For general discussion that would be off-topic in other threads!
The Skeptical Zone
"I beseech you, in the bowels of Christ, think it possible that you may be mistaken."
For general discussion that would be off-topic in other threads!
Moral Turpitude : Unicorn ? Horse
Just trying a LaTeX plugin thinking it might be useful. It’s Quick Latex and should work in posts and comments. Instructions are on the Quick Latex website. Here
Thanks, Alan. That will make things a lot easier in threads like the ASC one.
Actually, it is the same package Lizzie installed a while ago and was deleted along with all the other plugins after the last crash.
It does seem as easy as
Testing, 1, 2, 3…
Alan, I’d like to offer some pro-bono WordPress consulting for this site. No reason to have troubles like this. Here’s some insight on qualifications for me: http://goo.gl/adl6Lm & http://goo.gl/Ih4GU1
Paul,
Hi Paul, welcome to TSZ and thanks for the offer.
Elizabeth Liddle is the blog owner so it would be up to her. I’ll alert her to your comment.
Hi Paul,
Welcome to TSZ.
We already have a white-collar linguist, so I hope you’ll stick around for balance. 🙂
Ooh yes please! Thanks!
Hi,
I have what I suspect is a dumb question. Is there a way for me to find my previous posts? I searched for them by my user id but couldn’t find them. I’m guessing (hoping) that there is some way for me to look up what I posted in the past. I was only on a few times about two weeks ago, so I don’t have much to look for, but I wanted to check on some responses.
Thanks.
Hi onetime,
I found your comments starting here.
keiths,
Thanks Keiths, How did you find them? I don’t get on here a lot, and I’ll probably need to find my comments again .
Thanks.
onetime,
I googled the following:
The quotes around “onetime” keep Google from returning results for “one time” and “one-time”.
Keiths,
Thanks!
Today will go down in infamy. “Genomics” is publishing an issue edited by James Shapiro and Shi Huang. Two of the “Third Way” kooks. Shi Huang has been doing a Kariosfocus imitation at Sandwalk for a week or two.
And if that’s not enough to make the day, Jerry Coyne has been banned from FaceBook for unspecified reasons.
I used to like Shapiro.
Good example of emergence from his wikipedia page:
“Shapiro has also studied pattern formation in bacteria, an area where he feels that there are new mathematical principles to be discovered that also underlie the growth of crystals and the shape of cosmological structures.[6] For instance, he found that the gut bacterium proteus mirabilis forms complex terraced rings, an emergent property of simple rules that the bacterium uses to avoid neighboring cells.[19]”
There’s nothing nutty about asserting that physical chemistry constrains evolution and probably shapes some structures.
But Shapiro thinks stuff like this is evidence of a Designer. Not just design.
I do not wish to step on some excellent ongoing threads, but I think it’s time to crank up another Wesel thread. The level on non-comprehension at UD is inconceivable. So to speak.
No, I do not agree. The typical versions of Weasel operate on fixed length strings, but that is not a requirement for such a program. Nor is it necessary to have a target.
Perhaps when things are slow, I’ll post.
IBM’s Watson computer can now do in a matter of minutes what it takes cancer doctors weeks to perform
Read more: http://uk.businessinsider.com/r-ibms-watson-to-guide-cancer-therapies-at-14-centers-2015-5#ixzz3aWATSvGN
Google plans to have autonomous cars for sale in five years.
petrushka,
Winston Ewert’s response to Joe and me indicates that the Evolutionary Informatics Lab has finally caught on to the fact that the Weasel program doesn’t actually search for a target. However, it’s still easy to make things hard for them. Choose the initial target uniformly at random, mutate the target from time to time, and let the program run forever. I called this the Wandering Weasel a long time ago, and somebody at AtBC implemented it. (It’s old hat in the theory of evolutionary computation. I don’t like doing lit reviews for creationists.)
These days, I believe that the real issue is the difference between a technologist using a process to cause a desired event to occur, and a scientist using a process to model the known occurrence of an event. For instance, a technologist may know the properties of a solution to a problem, and seek to identify an object with those properties. It is then the technologist, not the process initiated by the technologist, that seeks to cause the event (solution set of the problem) to occur. That’s not at all what’s going on with the scientist. I’ve been struggling to come up with a clear explanation, and I’d appreciate the help of you and others.
Fun little bit from UD. Larry Moran defending the term IDiot.
Is it assumed in Wandering Weasel that the target string keeps changing, with the scores reflecting how far the current string is from that target string?
Of course, in Dembski, Ewert, and Marks’s paper the target is chosen possibly independently of the fitnesses. The Weasel could then be signaling how close one was to something that is not, or is no longer, the target. I think in your Wandering Weasel these are kept coordinated, right?
In my word game program the “target” is words and substrings of words from Scrabble dictionaries in five different languages. My math is limited, but I believe this is a markov selector. No word is given preference in the selection weight, but the more and longer matches, the higher the score.
You can change language in the middle of a run, and the population will shift.
There is no specific target, so the population will continue evolving indefinitely.
Joe Felsenstein,
I wrote a long response, but lost it because my TSZ session expired before I clicked “Post Comment.” I don’t have a repeat in me at the moment.
Yes. It is very easy to modify an existing program to do this. In each generation, pass the target to the mutation routine, along with a relatively low mutation rate, to obtain an “offspring” target. Replace the target with its “offspring.” Now it makes sense only for the program to run interminably. Existing theory addresses how well an evolutionary process tracks the moving target. The framework of DEM doesn’t allow for it. The biggest problem for them is that the Wandering Weasel does not halt and produce an outcome.
Of course. And we have to be careful here with the term target, because the “target” in terms of which fitness is defined is not the “target” in terms of which bias (“active information”) is defined. It seems to me that Dembski, Ewert, and Marks thought, in most of their work, that the two coincided. That’s the only way I see to make sense of their claims that the fitness function supplied information about the target.
The coordination you’re referring to is there. However, the signaling is in the program, not the modeled system. The offspring have their fitnesses because of what they are, not because an oracle (to use DEM’s term) is indicating how close they are to the presently-ideal type.
Not sure I get this. In Wandering Weasel there is an “ideal sequence”. The program continually measures how far a sequence is from that.
The “fitnesses” are a function of the number of differences. So the program doing that is not an oracle?
I meant also to indicate that I understood that the “ideal sequence” kept changing.
It is bizarre, we’ve agreed, for DEM to treat the fitness landscape as part of the “search.” This derives from Dembski’s misunderstanding of the Weasel program. He thought that the point was to hit the target, and that Dawkins had guided the program to the target by supplying it with a fitness function.
Dawkins actually showed how fit characters accumulate in a process alternating between reproduction-with-variation and survival of an offspring of maximum fitness. I’ve said survival instead of selection because Dembski and Marks speak as though something literally processes fitness data to decide which of the offspring becomes the parent in the next generation. It is only the program that does this. In the process simulated by the program, the offspring compete to survive, and how well an offspring competes is determined by the number of adaptive traits it has. (Dawkins may have say something about a breeder. If so, ignore it.) The fitnesses of offspring do not exist as data in the simulated process. Even if they did, nothing exists to register them. There is no information going on.
In short, it’s what the simulated system does, not how the simulator works, that matters. Am I making sense? I’m more than a bit tired at the moment. Ask me some followup questions, and I’ll try again later.
It seemed to be necessary to get their result, but their mathematical results (as we showed) do not imply that any Active Information must result from Design, since simply having fitnesses and reproduction gives one a substantial amounf of that, and smoothness of the fitness surface “because physics” gets you more.
Modeling “white noise” as the default fitness landscape is unreal. Their default model is even more unreal than that, as it doesn’t even have organisms that reproduce. And their scheme is supposed to apply to situations where an existing organism reproduces.
The internal machinery of Weasel is obviously irrelevant.
Your Wandering Weasel models situations where the fitness surface changes continually. That is an interesting case, but not one that can be said to have an overalll single fitness surface, let alone an overall Target.
Precisely. I was trying to expose the absurdity of DEM’s approach with a minimal, and furthermore meaningful, extension to a model everyone knows about. The problem, I see now, is that Dawkins didn’t present the Weasel program as a model. He simply explained what it did, illustrating his notion of cumulative selection.
When I take another shot at this, I’ll provide a minimalistic model, with traits that are either present or not, and that are either adaptive or maladaptive, depending on current conditions. (The model is implemented by the Wandering Weasel program with a binary alphabet. But saying that won’t help me straighten things out.)
DEM can plug the changing fitness function into their “search.” But what are they going to do — attempt to track the “search” by moving the “target,” and then say that the “search” tracks a moving “target”?
Test. You ordinarily can’t enter $\LaTeX$ and get
Tom English,
I have little luck with that. If I enter
what will happen?
Joe Felsenstein,
It worked! (Faints).
Yes, Dawkins’s Weasel was not any attempt to fully model evolution, but a teaching example, a very clear one. It was an attempt to counter the lies that creationist debaters tell when they describe evolution as “random”. They want their audiences to conclude that impressive adaptations could not be accomplished just by throwing stuff together at random. The Weasel was intended to teach people that the directional biases of natural selection could be far more effective than randomly mutating genomes.
Of course creationists and ID advocates have ignored this and presented Weasel as a “model of evolution” and then argued that it is a very bad model.
I think that implicit in their argument is that regions of high fitness are the target. If we rank genotypes by fitness, that a cutoff somewhere in that ranking denotes a set (all genotypes of higher rank than that) which is to be the Target. But boy, they are awfully unclear about that.
As we know, in the DEM argument they can’t do that, because if they confine attention only to cases where there are genotypes, and those have fitnesses, they will be unable to prove their theorems.
I was unclear. Let me reword:
[Creationist debaters] “want their audiences to conclude evolutionary biology is saying that impressive adaptations are accomplished just by throwing stuff together at random. And the audiences know that this is no explanation.”
The lie is the assertion that evolutionary biologists are saying that evolution is a purely random process. Creationist debaters hammer away at this, throwing around the word “random” a lot, but never clarifying how natural selection makes the result very nonrandom.
Joe Felsenstein,
Natural selection is not a mechanism. That is part of the Darwinist cabal lies-because obviously admitting that your theory is random, is really hard to make your case believable. So its only random when it convenient to rule out the designer.
phoodoo,
Not that again.
Sure, and there are no such things as landslides. Because those are just lies of the geologists’ cabal. They are really just rocks, mud, and dirt falling down a hill.
You presumably do agree that differential fertility and differential viability exist — and that when they do, the changes that occur are very nonrandom. So when creationists and ID supporters repeatedly use the word “random” to describe the processes of evolution, and never explain the nonrandom part, they are massively misleading their audience.
You agree with that, right?
Right.
Joe Felsenstein,
You think that attractive people have more babies? People who can sing well, procreate more? People who have more symmetrical faces are more likely to marry? Ladies with “birther hips” get laid more? Ladies with small breasts do get as many chances to have babies? Smarter people are more likely to pass on their genes?
Nonsense.
Well, perhaps you have a point there.
Mystified. Did not discuss any of phoodoo’s traits.
I ask again. Phoodoo, do you agree that differential viability and differential fertility exist? And that these bias the outcome of evolution?
I presume that you agree. Right?
Joe Felsenstein,
I just answered you Joe. The suggestion that some higher level organisms pass on their traits better because of some tiny advantages in their physical make-up in nonsense. Alligators don’t mate more often because they have a better shaped snout. Cockroaches don’t make more or less babies depending on the shape of their feet. Intelligence plays no role in how often snakes breed. The color of ones skin is not a deciding factor for the number of offspring you are going to get from a lizard.
You idea that it is a physical embodiment that determines the likelihood of procreating is unsupported assertion that has no bearing to reality. Luck. Happenstance. Their place in the world at the right time, that is what decides whose genes get passed on. It has nothing to do with being more or less fit (unless we use that that awful concept your side thought up, which only means that which was at the right place at the right time).
Does phoodoo have a better explanation of why there are alligators, cockroaches and lizards?
phoodoo,
Genes have no effect upon their own persistence. That is phoodoo’s supported-in-detail counter to the ‘unsupported assertion’ that they can.
Alan Fox,
A better explanation than accidental rocks, accidentally making more copies of rocks? Well, sure
A little old man under a bridge knitted them using shredded micro-film of cactus photos.
Because pretty much ANY theory is better than the one about rocks somehow copying themselves poorly.
phoodoo,
Clearly you don’t! 🙂
Allan Miller,
Do you think it is genes which decide which humans have the most babies?
I don’t speak for Allan Miller, but my view is that it is the environment that makes these decisions. The environment is the designer.
I see, when you argue that “natural selection” isn’t a force, you aren’t just saying that what’s doing the job is differential fertility and differential viability. You actually think that there are no differences in viability and fertility among genotypes !
There are of course many studies where people have measured the differences in fitness between animals (or plants, or protists, or bacteria) with different phenotypes, and/or different genotypes.
But you seem to somehow know that when two alligators have different-shaped snouts this can’t possible make one more likely, on the average, to catch prey and be better-fed and more likely to survive or reproduce. Of course chance happens to all alligators, but we’re talking about averages over those chance events.
It’s rather like playing poker. Any given game has a large element of chance, but on average some players are more likely to win than others.
I’m fascinated by phoodoo’s wisdom. We need phoodoo to explain this more, to let us know why phenotypes of organisms don’t matter to their chances of surviving and reproducing. And why it’s not true that some poker players are better than others.