…and this is a 
The more ignorant you are, the easier it is to disprove evolution
Category: Creationism
Posted on: January 31, 2007 7:07 PM, by PZ Myers
Uh-oh. Evolution has just been refuted by a very sophisticated simulation. Try it; you'll quickly discover how frustratingly boring evolution can be, and you'll give up on it.
The 'simulation' is simple: put some random text in a box, click on a button, it randomly substitutes a random letter for some other letter, and whoa…you'll notice that your excerpt from the libretto of Figaro hasn't been transformed into the Gettysburg address. Therefore, evolution is false.
Seriously, it's that bad, and the author actually does think he has accomplished something significant. It's a simulation that requires the user to make multiple clicks per trial; by sheer clumsiness of the interface it reduces the number of trials that can be done. It ignores fitness functions, synonymy, ranges of functionality, multiple functions, etc., etc., etc., all in the name of slavishly and crudely mapping English to protein evolution, and doing it all with a program that looks like something a sixth grader would slap together. Oh, but it is so much better than Avida or Tierra.
Read through the writeup—it's appalling how ignorant the fellow is of basic biology. This kid really needs to read the Evolution and Chance FAQ, the Evolution Proceeds by Random Chance FAQ, and Musgrave's ripping apart of bogus abiogenesis calculations. It's probably good enough to convince your standard issue clueless creationist, though.
(hat tip to Tobasco da Gama)
Comments
I managed to change "Creationism is a crock of shit" to "Creationism is a crock of shat" in just one mutation! There must be something to this random mutation thing :)
Posted by: Teapot | January 31, 2007 7:17 PM
My first click went from "The quick brown fox jumped over the lazy dog" to "The quick brown fox Pumped over the lazy dog". I can't see the creationist Christian nutcases being too happy with filth like that being promoted in their name!
Posted by: Orpheus | January 31, 2007 7:22 PM
I changed "when you were born they threw out the baby and kept the placenta" ... but I wrote it in Spanish and nothing interesting cam out =(
Posted by: Tinni | January 31, 2007 7:23 PM
The stupid, it burns! (HT Atrios)
Posted by: afterthought | January 31, 2007 7:24 PM
You know, if the guy had actually written his "random mutation simulator" to do many trials (thousands, millions?) at once and incorporate some sort of selection, it might have been cool.
But this thing is just crap.
Posted by: Rick @ shrimp and grits | January 31, 2007 7:51 PM
Posted by: df | January 31, 2007 8:17 PM
Perry Marshall is Pat Boone's clone child.
Posted by: Matt Platte | January 31, 2007 8:28 PM
Ok, I give up: what does the "Select" button do?
Since the experiment is running on a population of 1, I suppose it's lack of doing anything is sort of inevitable...
Posted by: Millimeter Wave | January 31, 2007 8:35 PM
The "Select" button seems to save the current string. Clicking on "Revert" turns back time to that point.
Posted by: Amos | January 31, 2007 8:44 PM
I changed "fuck" into "shits" in under 8 minutes.
Posted by: Jeff | January 31, 2007 8:53 PM
Okaaaaay...
What a doofus. Still, it probably won't prevent approving writeups of his wonderfully clever and insightful website. :-o
Posted by: Millimeter Wave | January 31, 2007 8:53 PM
Ummm...the first 2 links for the FAQS (Evolution and Chance FAQ, the Evolution Proceeds by Random Chance FAQ) don't work.
Posted by: Krystalline Apostate | January 31, 2007 8:57 PM
I put in a simple, primitve organism, the letter "a", and in 3 clicks I was at "I". That's like from amoeba to people in 3 mutations! Evolution works, and WAAAAY faster than I ever thought it could. Imagine if there'd been natural selection.
Posted by: QrazyQat | January 31, 2007 9:03 PM
Wow.
To think he missed one of the most important prerequisites of evolution, and that's LIFE. Last I checked, a string of letters isn't alive.
That really is pathetic.
Posted by: thaumaturge | January 31, 2007 9:05 PM
It's actually pretty nice, I think. Hit "mutate" once, save it if it's a real word. Otherwise, revert. It might take a lot of work to get the sentence you'd hoped to get, but that's to be expected; evolution isn't teleological, is it?
Of course, he says only that you usually get gibberish after one click, then goes on to claim that this disproves evolution. The appropriate analogy here would be: "Evolution is impossible because a new species can't evolve with only one mutation!"
Hey, PZ, if you can get the code behind the mutation generator, can you add features for mutations other than the usual point mutation? Add a feature that copies the entire string (or part of it) into a new box; with a duplicate of part of the string, one of the two can mutate into gibberish with no ill effect. Or maybe simply write a bot that will activate this thing repeatedly and select certain words and phrases. Use his own weapon against him!
Posted by: Maronan | January 31, 2007 9:07 PM
randommutation.com
Registrant Contact:
Perry S. Marshall & Assoc
Perry Marshall (info@perrymarshall.com)
+1.7087884461
Fax: +1.7087884599
1508 Ridgeland Ave
Chicago, IL 60402-4900
US
Posted by: thaumaturge | January 31, 2007 9:10 PM
Yeah, this guy's a piece of work, an Internet marketing guru who has decided to branch out into religion lite. If you like, he'll send you misleading e-mails that will make it seem as if he's just a honest seeker after truth with no ax to grind, etc. Correspond long enough and suddenly he's got some Jesus to sell ya.
Which, I think, is just dishonest. If you're a Christian (like I am), say so. Don't hide it with marketing gimmicks, then spring it on people after you've hooked them. I sent him an email complaining about it, and got a form letter reply that appeared to have been generated by a program.
SH
Posted by: Scott Hatfield | January 31, 2007 9:12 PM
This just seems like a ridiculous craptaculation of the Richard Dawkin's "methinks it is like a weasel" simulation.
Posted by: YuppiTuna | January 31, 2007 10:36 PM
But "he's trained in engineering"! How can you ignore the arguments on someone who is "trained in engineering"? Engineers are the leading intellectuals of our age!
Posted by: Zeno | January 31, 2007 10:53 PM
Perry Marshall has been at it for more than a year at IIDB in a thread in their EvC forum titled Proof of God via DNA and Evolution. He claims on his website to have refuted every challenge of the atheists. I know there are many who disagree, including me. His 'proof' is actually a faulty inference, but each time one of the many flaws is brought up, he just requotes his same old Yockey material, which even Yockey himself claims does not indicate a god or designer. He seems to be just another of those who know the answer and try to fit the evidence to his presuppositions.
Posted by: Ray S | January 31, 2007 11:04 PM
Well, he says he has a degree in engineering... I wouldn't say that says too much. But I don't know; would you say somebody who has a physics degree and then went into sales can be reasonably called a "scientist"?
Posted by: Millimeter Wave | January 31, 2007 11:16 PM
All these pathetic attempts by creationists at trying to simplify the "debunking" of evolution shows their contempt for their fellow human beings. Apparently they believe nobody has tried these obvious and intellectually lacking first level experiments. In fact, how they come up with these "experiments" at all is beyond me. That someone spent time on this coding it, putting it on the internet, and making up random formulations of fragments of thoughts seen in books, the internets and TV, is quite obviously the wheel spinning of someone with too much time on their hands because they didn't bother to go to school, get educated, and get to the work of pushing humanity along.
Posted by: BlueIndependent | January 31, 2007 11:24 PM
ARGH! He's an ENGINEER? I totally missed that. Dammit, that pisses me off even more than his complete ignorance of experimental methodology and the ridiculous application of teleology to evolution.
Posted by: Joshua | February 1, 2007 12:09 AM
He's an ENGINEER? I totally missed that.
Old saw:
A mathematician, a physicist and an engineer were asked to prove that all odd numbers are prime.
Mathematician: "1 is prime. 3 is prime. 5 is prime. 7 is prime. The proof follows by induction."
Physicist: "1 is prime. 3 is prime. 5 is prime. 7 is prime. 9 is... an experimental error. 11 is prime..."
Engineer: "1 is prime. 3 is prime. 5 is prime. 7 is prime. 9 is prime. 11 is prime..."
(Disclaimer: I went to engineering school.)
--
Posted by: marquer | February 1, 2007 1:01 AM
YuppiTuna said:
PZ, I think you missed a great teaching moment. Dawkin's WEASEL (online version) is a much better simulation. Granted, you have to be careful trotting out the program among creationists as it's directed to a goal, but it's good as a demonstration of random mutations and natural selection. Also, more complex simulations that work more like living organisms, like Lenski et al's (2003) Avida are pretty cool tools as well.
Posted by: jpaulr | February 1, 2007 1:06 AM
Heres a youtube clip I saw a while back that adds selection to the random part of one of these simulations and quite neatly shows how you can rapidly come up with complicated solutions from an initial starting point.
http://www.youtube.com/watch?v=M2SVMKZhV2g
Posted by: MartinC | February 1, 2007 2:17 AM
Funny thing, I just had my monkey click it an infinate number of times, and he wrote the play "Hamlet"
Posted by: Revy | February 1, 2007 2:20 AM
Childish! That was the first word that came to mind when I saw this, just like PZ says it:
Apart from reading up on basic stuff about evolution, the kid should perhaps also widen his scope a bit. May I suggest a bit of Oscar Wilde, such as the following quote?
"Alas, I am no longer young enough to know everything."
True learning is not only about expanding what you know, but also about understanding the (present) limitations of that knowledge.
And yes, even an engineer can understand that...
Posted by: Thinker | February 1, 2007 3:02 AM
"The more ignorant you are, the easier it is to disprove evolution."
I ran across this saying last week. It must be a corollary to your title.
Law of Logical Argument:
Anything is possible if you don't know what you are talking about.
Posted by: djmullen | February 1, 2007 5:01 AM
Marshall is, of course, an engineer....
Posted by: slpage | February 1, 2007 8:07 AM
I tried typing in "Dembski sucks". You can probably guess what the very first mutation turned it into.
Posted by: Iain Walker | February 1, 2007 8:35 AM
I like the Weasel Applet. It went from
"YEv8?gXgZ8V h8,YctRY1smaAJ9GBRWhiPClkyekH9 ffmt0oTtnweZY161"
to
"Intelligent designists are neither intelligent nor designed"
in 618 generations and 159232 tries.
Posted by: N.Wells | February 1, 2007 9:18 AM
I got it to say "YOU'RE THE MAN NOW, DOG!" after only three tries. Srsly.
Posted by: Fatmop | February 1, 2007 9:25 AM
"It's probably good enough to convince your standard issue clueless creationist, though."
They come pre-convinced. What they are looking for is excuses.
Posted by: Alexandra | February 1, 2007 9:44 AM
You know... not only does this suffer from the lack of a fitness test, it's expecting too specific a result. "Randomly change a word until it becomes a recognizable English word." This is like saying that evolutionists claim that humans, canaries, or gigantic cephalopods were inevitable. If you wanted to make this an interesting trial, you'd have to come up with some meaningful general definition for a language and test for that. Of course, that would involve an interest in doing actual science to make your case :p
Posted by: Beren | February 1, 2007 10:09 AM
Man. That is amazing. :-o
Posted by: David Marjanović | February 1, 2007 10:50 AM
Step 0: begin with a fixed dictionary of words (your choice of alphabet, language, and corpus).
Step 1: generate the set of strings of length n which are 1 letter different in one location from the dictionary in Step 0, i.e. one point-mutation away from the dictionary, i.e. strings of Hamming distance 1 from strings in the dictionary. Note that this overlaps the dictionary, but is not the same nor the complement.
Step 2: generate the set of strings of length n which are 1 letter different in one location from the dictionary in Step 1, i.e. one point-mutation away from the dictionary.
lather, rinse, repeat.
By step n, you have the set of all x^n strings of n letters, where x is the number of letters in your alphabet.
The next level of sophistication is to replace point mutation by a set of other operators. For instance, the Levenshtein distance, or "edit distance" where the operators are insert a letter, delete a letter, or point mutate a letter.
The next level is to use a measure of edit distance which takes into account the actual operations on the genome, weighted by the relative likelihood of adding a trait versus losing a trait.
These make simulations which are much more fun. Reference also Lull (a.k.a. Lully), "The Library of Babel" by Jorge Luis Borges, and the phrase "O! Time they pyramids", and the letter-frame cranked device on Swift's Island of Laputa.
Posted by: Jonathan Vos Post | February 1, 2007 12:41 PM
What does it mean that my input "Creationism is ridiculous" mutated into "666"?
Posted by: PTW | February 1, 2007 12:55 PM
PZ, there is no need to insult six graders. You have no idea what they are capable nowadays in a modern society so long their natural curiosity is satisfied in a sufficient manner.
Posted by: lo | February 1, 2007 1:28 PM
Peter M. Huggins, Lior Pachter, and Bernd Sturmfels, "Towards the Human Genotope", received 26 Sep 2006, Abstracts of the AMS,
Vol. 28, No. 1, Issue 147, 2007, 1023-11-1803, p.211.
Given a collection of genotypes, their genotope is the polytope defined as the convex hull of all allele frequency vectors that can arise from populations over the collection of genotypes. On the theoretical front, Berenwinkel et al have shown that regular subdivisions of genotopes encode shapes of fitness landscapes and generalize the concept of epistasis to arbitrary numbers of genes. Now on the parctical side we aim to show that it is computationally feasible to compute certain projections of subpolytopes of the human genotope. We report on three classes of low-dimensional projections: projections specified by principal component analysis, by restriction to few SNPs, and by archetypal analysis.
Posted by: Jonathan Vos Post | February 2, 2007 2:10 PM
Here's my weasel DNA code: (a lot simpler than a real weasel's DNA I'm assuming)
kjfhwe8yrg9uweijgw'eprogiw[-509ihkw'e5oiw[905u5'wyjw[-4509jwe'hjw'eopgkpgG(UPGJ9p4TQ3P'g(j($gj$pohjohju$jgj:geih"o:eig*hoi#$hjg)(*hngiwjg"(@)#$gh"I$H{*)$H"OIH)*#$hgnbg"ipbh"o$bjV{M V{()$#UF{)UG{"IJg{m$()UG{)$(UGM{"PWJMjvd'spigjao'rpigja[0r8uyg"erijgw895u4g'WIRJG[098UGJQ34IJGA'OERIGUBA[8RGHAEL5NGQGJ'OTU[A0RGU[0QAE9TGA08UN[)(*&$NW[0YU[0W4YUQ5[E9RU"RPOJS;OHIUJ AE58U YQ5IJJ13TU[0T9QU0[29UYQ[098YUNH A[0ER9YUA [EI9GAM TEHPOIAE[H09UMG[-9GTIM[]A-98HNGQ935 'ALKFO"pj"pijo:h prjGPG098NA 5IGU IUGAG;OAGJ'APRIUGA[NIG APRIGA'RP9GUNA[0RGJ RESOJG AEROIGJAR'GOIPJAR'HAOPRIEJGAR'GOPIA ERG'AREPG9UIA[09RUGQ43I9GUQ098U7RNA7G4A7T490TN84
the online weasel creator tool has been stuck at 84% and a score of 95 for half an hour or so now and 40,000+ generations.
Anyone else want to see if they can evolve me more than 84% of my "weasel"?
In the mean time I'm going to take my 100% of a Dog for a walk and see if my weasel gets any better evolved when I get back. (Maybe I'll walk to the book store and grab one of Perry's books to read while I'm waiting too)
The only point of that simple random mutation app he provides is to demonstrate entropy. The point is random mutations alone degrade information always. Period.
Seriously, I'd like to know if anyone with a "real" computer can get it past 84% and how many generations it took to get to 85%. Anyone game? Anyone got a super computer handy?
-Greg.
Posted by: Greg M | February 4, 2007 12:41 PM
I was able to turn 'crap' into 'DNA,' in a few minutes. It is pretty cumbersome, but I suppose it could be made to function with some effort on a larger scale.
I'm glad to see the ID folks trying to illustrate what they are saying with simple simulations to compare with what evolutionists use to illustrate their points. It's a good way to read back their understanding of natural selection, and to focus on points of misunderstanding.
Evolution critics are evidently claiming that in the weasel simulation (or similar), the "correct" parts of the string are frozen and simulated mutations only occur in the parts that are not yet "correct". I haven't studied specific versions of the weasel code but I assume that this is accomplished by discarding "bad" mutations and retaining good ones, rather than actually freezing the good characters. From this simple simulation, it would appear to take quite a few discards to get a single step in the right direction. This is somethign I believe the ID folks refer to as the "cost of variation".
This is perhaps a simple but fundamental point that if clarified could go a long way to convince folks about these simple simulations. It could certainly be explained with considerably more patience than folks here seem capable of mustering.
I for one would like to see this comparison taken seriously enough to lay it out in neutral language that isn't designed to be insulting. You don't see textbooks dissing Ptolemy as they explain how we got to Kepler and Copernicus, do you? If the point is make clearly and powerfully enough, one doesn't need an insult to influence the onlookers.
Posted by: Tom Moore | February 7, 2007 8:53 PM
ur a fag...evolution is false...athiest person
Posted by: r | April 10, 2007 3:18 PM
That's gotta be a joke. hehe.
Posted by: Steve_C | April 10, 2007 3:21 PM
yet another example of someone who quite blatantly doesnt understand even the BASICS of evolution, yet attempts to disprove and discredit it anyway...
Posted by: John | October 9, 2007 10:43 AM
It may interest some that he's responded to this blog at:http://randommutation.com/darwinianevolution.htm.
Besides calling him a 6-yr old, anyone have some way to prove his hypothesis wrong? The bold hypothesis he says this random generator displays is:
"To demonstrate as clearly as possible that the notion of Random Mutation as a source of evolutionary progress is utterly false and absurd. Evolution by Random Mutation is nothing more than a fairy tale. It is the most widely circulated myth in 500 years of science. The dogma has been repeated ad nauseum but is always assumed, seldom examined, and never demonstrated by anyone to actually be true."
Posted by: phinehas | February 25, 2008 12:24 AM
he has a degree in engineering
Of course he does! =D
Posted by: Kseniya | February 25, 2008 1:05 AM
I'd advise you all to take a look at http://www.cs.odu.edu/~cmorris/G-DES/, a beta software that I've been developing / playing with. It was designed to debunk this Perry fellow, before I realized that his "religious views" were nothing more than a marketing gimmick and his "science" nothing more than speculation.
Originally I was trying to help him safe face, and prove to him that he was wrong without dragging it out in the public forums of the net.. but he ignored my emails.
What do you want to bet that he's doing some analysis on the hits on and responses to his site? Marketing through religion, re-invented copyright 2007 Perry S. Marshall....
Posted by: Charles A. Morris | June 24, 2008 2:59 PM