Marshall at his Cosmicfingerprints website: Could I suggest that you reverse the order of your "reply first, research later" approach?--Percy

No one's saying there's no such thing as semantic information. Of course semantic information exists.But Shannon's theory of information does not include semantics. It only covers aspects of information that are quantifiable. You cannot adapt Shannon information to make arguments about semantic information. Anyone making claims about semantic information cannot cite Shannon.--Percy

Hi Lucy!This sounds like misinformation to me. Where did you find it?--Percy

In case it helps, my example in Message 1 was concocted solely to illustrate how a point mutation can increase the amount of information in a genome by adding to the number of alleles for a gene. I didn't go to the trouble of actually making sure my codons were valid because it wasn't relevant from an information standpoint.Would it be helpful to edit Message 1 to remedy this problem? I could include an explanation at the top that such an edit was performed, and I could include the entire original message in a hide section.--Percy

There are no mistakes. The organism was hypothetical, its genes were hypothetical, its alleles were hypothetical, the codons were hypothetical, and the amino acids they code for were hypothetical. The example was an illustration of how information increases in a population's genome, not of how DNA codes for proteins.I could change the example slightly so that the codons are real, but everything else can only be hypothetical. To take the example any further into the real world would take away the simplicity that makes it an easily grasped example of how information increases in a population's genome. Anyway, here's the example again, this time with codons that represent different amino acids:These are the alleles of the original gene:

• GGAAGC (green eyes)
• GGAAGA (blue eyes)
• GGCACG (yellow eyes)

These are the alleles after the mutation produces a new allele:

• GGAAGC (green eyes)
• GGAAGA (blue eyes)
• GGCACG (yellow eyes)
• GGCAAG (brown eyes)

The number of alleles was 3, which can be represented in log₂3 = 1.585 bits, and then it increased to 4, which is log₂4 = 2 bits, and that's an increase in information of .415 bits.--Percy

We're on the same wavelength, but I'm going to explain at greater length for the benefit of others before moving on to the rest of your message.You are correct, except that my example is for a hypothetical population where we know each and every allele of each and every gene. This is because I don't want to deal with the complexity associated with the possibility of incomplete knowledge.So every organism in the population has this gene, and if you examined this gene across all the organisms in the population, you would find only 3 unique alleles. Then I postulate that one of the offspring receives a point mutation in this gene whose result is a unique allele, and naturally it will pass this allele on to its own offspring. A few generations go by, and if you now poll this gene across the population you will find 4 unique alleles. The increase from 3 to 4 alleles in this gene represents an increase in information in the genome for the population.At heart communication is the problem of communicating one message from a set of messages. The size of the message set (and the probability of each message, but in crafting examples I always make the probability of each message the same) determines how much information is communicated in a single message. 3 messages takes 1.585 bits, 4 messages 2 bits.So in biology, the problem is how to communicate an organism's genes to its offspring. For the gene in question there were 3 alleles, and it takes 1.585 bits to communicate one message from a set of 3. When the allele reaches the offspring, 1.585 bits of information will have been communicated. When a mutation produces a 4th allele then the message set is of size 4, and communicating one message from a message set of size 4 requires 2 bits, so when an allele from a gene with 4 alleles reaches the offspring then 2 bits of information will have been communicated. Yes, but I try to avoid mentioning this in order to keep things simple. Yes. For others, let me add that .33 is the probability of one of the messages from a message set of size 3, and that it is necessary to summarize across the message set in this fashion when the probability of the messages is not equal. Oh, and that the probability is actually 1/3, not .33. 3*.33=.99, which isn't 1, and the sum of the probabilities must equal 1. I only mention the roundoff error because it's a minor factor in what you say next: When compared with your previous equation the rounding issue is now apparent, but this looks right to me. You're more correct than me. I'm erring on the side of simplicity, though I do say "after a few generations" to give the allele time to spread through the population.--PercyEdited by Percy, : Spelling.

Yes, exactly right, noise can increase information. And I not only assumed it, I proved it mathematically. If you'd like to challenge this then you have to respond to the example I provided in Message 154.It's analogous to the children's game of telephone, where the message delivered at the end of the line is usually different from the original message. Sometimes the modified message is shorter (less information), sometimes longer (more information).When additional messages are added to a message set, as can happen during copying, then the message set size increases and the amount of information communicated increases. The Bible is a good example. Some Bibles contain more information than others, and one common way this happened is through scribal error. Noise, if you like.--Percy

The Wikipedia article on spontaneous abortions cites research indicating that the rate is around 30%. Like you I thought I remembered hearing of higher rates, but of around 50%.--Percy

When you're replying to a message, the full text of the message you're replying to is displayed immediately below the text box where you type your reply. If you scroll down a little to see it, you'll see that it's an exact replica of the message, including all annotations such as the author, the message number, who it's a reply to, and so forth. You can copy-n-paste from this text into your own message. And if you need the precise exact raw text that he used, perhaps because you want to keep his use of bold and italics or to maintain his URLs, then click on the "Peek Mode" radio button at the top right of his message and cut-n-paste from that.--Percy

I liked the way you framed the language analogy, with each child receiving the message performing selection according to whether it made sense. In this analogy mutations occur at the level of words, or at least syllables.Your press release example is a different analogy. I haven't found this approach very useful in the past, I think because to creationists it seems too obviously like gradual degeneration of information.I don't know if Cavediver was already aware of this when he introduced the computer instruction analogy, but this is a common approach to artificial life. My memory may be off, but I think Tierra and Avida may be examples of this approach.--Percy

Sure it can, and I provided an example back in Message 154.Everyone already agrees that an intelligence can create information. Why are you trying to prove something everyone already agrees with?You're ignoring the central issue: whether random chance can create information. If you think it can't then you have to find the flaws in the many examples provided to you, such as the one in Message 154.--Percy

This is the whole argument from ID in a nutshell. Life is incredibly unlikely, therefore some intelligent agent must have created it. Because any intelligent agent must itself have been at least as complicated as the life it created, it couldn't have come about naturally and must itself have been created by an even more intelligent agent. We can't keep inventing intelligent agents forever, so at some point the intelligent agent must have been God. The whole argument comes down to, "Life is incredibly complex, therefore God was the ultimate creator of life."It's an interesting idea that an intelligently creative process might leave behind detectable signposts, but instead of demonstrating their principles with simple examples involving a few nucleotides, investigators like Dembski, Durston, Abel and Trevors instead begin with real DNA or entire protein families. Aside from whether what they're doing is a measure of complexity, they haven't shown that complexity requires an intelligence. They haven't even shown the ability to distinguish between a random DNA codon sequence and an actual one. Being able to do that would be the first real demonstration that they're on to something.Now here's the real kicker from the point of view of science. Instead of building a growing and increasingly persuasive body of research, IDists declare that their conclusions are true before their research has demonstrated this, then when scientists object they demand that schools teach the controversy. The fact of the matter is that there is no controversy within science. IDists and creationists in general are creating a social controversy by placing one group of Christians at loggerheads with the rest of society because they're concerned that science is a threat to faith. There's no scientific controversy.But how do we convince people like yourself that Dembski et. al. have not yet demonstrated their claims? Certainly Dembski is claiming in no uncertain terms that he's demonstrated CSI, and why would he lie?I don't believe Dembski is lying. I think few people in the creation/evolution debate are lying. I think they all sincerely believe what they are saying. So how do you decide if Dembski and company are right or wrong?The fact of the matter is that whether Dembski is right or wrong is not what's important. If you want to believe that Dembski and company are right, then that is your prerogative. Many new scientific ideas begin with someone thinking he is right when everyone else believes he is wrong. What's important is how do you get from just one or a few thinking they're right to most people thinking they're right.The answer is the scientific process of research, peer review, publication, wider peer review, replication, more research, and so forth. This is the very process that just isn't happening with ID. You can believe that Dembski and company are right if you want, but they haven't convinced many in the scientific community that they are right, and by and large, especially in the case of Dembski, they just bypass the scientific community and take their arguments to those worst qualified to assess them: the general public.Abel, Trevors and Durston et. al. are doing the right thing by publishing research papers in the legitimate technical literature. The best that can be said for ID right now is that it hasn't caught on yet within the scientific community. I agree with you that motives should be left out of the discussion, but the fact of the matter is that very few in the creation/evolution debate who are on the side of science have any bone to pick with religion. For most it just isn't something they think much about. It is only when creationists muster a threat to science or science education that they become resentful, and usually not of religion, but of those working under its influence. If you would prefer that people not mention religious motivations, then you have to refrain from saying things like, "Maybe if they look harder they'll find evidence of the creator."--PercyEdited by Percy, : Typo.

Science is tentative. It never declares it has the once and for all final truth. Science seeks what is most likely true about the universe we live in through empirical research, but it is tentative and will always change in light of new evidence and/or improved insights.What IDists must do to influence accepted scientific theories is take their evidence and arguments to scientists, not the general public as Dembski does. If scientists find his evidence and arguments persuasive then theory will change. I mean scientists are not arguing about whether a designer is responsible for the diversity of life on earth. Science class teaches science, and therefore science class would be the proper venue to teach about scientific controversies, as opposed to history or English class. But there's no controversy within science pitting those who accept evolution against those who accept design. There's only a social controversy affecting the schools that pits conservative Christians who see science as a threat to faith against school boards who have been mandated with teaching currently accepted scientific views to students.It would be a lie to teach students that there is a scientific controversy over evolution and design, for the simple reason that no such controversy exists.It would equally be a lie to teach students ID as accepted science, for the simple reason that it is not.The ID insistence that new information cannot be created has to demonstrated through research before it can be taught as accepted science.--Percy