Jerry Coyne marvels at Rubik's Cube:
Coyne:
Like everyone else, I once had a Rubik’s cube (the world’s best-selling toy, 350 million of them had been sold by 2009), but I am simply puzzle-illiterate, and gave it up quickly.
Coyne quotes Wikipedia:
The original (3×3×3) Rubik’s Cube has eight corners and twelve edges. There are 8! (40,320) ways to arrange the corner cubes. Seven can be oriented independently, and the orientation of the eighth depends on the preceding seven, giving 37 (2,187) possibilities. There are 12!/2 (239,500,800) ways to arrange the edges, since an even permutation of the corners implies an even permutation of the edges as well. (When arrangements of centres are also permitted, as described below, the rule is that the combined arrangement of corners, edges, and centres must be an even permutation.) Eleven edges can be flipped independently, with the flip of the twelfth depending on the preceding ones, giving 211 (2,048) possibilities.
which is approximately forty-three quintillion.
The puzzle is often advertised as having only “billions” of positions, as the larger numbers are unfamiliar to many. To put this into perspective, if one had as many standard sized Rubik’s Cubes as there are permutations, one could cover the Earth’s surface 275 times.And Coyne is right. The combinatorial possibilities are astonishing, and the likelihood of arriving at a combination that successfully solves the puzzle is nil, unless intelligence is applied.
Living things are an immensely more complex "solution" to adaptive puzzles than a solved Rubik's Cube.
Darwinists will argue that evolution is not random. They will insist that while variation is random, natural selection is not.
Honest people will reply that we must look at evolution as a whole-- random-variation-and-natural-selection. Evolution is a solution to an adaptive puzzle, just as a correctly aligned Rubik's Cube is a solution to a puzzle.
Both solutions are impossible without intelligence, because quintillions of undirected combinatorial possibilities preclude success.
Life, like a solved Rubik's Cube, is incontrovertible evidence for intelligent design.
'Life, like a solved Rubik's Cube, is incontrovertible evidence for intelligent design'.
ReplyDeleteNo, it isn't. Life is like an extremely large number of Rubik's Cubes, which undergo random single moves, and only the ones which are likely to directly lead to a solution are retained (all the others being eliminated) and duplicated to return to the original number. And repeated, over and over again.
Eventually, you'd get a very large number of solved Rubik's Cubes, all solved by random moves and selection for the better ones.
I thought you'd previously stated that you don't believe in intelligent design, believing instead in something called teleological evolution. Do you lie straight in bed?
It's true that the odds against having arrived at precisely the same species that we currently have on Earth is astronomical, but then again, there are an astronomical number of possible arrangement of species, making any combination certain, and not evidence of ID.
As an analogy, there are currently 7 billion people living on the Earth. The current distribution of humans, their precise locations (whether they're inside their homes or visiting friends for example), is astronomically improbable. But there has to be a single arrangement of humans, no matter how extremely improbable, and this is it. It's not evidence for ID (intelligent distribution of humans).
bachfiend:
DeleteLife is like an extremely large number of Rubik's Cubes, which undergo random single moves, and only the ones which are likely to directly lead to a solution are retained (all the others being eliminated) and duplicated to return to the original number.
Whence the foresight to determine which single moves are "likely to directly lead to a solution"? Whence the retention system ("are retained")? Whence the duplication system ("and duplicated")? Foresight, retention, duplication -- these all seem to me to be prerequisites to evolution, as you describe it. But materialistic evolution can't explain how such foresight or such systems arose, can it?
Kent,
DeleteReproductive success. The small changes result in better survival, more offspring, which then go on to having better survival and more offspring themselves.
There's no foresight. The ancestors of the dodo on Mauritius Island (and other flightless birds on oceanic islands) lost their power of flight because powered flight is expensive. If you don't need to fly, because there are no predators, then the birds which can't fly can put more energy into reproducing and outbreeding individuals with flight.
And then bored Dutch and English sailors arrived on Mauritius Islan in the 17th century and wiped the dodo out.
An insurance adjuster was sent to assess the damage to a Lego factory in Oklahoma. As he drove up to the factory, he was astonished by what appeared to be a 100 ft tall Lego model of a magnificent tower. An old Okie security guard met him at the gate, and he couldn't help but blurt out "Who built that incredible model?!!"
ReplyDeleteThe old guy laconically replied, "Nobidy. Nobidy done it."
The insurance adjuster looked at the guard with a gimlet eye and said, "OK, smart-ass. Then how did it get there?"
The guard replied, "Tornado."
But an even more compelling - and true - illustration of the nitwittery that passes for technical commentary these days is...
Steven Vogt, professor of astronomy and astrophysics at the University of California, Santa Cruz, and one of the leaders of the team that discovered the planet, said he had "no doubt" there could be life on the planet.
"Personally, given the ubiquity and propensity of life to flourish wherever it can, I would say, my own personal feeling is that the chances of life on this planet are 100%," Vogt said.
--- Reuters
Vogt is a man of deep faith.
Geneticist Dean Hamer has proposed the existence of a God gene to explain why the majority of humans believe in God.
ReplyDeleteAfter reading bachfiend's comment, I propose the existence of the a-God mutation to explain why about 2 to 3 percent of the population are atheists. Like most mutations, this mutation results in the lost of a very important human function, the ability to differentiate statistical information from semantic and apobetics information, in other words, the ability to think clearly.
Here is a book that fully describe this ailment: Illogical Atheism
It's been demonstrated that Rubik's cube can be assembled by an evolutionary algorithm: random mutations + selection.
ReplyDeleteN. El-Sourani and M. Borschbach, "Design and comparison of two evolutionary approaches for solving the Rubik’s cube," in Parallel Problem Solving from Nature, PPSN XI Lecture Notes in Computer Science Volume 6239, 2010, pp 442-451. doi:10.1007/978-3-642-15871-1_45.
Hoo
If evolutionary algorithms are that good, then Why doesn’t the software industry use evolution to write more advanced programs?
DeletePépé, Intelligent computer scientists may eventually design programs to do just that.
DeletePépé:
DeleteIt's the frame problem. To write an evolutionary algorithm, you have to know the parameters of the solution you're looking for, how to mathematically compare two proposed solutions based on the desired parameters, how to generate a proposed solution from previous solutions, and finally, how to represent the desired parameters, comparisons, and generation of solutions in code.
You're basically writing an algorithm to automate what you would do if you had time to compare tons of potential solutions in a trial-and-error manner yourself to find the best one. It's not fundamentally different than what you do when you use a calculator to find the exponent of two large numbers, just because you don't have the time to do it yourself. That's what all problem-solving programs do, ultimately.
The representations you code of potential solutions, how to compare them, and the parameters of the desired end result are the frame of the algorithm, which define the limit of what it can find.
If you don't know ahead of time exactly what the desired parameters are, or what a desired solution is, or how to compare them, or how to represent all those things in code, you can't write an evolutionary algorithm to find them for you. It's not like an algorithm (evolutionary or otherwise) can just read your mind, and understand what you want, and magically create it for you.
You are 100% right Admiral. Microsoft Visual Studio 2012 has basic code generators for a variety of applications. It's up to the software developer to then put meat on these bones.
DeleteIt's Intelligent Design from start to finish!
The Deuce, here is a post I think you will enjoy: The Darwinist and the computer programmer
DeleteDeuce: You're basically writing an algorithm to automate what you would do if you had time to compare tons of potential solutions in a trial-and-error manner yourself to find the best one. It's not fundamentally different than what you do when you use a calculator to find the exponent of two large numbers, just because you don't have the time to do it yourself. That's what all problem-solving programs do, ultimately.
DeleteThis summary indicates that Deuce does not understand how an EA works. It omits one important component: cumulative selection. Ever heard of it, Deuce?
Hoo
Deuce: The representations you code of potential solutions, how to compare them, and the parameters of the desired end result are the frame of the algorithm, which define the limit of what it can find.
DeleteThat's quite a howler, Deuce. The programmer of an EA does not necessarily know in advance the parameters of the desired solution. You specify a fitness function, which does not point to any specific solution. Instead, it compares how one "organism" preforms relative to another. In other words, the designer knows what function "organisms" have to perform, but he does not know how it will be implemented in the best "organisms." See this example.
Hoo
Doc: "The combinatorial possibilities are astonishing, and the likelihood of arriving at a combination that successfully solves the puzzle is nil, unless intelligence is applied."
ReplyDeleteIn this paper we design, benchmark and compare two different evolutionary approaches to solve the Rubik’s Cube. One is based on the work of Michael Herdy using predefined swapping and flipping algorithms, the other adapting the Thistlethwaite Algorithm.
--- El-Sourani, & Borschbach (2010)
Some computer scientists are very intelligent and creative.
They surely are. That said, they did not solve the puzzle themselves. They got a Darwinian process (random mutations and selection) to do it.
DeleteWhich brings us to a point that stumbles many ID scholars and their fans.
Suppose God set the rules of the game and then kicked back and watched life evolve without ever interfering (much like El-Sourani and Borschbach). Are you, guys, OK with this perspective? Dembski, Meyer, Behe et al. are clearly not. They insist that God intervenes in the process.
Hoo
"[T]hey did not solve the puzzle themselves..."
DeleteThen who wrote the software?
It's simple, Grandpa. The scientists did not solve the puzzle. They wrote an evolutionary algorithm that did.
DeleteThat should answer your question. Now it's your turn to answer mine. (Prediction: not gonna happen.)
Hoo
"The scientists did not solve the puzzle. They wrote an evolutionary algorithm that did."
DeleteI think you off on another flight of fancy comparing the writing of a puzzle-solving algorithm to the creation of the Cosmos. That's more ridiculous than Pintos and iPods.
But in regard to your question whether I'm "OK" with Deism... Sure, I'm "OK" with it. I don't believe it myself, but what other people believe is their business, not mine.
And at any rate, a puzzle-solving algorithm, or an autopilot that adapts to changing atmospheric conditions, or my rice cooker that uses fuzzy logic to cook a perfect pot of rice, brown or white, soft or firm, without any intervention from me is not going to convince me of either atheism or Deism.
Grandpa: I think you off on another flight of fancy comparing the writing of a puzzle-solving algorithm to the creation of the Cosmos. That's more ridiculous than Pintos and iPods.
DeleteThat was Egnor's analogy: Life, like a solved Rubik's Cube, is incontrovertible evidence for intelligent design.
Short-term memory loss, Grandpa?
Hoo
What does Egnor have to do with your flights of fancy?
DeleteWhy not just blame Bush? I do.
Egnor, Grandpa, is the owner of this blog. The thread in which we are commenting begins with an opening post that compares designing life to solving Rubik's cube.
DeleteLet me know when this sinks in. I'll explain what I did in my comments.
Hoo
I know who owns this blog and you don't need to "explain". Your comments speak for themselves. :-)
DeleteIf you raise a different subject to discuss I'll be watching for it to see if I'm interested. I'm not interested in discussing the ownership of this blog. Or you.
Bye.
Don't let the door hit you on the way out, Grandpa! I'm certain you'll be back before lunchtime!
DeleteHoo
They surely are. That said, they did not solve the puzzle themselves. They got a Darwinian process (random mutations and selection) to do it.
DeleteAdd "what evolutionary algorithms actually are and how they work" to the list of things Hoo is ignorant of, yet very smug and opinionated about.
Indeed, Sir. The absurdity of the claim that optimization software capable of solving toy problems yields insight into the evolution of life is beyond pathetic.
DeleteYo, Deuce. I know a thing or two about evolutionary algorithms. Never used them in my professional work, though I have a couple of published papers (one of them in PRL) that rely on Monte Carlo simulations. Those are physics analogs of EAs as they are based on similar concepts: random changes are filtered by external feedback.
DeleteWhat's your experience with EAs, Deuce?
Hoo
Grandpa, you're back so soon! Nurse yanked your Maxim magazine?
DeleteHoo
Toy problems, toy curriculum vitae. Typical troll.
DeleteGrandpa, what do you know about EAs? Do you agree with Deuce's description of how an EA works?
DeletePrediction: Grandpa can't tell and won't tell.
Hoo
Some computer scientists are very intelligent and creative.
DeleteOh no, the old howler that it takes intelligence to design models of natural processes - therefore nature was intelligently designed.
Oh, and I published in Nature, Science, PNAS, Evolution, etc etc on EA's, so I am God here.
Troi: "the old howler that it takes intelligence to design models of natural processes - therefore nature was intelligently designed."
DeleteYour words, not mine.
"I published in Nature, Science, PNAS, Evolution, etc etc on EA's, so I am God here."
More Troll C.V. creativity.
By the way, Troi, did you know that when the occasional American does decide to emigrate to the Netherlands, the mean IQ of both countries goes up?
Grandpa, what do you know about EAs? Next to nothing, I'm guessing.
DeleteHoo
Guess away, Hoots. It's free.
DeleteI don't need to guess. I can look up your list of publications. There is some acoustics and psychology, but nothing about EAs.
DeleteHoo
More Troll C.V. creativity.
DeleteGod-honest truth. What's your h-factor? About 3 I'm guessing, Egnor ball park.
By the way, Troi, did you know that when the occasional American does decide to emigrate to the Netherlands, the mean IQ of both countries goes up?
I did not know that. Frankly, it surprises me. Having spent some time in US labs, it seems to me that the US has to import scientists from abroad, seeing as US citizens are too stupid or not interested because there is more money in other endeavors.
His h factor is 4, and that is if you include his namesakes.
DeleteHoo
Hoots: "I can look up your list of publications."
DeleteTranslation: I have google goggles.
Worked well, for you yesterday, didn't it Hoots? Michigan! Virginia! South Carolina!
Silly troll.
It's not google, Grandpa. It's ISI's Web of Science. Bet you haven't heard of it. Look it up.
DeleteHoo
4? Ouch. Why have other scientists so wrongfully ignored the Admiral's PhD work?
DeleteTroi: "it seems to me that the US has to import scientists from abroad"
DeleteRight. That's why we have to beg them to stay. You don't read much international news, do you?
It's because he published, like, 5 papers, and the rest are conference abstracts.
DeleteHoo
Anyway, whatever Grandpa's background, he knows squat about EAs. He is here just to puff his chest.
DeleteHoo
Virginia! South Carolina! Michigan! 3-0!
DeleteI'm disappointed in you, Hoots. Didn't learn a thing.
Better luck next time.
Yeah. It's all hat and no cattle with the Admiral, if y'all forgive me using some Texan vernacular.
DeleteOf course you didn't learn a thing, Grandpa. Old people lose the ability to learn. Sad but true.
DeleteHoo
And of course, the standard for proving fraud in every state includes the intent to deceive. I have cited cases from three states, and the old fart has not cited any counter examples.
DeleteThe same goes for federal cases. Here, for instance, we read:
In order to prove fraud in a Board inter partes proceeding, as set forth by the
Court of Appeals for the Federal Circuit in In re Bose Corp., 580 F.3d 1240, 91
USPQ2d 1938 (Fed. Cir. 2009), a showing of the following four elements is required:
(1) applicant/registrant made a false representation to the USPTO;
(2) the false representation is material to the registrability of the mark;
(3) applicant/registrant had knowledge of the falsity of the representation;
and
(4) applicant/registrant made the representation with intent to deceive the
USPTO.
Plain and simple, intent to deceive must be establish to prove fraud.
Grandpa, feel free to add proper citation of that case. I know you're good at that.
Hoo
United States Court of Appeals, Fifth Circuit:
DeleteTo succeed on a claim of fraudulent registration, the challenging party must prove by clear and convincing evidence that the applicant made false statements with the intent to deceive the licensing authorities.
Hoo 4, Grandpa 0.
Hoo
Egnor's argument for design from Rubik's cube is not very convincing. He mentions the number of combinations, forty-three quintillion, and we are supposed to be impressed! The number is large, all right, roughly 10^{20}, but it only constitutes 65 bits of information. That does not even rise to Dembski's universal probability bound (500 bits).
ReplyDeleteTo be sure, a random walk will take a long time getting there. But, as many will surely suspect, environmental feedback (selection, fitness, that kind of thing) can make the task a piece of cake.
In a magnet, atomic magnetic moments can point up or down. The number of possible configurations is humongous: 2 to the power of the number of atoms. This DWARFS Rubik's cube numbers. Yet magnets have no problem finding the unique state with all magnetic moments pointing up when they are cooled. Why? Environmental feedback. The fitness function is energy, so they are pushed toward the low-energy state.
With Rubik's cube, the same thing works (see that paper).
Hoo
Hoo, you seem to suffer from the a-God mutation!
DeleteThis comment has been removed by the author.
Deletea-God mutation confirmed!
DeleteMy own H factor is about 840m with low loss compensation for drop.
ReplyDeleteIt literally blows my opponents minds.
Can we lay of the dick measuring please. It makes the folks who do not compete in these games sick to the stomach. It's like listening to kids compare video game scores.
Cheers.