Dawkins’ WEASEL: Proximity Search With or Without Locking?

_{William Dembski
March 16, 2009

Darwinism, Evolution, Informatics

3}_{Categories
Darwinism
Evolution
Informatics}

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

On pp. 47-48 of THE BLIND WATCHMAKER, Richard Dawkins gives two runs of his WEASEL program (note that there were typos in both initial seeds — one had 27 characters, the other 29 whereas they should have 28; I’ve corrected that). Here are the two runs using the Courier typeface, which assigns equal width to each character (spaces are represented by asterisks):

WDL*MNLT*DTJBKWIRZREZLMQCO*P
WDLTMNLT*DTJBSWIRZREZLMQCO*P
MDLDMNLS*ITJISWHRZREZ*MECS*P
MELDINLS*IT*ISWPRKE*Z*WECSEL
METHINGS*IT*ISWLIKE*B*WECSEL
METHINKS*IT*IS*LIKE*I*WEASEL
METHINKS*IT*IS*LIKE*A*WEASEL

Y*YVMQKZPFJXWVHGLAWFVCHQXYPY
Y*YVMQKSPFTXWSHLIKEFV*HQYSPY
YETHINKSPITXISHLIKEFA*WQYSEY
METHINKS*IT*ISSLIKE*A*WEFSEY
METHINKS*IT*ISBLIKE*A*WEASES
METHINKS*IT*ISJLIKE*A*WEASEO
METHINKS*IT*IS*LIKE*A*WEASEP
METHINKS*IT*IS*LIKE*A*WEASEL

These runs are incomplete. The first, according to Dawkins, required 43 iterations to converge, the second 64 (Dawkins omitted the other iterates to save space).

As you can see, by using the Courier font, one can read up from the target sequence METHINKS*IT*IS*LIKE*A*WEASEL, as it were column by column, over each letter of the target sequence. From this it’s clear that once the right letter in the target sequence is latched on to, it locks on and never changes. In other words, in these examples of Dawkins’ WEASEL program as given in his book THE BLIND WATCHMAKER, it never happens (as far as we can tell) that some intermediate sequences achieves the corresponding letter in the target sequence, then loses it, and in the end regains it.

Thus, since Dawkins does not make explicit in THE BLIND WATCHMAKER just how his algorithm works, it is natural to conclude that it is a proximity search with locking (i.e., it locks on characters in the target sequence and never lets go).

Interestingly, when Dawkins did his 1987 BBC Horizons takeoff on his book, he ran the program in front of the film camera:

www.youtube.com/watch?v=5sUQIpFajsg (go to 6:15)

There you see that his WEASEL program does a proximity search without locking (letters in the target sequence appear, disappear, and then reappear).

That leads one to wonder whether the WEASEL program, as Dawkins had programmed and described it in his book, is the same as in the BBC Horizons documentary.

In any case, our chief programmer at the Evolutionary Informatics Lab (www.evoinfo.org) is expanding our WEASEL WARE software to model both these possibilities. Stay tuned.

Comments

Mr Kellogg: before I leave for the day. 1 --> you will see above that there is some fresh evidence, albeit of course hearsay, that Mr Dawkins states that he has not used explicit locking. Fine, this is not a courtroom: on charity, I have taken that. 2 --> the import of that is that as Joseph says, the program implicitly latches in the case in view. 3 --> That no "U" happens to appear is irrelevant to the observed fact that with 200+ out of 300+ letters in play, no correct letter reverts, regardless of want of functionality. (Indeed, the absence of a U is easily accounted for by the rapidity of convergence by which 2/3 of the available slots are effectively out of play. Wrong letters are being rather rapidly squeezed out by the force of the proximity rule, probably multiplied by a modest per letter "quasi-functional" mutation rate -- 5% [which is utterly huge for realistic mutations in bio-forms!] means 1 - 2 letters typically vary per population member. I use "quasi-functional" as in effect any mutation is allowed into the population to compete. the nearest tot he target -- regardless of want of overall linguistic funciton -- is promoted, which means that the trend is to converge to the target. (The circumstances would bias outcomes sharply towards those that have 1 - 2 functional additions, and do not derange the existing ones. These circumstances are utterly unrealistic, of course. Thus, implicit locking [and with suitable shift of parameters, quasi-locking] as I discussed at T3 in my previously linked 346 - 7 in the earlier thread.) 4 --> Thus, the targetted, proximity based search implicated by the behaviour and by Mr Dawkins' direct statements as cited, is irrelevant to the capacity of a BLIND watchmaker such as RV + NS. Hoyle's FSCI challenge still stands unmet. 5 --> Similarly, as a third level matter, in the 1987 video, the program's observed behaviour, with fairly frequent flickbacks, is significantly different from the case of the 1986 printoffs, which we have good reason to see as representative of the overall pattern of the program. 6 --> that suggests at minimum, that probabilities and populations have been shifted, which is a material difference to the program. Such an adjustment does not obviate the force of the point at 4 just above. GEM of TKIkairosfocus_{March 17, 2009
March
03
Mar
17
17
2009
06:09 AM
6
06
09
AM
PDT}

The message was derived by the messenger. And it was derived just to prove the messenger can act like a little baby. My qualifications: 30 years of working with computers, writing and editing codes. Also, as evidenced by responses to this question, I understand perfectly what is going on with this "weasel" program.Joseph_{March 17, 2009
March
03
Mar
17
17
2009
05:57 AM
5
05
57
AM
PDT}

Don't like the message? Dismiss the messenger. Your qualifications are precisely what, Joseph? Please be specific.David Kellogg_{March 17, 2009
March
03
Mar
17
17
2009
05:53 AM
5
05
53
AM
PDT}

David, No one cares what an English professor has to say about a computer program unless tat program was written by thta English professor. And if all you can do is act like a little child about it, then why bother?Joseph_{March 17, 2009
March
03
Mar
17
17
2009
05:48 AM
5
05
48
AM
PDT}

Joseph, how is the allegation of locking any more evidence-based than my allegation about the letter U?David Kellogg_{March 17, 2009
March
03
Mar
17
17
2009
05:41 AM
5
05
41
AM
PDT}

David, No one cares what an English professor has to say about computer programming. No one.Joseph_{March 17, 2009
March
03
Mar
17
17
2009
05:38 AM
5
05
38
AM
PDT}

I have noticed that in the WEASEL runs printed above, there is no instance of the letter U. Nowhere, not in almost 450 characters and spaces. From that observation, I propose that the WEASEL program will never change another letter to a U. The program must be designed to avoid Us. Silly, you say? A pointless distraction? Dawkins mentioned nothing about it? I. Don't. Care. I don't see a U in these partial prints of runs. Therefore, the program can allow no Us at all. But, you say, there are Us in the 1987 video. Ahh. 1987. That's a different matter. If a U shows up in 1987, it must be a different program.David Kellogg_{March 17, 2009
March
03
Mar
17
17
2009
05:32 AM
5
05
32
AM
PDT}

Challenge to evolutionists: Can you salvage Dawkins' WEASEL words (or similar attempts) from the War Games fallacy? At a dramatic point in the movie War Games, a computer endeavors to crack a password code. After working some time at blinding speed, it manages to solve one character of the password. Some time later, it has solved another character, and so on. It does not take much reflection to realize this is nonsense. If the password is unknown, it cannot possibly determine it one character at a time, preserving and accumulating partial successes until it reaches a working password. A password works as a whole, or it doesn't work at all. It is all or nothing. How are these attempts by Dawkins or others (e.g. Zachriel) at letter and word manipulation not equally nonsensical as illustrations for biological evolution? For instance, in the second example series of the opening post, the sequence "LIKE" appears very early and is preserved, even before it is functioning as a word, let alone a word in a functional sentence. The same type of question would apply to "the Zachriel algorithm (which is actually a more refined and thus more relevant version of the weasel program)" according to JT's description. What is the biologically equivalent justification for preferentially preserving partial progress toward function even before function is reached (e.g. before we reach a functional sentence)?ericB_{March 17, 2009
March
03
Mar
17
17
2009
04:58 AM
4
04
58
AM
PDT}

There isn't any such thing as "cumulative selection" in nature. Cumulative selection requires a target. In nature the "target" does not (yet) exist. And as I said earlier the "locking" of the letters is a product of probability. That is given a small enough mutation rate and a large enough sampling size (number of "offspring" per generation), letters will appear to locked in place when only viewing the chosen one- that which is closest to the target. That is just the way it is. Of course that would change if Dawkins' original program has the parent copied unaltered in each generation, as well as others that can be altered.Joseph_{March 17, 2009
March
03
Mar
17
17
2009
04:33 AM
4
04
33
AM
PDT}

6 --> What best accounts for it? Candidate no 1 is partitioned search with explicit latching [my model T2 from previous]. No 2 is an implicit quasi-latching [my model T3]. However, it is to be noted that Apollos showed by code example that explicitly latched programs can show quasi latching if that is written in. 7 --> So, only code will be explicitly decisive, if demonstrative proof is required; which obviously will not be forthcoming. But, we can proceed on inference to best explanation. Ont he first look, the most credible explanation is that Weasel circa 1986 latched, but was improved by 1987 to not latch. 8 --> On charity, we will however take Mr Dawkins at his reported word, that he did not ever EXPLICITLY latch Weasel. So, the particular balance of parameters used, implicitly latched in the 1986 cases, and quasi-latched with fairly frequent flickbacks in the 1987 video. (The suggestion above that parameter tweaking for visuality advantage is an interesting suggested context.) 9 --> On number of generations to target, we have explicit statements in the 1986 cases that the program took 40+ and 60+. By sight, especially the rate at which flickbacks reverted in the topline display of what appears to be current generation [the bottom one that runs too fast to follow by eye is obviously current mutant), we are running at a generation per ~ 1/10 second or so. So, we probably have a much higher number of generations in 1987, which runs a lot longer than 40 - 100 gens at one per ~ 1/10 sec. [A generation per 1/10 second is also consistent with the report that the Pascal version ran in 11 seconds, which would run out in ~ 100 generations at that rate.] ______________ CONCLUSION: Weasel does not address the Hoyle challenge, on getting the increments of functionality required to be on the shores odf Dr Moreau's Island Improbable, for which hill-climbing algorithms can in principle climb, step by step up the easy slope side to peak performance at the crest of Mt Improbable. Instead, it rewards closeness to target of non-functional configs of nonsense phrases. Thus, it begs the question Hoyle raised and acts as a distractive strawman from the issue of FSCI/CSI; that is, it is one of the unfortunately long line of fundamentally misleading icons of evolution. Now, charity per Dawkins' reported statement, we can accept that the Weasel program does not explicitly latch, but instead creates output latching as seen in the 1986 printoff samples, by parameter balancing etc, i.e. quasi-latching with significant incidence of flickbacks in at least some cases. [Algor model T3 not T2 from 346 - 7 in the previous thread]. All said and done, that leaves the bottomline issue clear: Weasel does not answer to the FSCI challenge,as it rewards non-function on proximity to target. And that we canboth see and can understand from Mr Dawkins' direct published statements. Weasel -- sadly -- in the end only manages to be a distractive, only rhetorically persuasive irrelevancy. On substance, we still have only one credible explanation for CSI/FSCI: design. GEM of TKIkairosfocus_{March 17, 2009
March
03
Mar
17
17
2009
03:34 AM
3
03
34
AM
PDT}

H'mm: Seems I am late to the party. A few observations, but first, a clip 9courtesy Wiki) on what Dawkins said he was trying to do, and did:
[Citation from BW ch 3 by Dawkins, on Weasel:] We again use our computer monkey [NB: to select a specific target functional case form 28 27 state elements identifies to 1 in "only"~ 1.2 * 10^40, well below the credible challenge to get FSCI for OOL and body plan level biodiversity, which is what Hoyle -- who Dawkins was trying to rebut -- had raised], but with a crucial difference in its program. It again begins by choosing a random sequence of 28 letters, just as before … 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 [but just such single step to gain functionality is actually what is necessary for natural selection to compare on differential functionality]: 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.
1 --> Let us highlight key admission no 1: 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. 2 --> Admisison no 2: This [the time the PC would take to select at random something on the order of 1 in 10^40] is more than a million million million times as long as the universe has so far existed. 3 --> So, to escape the implications of the long odds against arriving at functionality by chance [and remember, these odds are far, far, far better than those for observed bio-functionality information . . . ], Dawkins illustrates a purportedly BLIND watchmaker, by intelligently designed, targetted search that rewards detected proximity to a target of non-functional, nonsense phrases. 3 --> So, Weasel illustrates, not the power of a BLIND watchmaker, but the capacity of an intelligent designer to find a target by a partially random search process. "Improved" versions could doubtless go all the way to GA's that define a functionality parameter and seek to get it to peak. 4 --> But in neitehr case would there be any proper dismissal of the key insight: Weasel-style programs instantiate intelligent design in action, not he power of some alleged BLIND watchmaker. 5 --> Therefore, let this major point be settled in our minds, before we turn to a secondary issue: the observed lockstep of Weasel circa 1986, where letters in the output latch and we see that 200+ out of 300+ letters in the program's output, show an evident pattern of o/p latching with no counter instances. (Note, it is the absence of counter instances that led tot he inference, far and wide that the program in question is a letter- by- letter search that rewards "success" on a letter by letter basis, which is also a very plausible reading of Mr Dawkins' words. For,t he minimal step forward is obviously one letter's step forward.) [ . . . ]kairosfocus_{March 17, 2009
March
03
Mar
17
17
2009
03:34 AM
3
03
34
AM
PDT}

David Kellogg, If languages existed independent of intelligent agents, I could agree with you. People are not just a vehicle, like a host, and language a parasite that is along for the ride. Without people using the language, and changing it, there would be no "change" of language. You may as well say that art and medicine have "evolved." But we know that all of these things are really perpetually created and designed.Clive Hayden_{March 16, 2009
March
03
Mar
16
16
2009
09:13 PM
9
09
13
PM
PDT}

[41]: I can certainly see the potential for teleological arguments concerning the probability of getting a fitness landscape so structured. In fact, any such argument would reduce to the observation that if f(x) = y, then f(x) cannot be more probable than y.JT_{March 16, 2009
March
03
Mar
16
16
2009
08:47 PM
8
08
47
PM
PDT}

Gil:
Why would a computer programmer search for something that he included in his code?
To illustrate cumulative selection. That's a no-brainer for anyone who has read the relevant section of TBW.
I’ll tell you the answer: So he can dupe people into thinking that dumb, random searches can produce interesting, innovative results. Which they can’t.
Apparently it's okay on this board to divine nefarious motives without offering a speck of evidence. And I don't know what you mean by "random searches", but genetic algorithms most certainly produce interesting and innovative results. Dawkins's algorithm, of course, produces no useful results, as the target is defined in the domain of the objective function. Useful genetic algorithms, on the other hand, find elements in the domain that meet desired criteria in the codomain.R0b_{March 16, 2009
March
03
Mar
16
16
2009
08:30 PM
8
08
30
PM
PDT}

ericB Darwinian evolution does not evaluate by anticipating what will be useful someday. Selection is ruthlessly oriented to the present competition So is the Zachriel scheme. If you randomly find a two letter word and from that randomly find a three letter word and from that randomly find a four letter word and eventually find a 10 letter word (and accomplish all this 200,000 times faster than could be accomplished via random generation of in tact 10 letter combinations), its because there's regularity in the fitness landscape. When you preserved that two letter word it wasn't with a mind to finding the 10 letter word you eventually hit upon by chance. But randomly selecting that two letter word (and then the three letter word and so on) did focus you in on a 10 letter word, but it wasn't by your own design. I.D. cannot require that the fitness landscape for evolution be random. If it is random, that has to be demonstrated. However, The fact that the human languages are structured to facilitate this type of random search (and also considering that human languages themselves evolved via a haphazard a process.) seems strongly suggestive as to the potential nature of biology. Clive Hayden: I don’t see the analogy between biological complexity and sentence complexity. The only similarity is that they’re complex. But it alludes to an implication that biology is approximating to something already in existence, such as a meaningful statement, already known. The statement "I see" has meaning on its own. It doesn't have meaning only on the basis of some other sentence it could potentially be a part of, but not yet in existence. But by combining it with other words gives the phrase increasing specificity (i.e. complexity) and increase its fitness for a specific function. "I see dead people". biological changes cannot be “read” and understood. It seems they are read and understood by reality itself. Consider an eye, but not the means by which it may or may not have emerged. Are you saying that reality cannot be allowed to determine how specific attributes of an eye's physical configuration dictate how the eye functions and how this function bestows on the eye its highly selective advantage. The issue isn't whether the Zachriel program is a perfect model of nature. Clearly however, its a vast improvent on the Weasel program, in considering the issue of intermediate functionality among many other things. Why would the Dawkins program be continually attacked for a deficiency in it that its author acknowledged from the beginning, and at the same time the Zachriel program be merely ignored.JT_{March 16, 2009
March
03
Mar
16
16
2009
08:05 PM
8
08
05
PM
PDT}

Gil, if it's so easily dismissed, why are people blathering on about partitioned search and latching and the like?David Kellogg_{March 16, 2009
March
03
Mar
16
16
2009
07:44 PM
7
07
44
PM
PDT}

The target search string is contained in the code. Therefore, no search is required. That's the end of that. Why would a computer programmer search for something that he included in his code? I'll tell you the answer: So he can dupe people into thinking that dumb, random searches can produce interesting, innovative results. Which they can't.GilDodgen_{March 16, 2009
March
03
Mar
16
16
2009
07:41 PM
7
07
41
PM
PDT}

I don't see the analogy between biological complexity and sentence complexity. The only similarity is that they're complex. But it alludes to an implication that biology is approximating to something already in existence, such as a meaningful statement, already known. No amount of biological changes, taken together, approximates to any guide or lexicon of meaning, equivalent to language. If a wing or a feather is produced, that is not similar to using parenthesis or a semicolon, or any other literary device--for a collection of biological changes cannot be "read" and understood. I think there are all sorts of problems with the analogy, and teleology is inherent in it. Sheer complexity doesn't equate to meaning, so there is no equivalent to increases in biological systems and meaning known prior to reading a sentence.Clive Hayden_{March 16, 2009
March
03
Mar
16
16
2009
07:18 PM
7
07
18
PM
PDT}

p.s. To complete the analogy, imagine someone trying to explain the existence of proteins unique to the bacterial flagellum by saying that those proteins are part of the landscape of bacterial flagellum proteins. Once mutation finds one of them, it can preserve it until the rest come along in similar fashion. Eventually a complete functioning flagellum can be constructed (after some trial and error combining the parts). This would be obvious nonsense. No one would seriously propose this. Yet when we substitute words and sentences, the audience can be taken in by the illusion. Such illusions work in part because, as users of English (or other languages), we take the language for granted. In its entirety, we are prepared to treat it as real and as natural as the physical landscape around us. We forget that without language translators (e.g. us) meaningful symbolic language could not exist.ericB_{March 16, 2009
March
03
Mar
16
16
2009
07:08 PM
7
07
08
PM
PDT}

Clive [20], I'm not sure how seriously you're posing this question. But all languages, including English, do evolve, one step at a time. All languages have developed by means strikingly akin to biological evolution: They mutate and acquire new function beyond the control of any individual or group. While sentences are designed, languages (with the exception of oddities like Esperanto) are wholly natural. Here is the first poem in English:
Nu we sculon herigean / heofonrices weard, meotodes meahte / ond his modgeþanc, weorc wuldorfæder, / swa he wundra gehwæs, ece drihten, / or onstealde. He ærest sceop / eorðan bearnum heofon to hrofe, / halig scyppend; þa middangeard / moncynnes weard, ece drihten, / æfter teode firum foldan, / frea ælmihtig.
(The slashes represent half-line marks in Old English poetry.) Stuff like this evolved into modern English. How? Reproduction, variation, selection. Rinse, repeat. When? Over time, one step at a time. Middle English -- after the Norman Conquest -- is a "transitional form" between Old English and modern English. The "body plan" of modern English stabilized with the introduction of print and the normalization of spelling that was codified -- but not caused -- by the introduction of dictionaries.David Kellogg_{March 16, 2009
March
03
Mar
16
16
2009
07:08 PM
7
07
08
PM
PDT}

Word and sentence juggling as "support" for the plausibility of biological evolution is confused and fatally flawed. There are multiple problems. To take one such problem as an illustration, Darwinian evolution does not evaluate by anticipating what will be useful someday. Selection is ruthlessly oriented to the present competition. However, if we imagine a world of functional sentences, the idea that isolated words can be preserved because they are part of English immediately breaks the analogy with biology. In English, we might mutate "of" to "off", knowing that both are part of the "landscape" of English -- i.e. subsets of the already existing fully functional language, even though by themselves they are not yet functioning as sentences. But that is not the position evolution is in. Biology has no already compiled "dictionary" to know what incomplete sentence parts might be useful to a fully functional future sentence. If you want to seriously attempt to perform some analogous thought experiment or computer simulation, you have to get rid of the idea that the non-functional parts (e.g. words in English) are predefined, knowable, recognizable -- and therefore preservable. The real function of such exercises has been to bolster hope that would normally be dashed against hard reality. Behe's The Edge of Evolution takes a serious step toward uncovering how hard that reality really is.ericB_{March 16, 2009
March
03
Mar
16
16
2009
06:39 PM
6
06
39
PM
PDT}

What I’m still not understanding is the analogy between when something has been, if you will, attained, or captured, such as biological viability, and how that’s analogous to sentences. ... What would this be? Is the analogy saying that “biological viability” (which I presume means just living, and maybe reproducing) is akin to a valid English sentence in some way? How would the analogy follow? Well language seems a very apt model for biological organisms. Languges are extremely complex and have themselves coincidentally evolved in a very haphazard incremental fashion. No one's saying that it is a perfect analogy, but what would be a better one? But if we're define fitness as an increase in biological functional complexity, the relevance of an increase in meaningful sentence or word length should be apparent.JT_{March 16, 2009
March
03
Mar
16
16
2009
04:37 PM
4
04
37
PM
PDT}

Clive Hayden: Do you mean to say that just by virtue of living and reproducing, mutating and evolving, that the English sentence is being crafted? one evolutionary step at a time? Wouldn’t that mean that there were some fixed property that the biological viability was approximating to in it’s “iterations”? Fixed? No. If the goal were "any legal C program" for example, that would be a dynamic goal, not fixed. And a context-free parser (something that could identify a C Program) is not in principle a complex program. (BTW, If your indirect point is that there is teleology inherent in the process, I would agree with you, but it is just extremely difficult to get people to see that if they are not so inclined - just an observation.) How would biological changes amount to anything “readable” over the course of the changes?–even in their own “language”? We already know when an English sentence is being approximated to, how do we have this equivalent from biological changes–how are the changed amounting to something that we already know and can read and pass judgment on their proximity to it? I think you would have to agree that there would be potential objective standard for gauging comparitive fitness of organisms in an environment. So that if you compare two otherwise identical organisms and one has better vision due to specific refinements in its eyes. So however those refinements came about, it would be a true fact that the organism possessing them was more fit. The Zachriel mutagenator program can generate a third of the english dictionary in a few hours. I don't know even how it took that long - given that in my experience it was generated what looked like a hundred words in a few seconds. But all the intermediary words are legal words themselves. And then longer and longer valid words are formed strictly through random mutations. For just a seven character word this incremental method is several thousand times faster than blind chance (because of regularities in the english language landscape). And incredibly the longer the word is, the more improvement there is over blind chance. It very quickly becomes 200,000 times faster for say a 10 character word than a 26^10 blind search. So now imagine you have a bunch of individual words floating around, and those that can hook up gramatically start doing that to form phrases. So then the challenge is to form phrases and sentences by combining words randomly, burt that's no different than combining letters to form words. And then as phrases reach a certain length a period becomes an option. And then certain sentences that make sense together hook up. And how would we get say a children's book "Bob's New Hat" out of this, for example? Well presumably there would be a lot of copies of Bob and Hat floating around and maybe it becomes sort of like Mad Lib and we get a whimsical children's book out of it. I go sidetracked here, but I will post this as is.JT_{March 16, 2009
March
03
Mar
16
16
2009
04:07 PM
4
04
07
PM
PDT}

Clive: "What is the analogy trying to show? What would showing it prove?" To use Dawkins' other analogy, it shows that you can find your way up the natural staircase at the rear of Mount Improbable rather than clearing the sheer front face in a single bound. It doesn't address the standard ID arguments that 1) there is no such smooth path requiring only manageably short leaps or 2) if there is, then that's an improbable enough contrivance to indicate some kind of design in itself. But then, in fairness, it isn't meant to. It's not a sufficient case for "RM + NS", but it is a necessary early step in one. (But all this is IIRC as I don't have the book to hand.)anonym_{March 16, 2009
March
03
Mar
16
16
2009
03:58 PM
3
03
58
PM
PDT}

R0b @18, my bad. I didn't see the "tries" wrapped on the next line. I now agree that it probably is showing every member of the population on the screen.Pendulum_{March 16, 2009
March
03
Mar
16
16
2009
03:54 PM
3
03
54
PM
PDT}

Clive @25 Doesn’t that mean that evolution is familiar with “sentences”? Or at least English grammar. The analogy in biology would be "evolution is aware of protein chemistry". Switching from magenta to yellow, bees and dandelions don't know what we call yellow, they are just agreeing on a signal that one can make and one can look for. The dandelion blindly paints itself a color, and the bee blindly wires its brain to look for a color. If they agree, they both survive and pass down their respective recipes. over much time, they evolve a fit as close as lock and key.Pendulum_{March 16, 2009
March
03
Mar
16
16
2009
03:46 PM
3
03
46
PM
PDT}

JT, What I'm still not understanding is the analogy between when something has been, if you will, attained, or captured, such as biological viability, and how that's analogous to sentences. Do you mean to say that just by virtue of living and reproducing, mutating and evolving, that the English sentence is being crafted? one evolutionary step at a time? Wouldn't that mean that there were some fixed property that the biological viability was approximating to in it's "iterations"? What would this be? Is the analogy saying that "biological viability" (which I presume means just living, and maybe reproducing) is akin to a valid English sentence in some way? How would the analogy follow? How would biological changes amount to anything "readable" over the course of the changes?--even in their own "language"? We already know when an English sentence is being approximated to, how do we have this equivalent from biological changes--how are the changed amounting to something that we already know and can read and pass judgment on their proximity to it?Clive Hayden_{March 16, 2009
March
03
Mar
16
16
2009
03:37 PM
3
03
37
PM
PDT}

Clive Hayden [25]: Apparently, evolution doesn’t have anything in “mind”, not even a “sentence”. Well, apparently it has in mind "any viable biological organism". And it has in mind "increasingly viable biological organisms". Increasing viability would imply increasing information in an organism. adaptedness would mean the wherewithal to handle more and more situations in the environment, which would imply more and more complex organisms. So in the context of "Any valid english sentence", presumably there should be a goal of longer and longer valid english sentences. I was discussing evolution with Kris on another thread, and he said that there was no “goal” to evolution. I believe it would be contended that the solutions are in the landscape and not the selection process, but this distinction may not be relevant. What is the shortest English sentence supposed to show? that evolution can do such a thing by keeping parts of the sentence until all of the sentence is reached? Well, in the Zachriel algorithm (which is actually a more refined and thus more relevant version of the weasel program) Only valid english phrases are kept as intermediary forms. Increasing fitness is measured as increasing length of a valid english phrase (or sentence).JT_{March 16, 2009
March
03
Mar
16
16
2009
03:27 PM
3
03
27
PM
PDT}

JT, "What if the goal were “any valid english sentence [of length n].” So of all the individuals on a given iteration, it would keep the closest phrase to any valid sentence." I was discussing evolution with Kris on another thread, and he said that there was no "goal" to evolution. Maybe this is my conceptual difficulty with the analogy. Apparently, evolution doesn't have anything in "mind", not even a "sentence". What is the shortest English sentence supposed to show? that evolution can do such a thing by keeping parts of the sentence until all of the sentence is reached? Doesn't that mean that evolution is familiar with "sentences"? That seems to me like saying that a person born blind mixes paints until he gets to magenta. If he didn't already know that color, how could he attain it? I appreciate your help in making sense of this illustration for me.Clive Hayden_{March 16, 2009
March
03
Mar
16
16
2009
03:07 PM
3
03
07
PM
PDT}

R0b, "The point of WEASEL is to illustrate the contrast between cumulative selection and random sampling. Of course, WEASEL doesn’t demonstrate that the conditions necessary for cumulative selection exist in biology..." I'm really trying to understand the Weasel analogy then if it is not really like trying to find the phrase that involves the phrase. I may have a conceptual block here, but I'm still not getting it. I appreciate you taking the time to explain it to me. What is the analogy trying to show? What would showing it prove?Clive Hayden_{March 16, 2009
March
03
Mar
16
16
2009
02:56 PM
2
02
56
PM
PDT}

Prev 1 … 9 10 11 12 Next