| | Merlin wrote: EXHIBIT A: For a given sized packet, more randomness in it implies less information. However, this does not imply that for a given packet, a small random addition or change to it decreases the information. It may increase it (or stay the same).
Then Grammarian responded:
EXHIBIT B
Sure. I fully admit, as does any statistician, that there is a non-zero, calculable probability monkeys making random additions of text to a blank page of paper might increase the information density on the page by typing intelligible sentences, and that these sentences might randomly be assembled into paragraphs, and the paragraphs randomly assembled into chapters, and the chapters randomly assembled into sequences, and the sequences into a full novel like "Atlas Shrugged." If you believe that it actually DID happen that way, then I have bridge in Brooklyn I'd like to sell you.
Then Merlin continues:
Well, you contradict yourself. First you grant its truth, then you say I make a contradiction.
Uhh, I didn't grant "its" truth. I posted a WHOLE PARAGRAPH EXPLAINING PRECISELY WHAT I MEANT, WHICH YOU DISHONESTLY OMITTED AFTER YOUR ELLIPSES (". . . "); TO WIT:
In post #92 Grammarian responds:
Sure. I fully admit, as does any statistician . . . [NB: Where the hell is the rest of my sentence, Merlin, ehhhhh???????]
There is no contradiction in my sentences.
Merlin's 1st sentence: For a given sized packet, more randomness in it implies less information. Merlin's 2nd sentence: However, this does not imply that for a given packet, a small random addition or change to it decreases the information. Randomness = [A] Information = [B]
1st sentence: More [A] implies less [B]. 2nd sentence: A small addition of [A] does not imply a decrease in [B].
Sorry, but that's what you wrote. Sentence 2 disaffirms sentence 1 by means of negating the copula and that is the essence of contradiction.
Are you going to quibble over distinctions-without-a-difference such as those between "more" and "a small addition" or "less" and "decrease in"?
Perhaps you meant to write, "Though big amounts of randomness will no doubt decrease the information of a system, a very small random addition need not." Is this what you meant?
If so, then it appears you're trying to argue that genomic information could have increased in the distant past as a result of very small random changes. And (for the 20th time) here's my reply:
1. The smallest possible random change to a genome is a single nucleotide substitution -- an error in DNA replication. After many decades of inducing such substitutions in fruit flies, nothing really interesting has happened; i.e., no new species. No flies-turned-into-wasps; no flies-turned-into-mosquitoes. Only a few monstrosities (flies with legs sticking out of their heads where their antennae should be.)
2. Many diseases in humans correlate with small, random changes in their DNA: Sickle-Cell Anemia, Cystic Fibrosis -- probably cancer, who knows? Show any correlations between random DNA mutations and positive additional traits to the human genome -- "trading up" -- which doesn't have some sort of protein/tissue/organ degradation associated with it. I don't know of any.
3. Other aspects of Darwinism say you're wrong. The field of population genetics demonstrates that very small mutations that try to "seep" their way into a population get sloughed off; they get rejected by the population as a whole, and the population maintains its original integrity (meaning: no increase in information in their genomes). Latest press release from the molecular biology department at University of Chicago claims that there is now direct experimental proof of this (rather than just statistical inference). They claim that they have shown (to their own surprise) that small mutations, occurring gradually, will be rejected. Conversely, lots and lots of mutations happening quickly are likely to be accepted. This is the opposite of what a Darwinist would expect. Darwinists don't like "catastrophism." They like "gradualism." It's bad enough to calculate the odds of ONE specific mutation happening at just the right time, in just the right order, at just the right place, with just the right enzyme to help things along, with just the right environment to ensure it gets selected; now to assume that MANY such mutations must happen in a very short amount of time if we are to get evolution to work at all, is to inflate already impossible odds.
So I will try again with an analogy.
I thought you didn't like analogy?
Suppose we have a sequence of 100 coin flips.
Already you've mischaracterized the problem and provided a faulty analogy. Don't give a hand-waving argument and say "suppose we have just any old sequence of 100 coin flips." Damnit, specify the outcome you're looking for! Give us in advance the sort of target you're trying to hit.
I no longer like you Merlin, but I feel sorry for you, so I'll do the math for you. 100 tosses are too many for an example. Let's take 10. Here's the scenario:
Somewhere in the universe there's a little known protein called "Merlinine" which is indispensible for the vision of space aliens. The protein consists of an impossibly short chain of amino acids, but that's our example. The sequence is this:
alanine+valine+leucine+proline+leucine+glycine+serine+valine+tyrosine+cysteine
That's the specified sequence. That's what we need. If we don't have exactly THAT, the aliens are blind. OK? What are the odds of forming that specific sequence -- not just any sequence we feel like, but THAT one -- by chance?
The math is the same as in coin tosses, except instead of 2 choices (heads/tails) we have 20 choices (the total number of amino acids, which forms a set of twenty discrete elements).
We start out with a bag of amino acids, we shake it up, and we reach into the bag to pull out the first one.
1. Odds of pulling out "alanine" are 1/20. [we pull out a 2nd] 2. Odds of pulling out "valine" are 1/20. [we pull out a 3rd] 3. Odds of pulling out "leucine" are 1/20. [we pull out a 4th] 4. Odds of pulling out "proline" are 1/20. [we pull out a 5th] 5. Odds of pulling out "leucine" are 1/20. [we pull out a 6th] 6. Odds of pulling out "glycine" are 1/20. [we pull out a 7th] 7. Odds of pulling out "serine" are 1/20. [we pull out an 8th] 8. Odds of pulling out "valine" are 1/20. [we pull out a 9th] 9. Odds of pulling out "tyrosine" are 1/20 [we pull out the 10th] 10. Odds of pulling out "cysteine" are 1/20.
Total odds = 1/20^10 of hitting THAT target (remember the archers? That specific amino acid sequence above is the bullseye we're trying to hit).
Whatever the exact sequence is, we can look it from three time perspectives - before (T0),
This is not a physics experiment in kinematics; there is no "T0" and time has nothing at all to do with this (except for the fact that we can take as much time or as little time as we wish to reach into the bag and pull something out).
sometime during such as after 95 flips (T1), and after 100 flips (T2). Accept at least for the moment that probability is an ex ante concept. At T0 the probability of that sequence is .5^100.
At all points along the selection process, the probability is 1/20^10 to hit the target, once we've decided that it's the bullseye; once we've specified it.
At T1 the probability of that sequence is .5^5. The first 95 are now given, and what the probability was at T0 is now irrelevant.
No. We repeat the experiment with a slight change: what are the odds of randomly creating an amino acid chain 10 residues long of the following specified structure:
TARGET = alanine+valine+leucine+proline+leucine+glycine+serine+valine+tyrosine+cysteine
1. Odds of pulling out "alanine" are 1/20. [we pull out a 2nd] 2. Odds of pulling out "valine" are 1/20. [we pull out a 3rd] 3. Odds of pulling out "leucine" are 1/20. [we pull out a 4th] 4. Odds of pulling out "proline" are 1/20. [we pull out a 5th] 5. Odds of pulling out "leucine" are 1/20. [we pull out a 6th] 6. Odds of pulling out "glycine" are 1/20. [we pull out a 7th] 7. Odds of pulling out "serine" are 1/20. [we pull out an 8th] 8. Odds of pulling out "valine" are 1/20. [we pull out a 9th]
(Here we break for lunch; we listen to some music; we read a book; we do some chores; we run some errands; we brush our teeth (and floss). Then we brew some coffee, and heigh-ho! it's back to work for us. We assume, of course, that the above sequence is now a given.):
9. Odds of pulling out "tyrosine" are 1/20 [we pull out the 10th] 10. Odds of pulling out "cysteine" are 1/20.
Probability of hitting that specific target, even with a lunch break = 1/20^10.
The fact that you've picked 8 of the 10 correctly in no way changes the odds, no matter where you decide to pause for lunch, or how long the lunch break is (analogy not parody). This is probability not physics. Time doesn't enter into it except if we become interested in how long it might take someone actually to perform this selection process.
Consider that in the real world, most proteins have amino acid chains of 300, 500, or more; not 10 as in the example above. Then there's the problem of the odds of creating all the different sorts of proteins, each one with a different sequence. Then there's the problem of creating all the different sorts of enzymes and regulation systems that integrate proteins together. Then there's the problem of why a protein would even be useful for anything, unless there was an organism and DNA to make use of it and to copy it. Darwinist theory has too many proteins and too many DNA sequencing it has to calculate odds for, and not enough time since the Big Bang to have searched through the "search space" of possibilities in a plausible way. If you tell me hey, maybe a bunch of amino acids just got lucky and formed first time around, then you endorse a miraculous view of the universe. Nothing beats odds of numbers like 1/10^65.
(Except, of course, intelligence, which is NOT governed by chance itself, and which routinely and easily beats odds like that.)
At T2 probability is inapplicable; all 100 are a given. You repeatedly take a T0 perspective, and I submit that this is improper applied to evolution.
You can assume anything you want. Why not assume that that all the amino acid sequences were already there in the distant past and all the requisite proteins were already formed, just waiting to be made use of by biological organisms that hadn't even appeared on the scene yet.
Grammarian wrote:
In order to get a computer to reproduce that specific pattern, you have to instruct it with all the original characters of the pattern: 1. print H 2. print H 3. print T 4. print H 5. print T 6 print T etc. etc. until the entire original pattern is given.
With the random (more exactly, pseudo-random) number generators in programming languages I have worked with, this is clearly false. The random number generator has a "seed" which feeds off the computer's clock by default. If the user sets the same seed for two different computer runs, the same sequence of random numbers is generated for each. Give 10 identical "seeds" to 10 computers; have them produce 10 identical random number strings of 100 digits.
Now, shut off the computers, and take out a piece of paper and pencil (we're writing BASIC) and create a short, short, short, single-step algorithm that compresses any one of the 10 identical random-number strings. You must be able to feed the BASIC program back into a computer and have the algorithm generate an 11th identical random number string.
(Your algorithm won't be short, it won't be compressed. It'll simply be an enunciation of the same 100 digits that were in the one of the random number strings. Your algorithm will have the same length as the target sequence it's trying to produce.)
I have no more interest in discussing ID with you.
Boo, hoo.
|
|