Uncommon Descent Contest Question 10: Provide the Code for Dawkins’ WEASEL Program

_{Denyse O'Leary
August 26, 2009

Darwinism

2}_{Categories
Darwinism}

Share: Facebook; Twitter/X; LinkedIn; Flipboard; Print; Email

Special invitation for Richard Dawkins – but any civil person is entitled to enter.

There’s been some discussion here and elsewhere whether the the recent IEEE article by Dembski and Marks correctly characterizes Richard Dawkins’ famous METHINKS IT IS LIKE A WEASEL program.

Does the program ratchet correct letters or does it let them vary? One is a partitioned or stair-step search, the other a more realistic evolutionary search. From The Blind Watchmaker, where Dawkins describes the program, its performance suggests that it could be either of these options (though he doesn’t say).

On the other hand, from a (video-run of the program , go to 6:15), it seems to be the latter.

It’s easy enough to settle this question: Make the code for the program public. Perhaps Richard Dawkins himself or his friends at RichardDawkins.net can finally provide this code (apparently a program written in BASIC).

The prize is a copy of either Stephen Meyer’s new Signature in the Cell or Richard Dawkins’ soon-to-be-out The Greatest Show on Earth.

Should the winner choose the latter, I will ask Dawkins’s publicist to mail the copy. Given that at his site, he calls himself “the most formidable intellect in public discourse,” I would assume that if he signs the copy, it will be worth millions.

But wait. Let’s see that code first.

Comments

DeLurker, 1- In TBW the weasel program never shows a reversal of charcters. Not one reversal is shown. 2- People reading the book would look up the word "cumulative" because "cumulative selection" is new (was new at the time). 3- Putting the two together- the definition of "cumulative", Dawkins description of slight improvements and the fact the illustration doesn't show any reversals, the inference is clear. Cumulative selection is a ratcheting process. If it wasn't he should have been explicit in saying exactly how it works. However had he been honest his point would have been lost.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:48 AM
6
06
48
AM
PDT}

PS: DL: “cumulative: “Increasing or enlarging by successive addition.”kairosfocus_{August 27, 2009
August
08
Aug
27
27
2009
06:45 AM
6
06
45
AM
PDT}

Joseph#100
I read the paper. Just because they do not explicitly say there is a population of size X and Y selection, doesn’t mean they didn’t use them both. Do you have the code they used? If you don’t then you just don’t know.
They describe a single string, not a population. They mutate every incorrect letter, rather than creating multiple daughter strings and selecting the most fit. No population, no selection. How can you read it otherwise?DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:44 AM
6
06
44
AM
PDT}

KF, Had Dawkins printed outputs that showed character reversals he would have screwed up many readers. IOW people would have noticed the contradiction.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:44 AM
6
06
44
AM
PDT}

DL: You are simply asserting that the world is as you wish it to be, not as it is. Denial is not a river in Egypt. Onlookers, Observe CRD again, as already excerpted -- and recall the showcased runs show evident latching and ratcheting to go with these remarks: ___________________ >> It . . . begins by choosing a random sequence of 28 letters ... it duplicates it repeatedly, but with a certain chance of random error – 'mutation' – in the copying. The computer examines the mutant nonsense phrases, the 'progeny' of the original phrase, and chooses the one which, however slightly, most resembles the target phrase, METHINKS IT IS LIKE A WEASEL . . . . What matters is the difference between the time taken by cumulative selection, and the time which the same computer, working flat out at the same rate, would take to reach the target phrase if it were forced to use the other procedure of single-step selection: about a million million million million million years. This is more than a million million million times as long as the universe has so far existed . . . . >> ______________________ Compare the plain vanilla defn of "cumulative: "Increasing or enlarging by successive addition." And you can refer above to the step by step discussion at 66. (One of the effects of objection waves, however irrelevant or distractive, is to bury substantial comments so that they are easily overlooked.) QED. Good day GEM of TKIkairosfocus_{August 27, 2009
August
08
Aug
27
27
2009
06:43 AM
6
06
43
AM
PDT}

Joseph#97
However reading TBW and looking at the weasel illustration, a ratcheting process is easily inferred.
You keep claiming this, but you have yet to support your claim with reference to Dawkins' description of the algorithm in The Blind Watchmaker. The text is clear -- mutation is independent of fitness. There is no ratcheting.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:42 AM
6
06
42
AM
PDT}

I read the paper. Just because they do not explicitly say there is a population of size X and Y selection, doesn't mean they didn't use them both. Do you have the code they used? If you don't then you just don't know.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:41 AM
6
06
41
AM
PDT}

Joseph#91
Indium:
D+M don´t use a population, they don't use selection.
You have said that before but never substantiated that claim. IOW that appears to be all in your head. Ya see if they referenced TBW, which they did, then it is a safe bet that they used the same stuff- population and selection- that Dawkins used.
Read the Dembski and Marks paper, section E, Partitioned Search:
Two of the letters {E,S} are in the correct position. They are shown in a bold font. In partitioned search, our search for these letters is finished. For the incorrect letters, we select 26 new letters and obtain OOT*DENGISEDESEHT*ERA*NETSIL. (21) Five new letters are found, bringing the cumulatie tally of discovered characters to {T,S,E,*,E,S,L} All seven characters are ratcheted into place. The 19 new letters are chosen, and the process is repeated until the entire target phrase is found.
No population, no selection. Indium's description is correct.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:39 AM
6
06
39
AM
PDT}

Joseph, the only case of EIL providing an algorithm for Weasel is Atom's adjustable weasel. (In the paper they are simply analysing the dynamics of ratcheted Weasel type programs; applicable to any latched version, implicit or explicit. But some strawman factories are running 24/7 at UD these days.) When they provide such a case, we can observe that EIL provides not only source code but text boxes for you to put in your preferred pop size and mut rate! Anybody prepared to bet that they did not do a LOT of simulation before sticking with what they know months ago was a controversial paragraph? If someone is, I have some prime, hot -- hot, hot, hot! -- real estate on Chances Peak Montserrat to sell for a great price . . . GEM of TKIkairosfocus_{August 27, 2009
August
08
Aug
27
27
2009
06:36 AM
6
06
36
AM
PDT}

DELurker:
Since there is no explicit code for latching in Dawkins’ Weasel algorithm, the IEEE paper by Dembski and Marks misrepresents it.
Except that Marks/Dembski never stated the program enlists explicit code for latching. However reading TBW and looking at the weasel illustration, a ratcheting process is easily inferred.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:34 AM
6
06
34
AM
PDT}

The usage by BillB may be standard but is also EXPLAINED. Dawkins did not explain cumulative as a process that could also lose what it had. Ya see saying that would noyt have helped his case. But I understand your position needs to redefine things on the fly. And I understand that commonly accepted definitions do not mean anything to you. But then again that is why your position is nonsense.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:32 AM
6
06
32
AM
PDT}

Joseph#88
There doesn’t have to be any explicit code for latching. Latching/ ratcheting occurs as a matter of course given a small enough mutation rate and a large enough population size.
Fantastic! We're in agreement! Since there is no explicit code for latching in Dawkins' Weasel algorithm, the IEEE paper by Dembski and Marks misrepresents it.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:32 AM
6
06
32
AM
PDT}

kairosfocus#87
DL, I am focussing on the showcased runs c. 1986 [as you cirted but seem to have missed the sifgnificancfe of in 41], which is what is to be explained as factual data.
That is obviously what you are focusing on, to the exclusion of the clear description of the algorithm that contradicts your claims. Look at the description, not the small amount of sample output.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:29 AM
6
06
29
AM
PDT}

DELurker, I know the strings mutate. That was never the question. Latching/ ratcheting occurs as a matter of course given a small enough mutation rate and a large enough population size. What part of that don’t you understand?Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:28 AM
6
06
28
AM
PDT}

Indium: Red herring. the facts in evidence to be explained are the showcased runs of champions c 1986. These on good empirical reasoning, show latching, and we have provided two mechanisms that can account for it. FYI, the champions emerge not when we have mutated a seed 50 or so times, but when we have filtered for closest to target. Thus mut rate, pop size AND filter affect the production ocf champions. After the ingredients are mixed, we see that under certain circumstances, already correct letters will be strongly preserved -- and that under others, they are most emphatically not. So, implicit latching is not a nullity. GEM of TKIkairosfocus_{August 27, 2009
August
08
Aug
27
27
2009
06:28 AM
6
06
28
AM
PDT}

indium:
D+M don´t use a population, they don´t use selection.
You have said that before but never substantiated that claim. IOW that appears to be all in your head. Ya see if they referenced TBW, which they did, then it is a safe bet that they used the same stuff- population and selection- that Dawkins used.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:27 AM
6
06
27
AM
PDT}

Joseph#86
IOW there isn’t anything in TBW that supports what you are saying. Nothing- not one word. If there were you would have posted it by now.
I have, as have others, several times. Again, in Dawkins' own words:
It now 'breeds from' this random phrase. It duplicates it repeatedly, but with a certain chance of random error - 'mutation' - in the copying.
The algorithm clearly specifies mutation without respect to the target string. That is the obvious reading of the text and the most biologically realistic mechanism.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:27 AM
6
06
27
AM
PDT}

Joseph#85
Without qualifying the word there is no reason to think that is how he used it. IOW if he is using a word in a way it is not generally accepted then it is up to him to qualify it.
The usage described by BillB is perfectly standard. It is your restricted definition that is not. Hundreds of programmers around the world have implemented the Weasel algorithm in a wide variety of languages, based solely on the material in The Blind Watchmaker. Based on that, it appears that yours is the idiosyncratic interpretation, unsupported by the actual text.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:24 AM
6
06
24
AM
PDT}

DeLurker, There doesn't have to be any explicit code for latching. Latching/ ratcheting occurs as a matter of course given a small enough mutation rate and a large enough population size. What part of that don't you understand?Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:23 AM
6
06
23
AM
PDT}

PS: DL, I am focussing on the showcased runs c. 1986 [as you cirted but seem to have missed the sifgnificancfe of in 41], which is what is to be explained as factual data. That data contains 200+ letters whose values could vary from correct a priori, but do not. (If you look up the demonstrated runs starting here, you will see cases where such reversions with detuned parameters are possible, and how long they tend to persist. If reversions were at all common in the relevant runs, reversions should have been easily observable in a sample of 200+ characters: taking Np = 0.1, and N = 200, p = 5 * 10^-4.)kairosfocus_{August 27, 2009
August
08
Aug
27
27
2009
06:22 AM
6
06
22
AM
PDT}

DeLurker, You don't get to tell me what I am doing. Also I am using the word "cumulative" how it is defined- the standard and accepted definition. If Dawkins is using it any differently than the standard and accepted definition it is up to him to explain that. As I said if Dawkins wanted people to know the reality behind evolution he would not have used the word "cumulative" and he wouldn't have used a program in which the output is always equal to or greater than the input. IOW there isn't anything in TBW that supports what you are saying. Nothing- not one word. If there were you would have posted it by now.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:22 AM
6
06
22
AM
PDT}

BillB:
Or he could have use cumulative, as in cumulative height, cumulative wealth…
Without qualifying the word there is no reason to think that is how he used it. IOW if he is using a word in a way it is not generally accepted then it is up to him to qualify it. He didn't so there is no reason to think he did.
If he wanted to describe a letter locking mechanism he could have just said that once a letter is found the search for it is over.
How he describes the difference between a random search and cumulative selection, especially given the weasel illustration, that is teh only inference to make. IOW there wasn't any need to come right out and say it. And taken in context what others say about mutations:
A mutation is a permanent change in the DNA sequence of a gene.
There just isn't any other inference to be had.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:17 AM
6
06
17
AM
PDT}

kairosfocus#81
The runs as shown together with the remarks on cumulative selection make latching and ratcheting a very reasonable and straightforward understanding of CRD’s words.
Here are Dawkins' actual words, again:
It now 'breeds from' this random phrase. It duplicates it repeatedly, but with a certain chance of random error - 'mutation' - in the copying.
Note that there is no suggestion of mutating only incorrect letters. It is a very simple statement. Dawkins continues:
The computer examines the mutant nonsense phrases, the 'progeny' of the original phrase, and chooses the one which, however slightly, most resembles the target phrase, METHINKS IT IS LIKE A WEASEL.
The selection mechanism is the only step of the algorithm that refers to the target phrase. The mutation mechanism does not. There is no explicit preservation of correct letters. It is very clear from a straightforward reading of Dawkins' own words that Dembski and Marks have misrepresented the Weasel algorithm in their paper.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:17 AM
6
06
17
AM
PDT}

kf, I will happily admit that the algorithm is good at preserving correct letters most of the time. Call that implicit latching if you must, but this is just a distraction on your part. Since it is an artificial algorithm this is artificial slection. Or natural selection if you call CPU, harddisk and RAM the natural environment of algorithms! ;-) Since you admit, that the latching is implicit the answer to the original question of the thread "Does the program let correct letters vary?" is YES! Most of the time these variations do not show up because of the selection routine, but they are still mutated. Now, this is sorted then. The next question then is if D+M misrepresent Dawkins. Since they use explicit latching instead of implicit, since they use no population and no selection the answer is also "YES!". I suppose that the active info concept can be applied to the real Weasel, and Random Mutation / Selection are indeed discussed in the paper! But discussing Weasel using a different algorithm where the only similarity is that it also preserves correct letters most of the time is a blatant misrepresentation. I ask again: Which algorithm from Atoms suite corresponds to Weasel and which one corresponds tpo the Partitioned Search discussed by D+M?Indium_{August 27, 2009
August
08
Aug
27
27
2009
06:13 AM
6
06
13
AM
PDT}

Joseph#70
I am addressing the full description made by Dawkins and illustrated by Dawkins in “The Blind Watchmaker”.
No, you are continuing to focus solely on the word "cumulative" and are applying your own, very restricted, interpretation of what that word means. When climbing a hill, one cumulatively increases one's altitude, even though an individual step may decrease it temporarily. With a genetic algorithm, one cumulatively increases fitness, even though one generation may be less fit than the previous generation due to the vagaries of the random number generator. I posted the full description of the Weasel algorithm in #75. Please show how that could possibly be interpreted, in context, as supporting the idea that letters are protected from mutation once correct.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:12 AM
6
06
12
AM
PDT}

DL: Simply read the excerpt above and the runs as shown. The runs as shown together with the remarks on cumulative selection make latching and ratcheting a very reasonable and straightforward understanding of CRD's words. Focus on the 1986 printoffs and the meaning of CUMULATIVE in its context. the general rule of hermeneutics is to take the plain meaning in context as the meaning, save where it makes no sense. And, plainly the obvious meaning -- cumulative, ratcheting, latching action as Weasel moves to target -- makes a lot of sense. GEM of TKI.kairosfocus_{August 27, 2009
August
08
Aug
27
27
2009
06:12 AM
6
06
12
AM
PDT}

DeLurker, The weasel program is only concerned with the output. Nothing else is relevant.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:11 AM
6
06
11
AM
PDT}

Also the program is evidence for ID as ID can be reduced to nothing more than a targeted search- along with the resources required to reach that target. yakky d:
This makes no sense.
It makes perfect sense if you understand ID. So what do you know of ID and how did you come to that?Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:10 AM
6
06
10
AM
PDT}

DeLurker, I am addressing the full description made by Dawkins and illustrated by Dawkins in "The Blind Watchmaker". Again if he didn't mean "cumulative" he should not have used that word. And his illustration of the process using WEASEL there is only ONE inference to be made- it is a ratcheting process. OTOH you have not provided anything from the book that would change nor challenge that inference.Joseph_{August 27, 2009
August
08
Aug
27
27
2009
06:08 AM
6
06
08
AM
PDT}

kairosfocus#69 You have apparently missed my explanation in #41 of how you are doing your math incorrectly. As I point out in that post, you are considering only approximately 0.047 percent of all the strings generated by Dawkins' Weasel algorithm. It is not surprising that you don't see a letter reversion. You need to read the entire description of the algorithm instead of referring to just a few sample lines of output.DeLurker_{August 27, 2009
August
08
Aug
27
27
2009
06:06 AM
6
06
06
AM
PDT}

Prev 1 … 8 9 10 11 12 13 Next