(Sorry this is so long – I am in a hurry)
Gpuccio challenged myself and others to come up with examples of dFSCI which were not designed. Not surprisingly the result was that I thought I had produced examples and he thought I hadn’t. At the risk of seeming obsessed with dFSCI I want assess what I (and hopefully others) learned from this exercise.
Lesson 1) dFSCI is not precisely defined.
This is for several reasons. Gpuccio defines dFSCI as:
“Any material object whose arrangement is such that a string of digital values can be read in it according to some code, and for which string of values a conscious observer can objectively define a function, objectively specifying a method to evaluate its presence or absence in any digital string of information, is said to be functionally specified (for that explaicit function).
The complexity (in bits) of the target space (the set of digital strings of the same or similar length that can effectively convey that function according to the definition), divided by the complexity in bits of the search space (the total nuber of strings of that length) is said to be the functional complexity of that string for that function.
Any string that exhibits functional complexity higher than some conventional threshold, that can be defined according to the system we are considering (500 bits is an UPB; 150 bits is, IMO, a reliable Biological Probability Bound, for reasons that I have discussed) is said to exhibit dFSCI. It is required also that no deterministic explanation for that string is known.”
(In some other definitions Gpuccio has also included the condition that the string should not be compressible)
These ambiguities emerged:
Some functions are not acceptable but it is not clear which ones. In particular I believe that functions have to be prespecified(although Gpuccio would dispute this). Also functions which consist of identifying the content of “data strings” (a term which is itself not so clear) are not acceptable because the string in question could have been created by copying the data string.
The phrase “no deterministic explanation for that string is known” is vague. It is not clear in how much detail and how certainly the deterministic processes have to be known. For example, it appears from above the possibility that the string in question might have been copied from the string defining the function by some unknown method is sufficient to count as a known deterministic explanation. This implies that really it is sufficient to be able to conceive of the very vague outlines of a determinist process to remove dFSCI. I think this amounts to another implicit condition: no causal relationship between the function and the string.
Lesson 2) dFSCI is not a property of the string.
It is a relationship between a string, a function and an observer’s knowledge. Therefore, it may be that dFSCI applies for a string for one observer with a certain function but not for another observer with a different function. The rules for deciding which function are not clear.
Lesson 3) The process for establishing the relationship 100% specificity of dFSCI and design is not commonly found outside examples created by people to test the process.
Gpuccio says thisabout the process:
“To assess the dFSCI procedure I have to “imagine” absolutely nothing. I have to assess dFSCI without knowing the origin, and then checking my assessment with the known origin.”
When challenged he was unable to name any instances of this happening outside the context of people creating or selecting strings to test the process as in our discussions. This is important as the dFSCI/design relationship is meant to be an empirical observation about the real world applicable to a broad range of circumstances (so that it can reasonably be extended to life). If it is only observed in the very special circumstances of people making up examples over the internet then the extension to life is not justifiable. To give a medical analogy. It might well be that a blood test for cancer gives 100% specificity for rats in laboratory conditions. This is not sufficient to have any faith in it working for rats in the wild, much less people in the wild. Below I discuss what is special about the examples created by people to test about the process.
A Suggested Simplification for dFSCI
dFSCI says that given an observer and a digital string where:
1) The observer can identify a function for that string
2) The string is complex in the sense that if you just created strings “at random” the chances of it performing the function are negligible
3) The string is not compressible
4) The observer knows of no known deterministic explanation for producing the string
Then in all such cases if the origin eventually becomes known it turns out to include design.
Given the rather lax conditions for “knowing of a deterministic mechanism” that emerged above, surely (2) and (3) are just special cases of (4). If (2) or (3) were present then deterministic mechanisms would be conceivable for creating strings.
So the dFSCI argument could be restated:
Given an observer and a digital string where:
* The observer can identify a function for that string
* The observer cannot conceive of a deterministic explanation for producing the string
Then in all such cases if the origin eventually becomes known it turns out to include design.
Conclusion
There are two main objections to the ID argument:
A) There are deterministic explanations for life.
B) Even if there were no deterministic explanations it would not follow that life was designed
For the purposes of this discussion I will pretend (A) is false and focus on (B)
No one disputes that it is possible to detect design. The objectors to ID just believe that B) true. The correct way of detecting design is to compare a specific design hypothesis with alternatives and assess which is provides the best explanation. This includes assessing the possibility of the designer existing and having the motivation and ability to implement the design. If no specific hypothesis is available then nothing can be inferred.
So is the dFSCI claim above true and if so does it provide a valid alternative way of detecting design?
The trouble is that there is dearth of such situations. One of the reasons for this is that digital strings do not exist in nature above the molecular level. At any other level it is only a human interpretation that imposes a digital structure on analogue phenomena. The characters you are reading on this screen are analogue marks on the screen. It is you that is categorising them into characters. So all such strings are created by human processes. It follows that design is a very plausible explanation for any such string. People were involved in the creation and could easily have designed the string. If you add the conditions that the function must be prespecified and there should be no causal relationship between the function and the string then design is going to be by far the best explanation. It goes further than that. It also means there almost no real situations where someone is confronted with a digital string without knowing quite a bit about its origin – which is presumably why Gpuccio can only point to examples created/selected by bloggers.
What about the molecular level? Here there are digital strings that are not the result of human interpretation. Now human design is massively implausible (except for a few very exceptional cases). The problem now is that carbon chains are the only digital strings with any kind of complexity and these are just the one’s we are trying to evaluate. There are no digital strings at the molecular level with dFSCI except for those involved in life.
So actually the dFSCI argument only applies to a very limited set of circumstances where a Bayesian inference would come to the same conclusion.
Mung 548
I wrote:
You wrote:
If you read what I wrote carefully, I didn’t say the definition resulted in the statement. I said it resulted in the statement being circular. Obviously definitions can make statements circular. If the statement is “All As are Bs” and I define an A as “not something different from a B” then the statement is circular.
I am examining the conditions when there is inference in Gpuccio’s reasoning and when it is circular.
Gpuccio
The same sticking point comes up again and again. I thought Keiths had made it crystal clear – but as I often said the judge of clarity should be the reader not the writer. This becomes not so much a debate as a challenge to try and explain something. So let’s try again. One of your responses is:
Yes -but not all groups know the same things. Therefore they do not have the same “present state of knowledge”. This difference in knowledge need not correspond to a scientific breakthrough. It could simply be a realisation that the laws of science can be applied in a particular way to that string. For example, a group may simply not realise that this sequence:
01101110110001101
can be generated quite easily by a natural process. Assumng it was much longer and there was a function they might apply the dFSCI procedure objectively and correctly and deduce there was dFSCI. Are they wrong?
It was you that put the word “known” into the definition. There must be a person or group that does the knowing. So you have necessariy made dFSCI relative to that group’s knowledge.
Gpuccio 561
All digital strings can be converted to a decimal number. That is not the point. I am not talking about a physical thing that happens to be expressed as a number. I am talking abut the number itself.
Who designed the number 3 and when did they do it?
But only to a human reader. In software, just as in biology, the runnable code, is native to the run-time processes at the physical level, i.e. it is not the descriptive source code.
But that is the point, that there is a “translating apparatus” already in place that understands “codes” that exist only in a very particular “chemical set”.
In order to do what you suggest, you would have to redesign an “apparatus” that exists today in biology that already does the “translation”.
At that point, you could still not “arbitrarily” use two different codes at the same time as you can with language.
When I was growing up, we spoke German, Yugoslavian and English at home.
Conversations tended to end in a different language than they started with, which was only acceptable because we could all encode and decode all three sets of “communication codes” at the same time.
Human language is different than what is represented in DNA because it is used for a different purpose.
Good one! 🙂
You are confusing the map with the territory.
Computers at the physical level operate with electrical charges on gates and current flow.
Biology does the same at the physical level as there are chemical processes involved.
On a computer motherboard, I cannot mix 3.3V and 5V devices because a voltage level that is logically high for one device might not be for another, and there is no guarantee the device will actually survive the mismatch without being stressed to the point of failure.
In biology, a different chemical makeup would result in different “life” and this means that the DNA “code”, which is chemical and not digital, is not arbitrary.
Empirically, we have never seen different “translation devices” in humans, other than DNA as it exists.
It strikes me that a lot of time and energy is being wasted over the term arbitrary.
With respect to the DNA code, arbitrary has at least two useful meanings.
One is that none of the bases are particularly favored by chemistry. Hence, mutations do not favor one or the other.
The other meaning is encapsulated in the phrase “frozen accident.” Which more or less means that the specific code and interpreter could have been different.
Gpuccio
I am going to drop out of this debate about whether dFSCI is relative to an observer’s knowledge. It has become circular and barren. I also withdraw my assertion that the statement
“Everything dFSCI is designed” is not circular.
I thought was true but I misunderstood your definition of dFSCI. In the light of what you have written over the last couple of days I believe that you use the term dFSCI in practice in such a way that the statement is circular.
It could not have been any worse for your side! 🙂
You came up with a design detection tool called “dFSCI”, that while independent of origin in your definition, was instead in practice used to exclude any means of attaining “dFSCI” by any origin other than design.
So “dFSCI” gets put on the scrap heap along with the Explanatory Filter, CSI and IC.
Try promoting the positive evidence for ID instead of constantly attempting to discredit “Darwinism”.
Until then, the onlookers are still short an ID mechanism.
gpuccio calls a halt (#578):
From my point of view here’s where we ended up:
All of which has caused me to lose interest in the dFCSI criterion, as there is no theoretical framework for it.
This may have been remarked before:
While CSI (Complex Specified Information) is purportedly a measure of information in bits, dFSCI (digital Functionally Specified Complex Information) is *not* a measure of information, but an assessment of sorts. This is one of the many areas of confusion.
dFSCI doesn’t qualify as a theory, though, even if it were valid in other respects.
Joe,
You and I have both said that “dFSCI” is not dependent on its origin yet gpuccio assesses “dFSCI” to a string only after he checks its origin.
Have you changed your mind?
Joe, read this response to gpuccio.
That means you cannot logically make the following conclusion.
“2″ must be assessed before “1c” if “dFSCI” is not dependent on its origin.
“dFSCI” cannot be qualified on the basis of its origin.
This what frustrates me, that I can get you to agree that “dFSCI” is not dependent on its origin, but as soon as you use it as a tool, you tell me that you check the origin before declaring a string has “dFSCI”.
Joe and I both believe “dFSCI” is independent of origin and it must be in order for the string to qualify as “testable for design”.
gpuccio,
Even when testing sample strings, you cannot exclude any known non-design mechanisms from being the origin of “dFSCI”.
If you run across a non-design mechanism that generates a string that you would consider has “dFSCI”, it simply means that “dFSCI” is not a design detection tool.
“dFSCI”, as any onlookers can see, is not falsifiable when you take the following words from gpuccio himself into account.
Before a string gets the attribute “dFSCI”, its origin is checked.
According to this logic, “dFSCI” is dependent on its origin and thus by definition, and by gpuccio’s whim, no string will ever have “dFSCI” that is generated by a mechanism he doesn’t like.
To all budding young scientists, this is not how science is practiced by actual scientists who put science ahead of their personal religion’s creation stories.
To all budding young lawyers, keep this in mind when you see another Dover unfolding.
There is no need to make an “attempt” because anyone can write a GA that will generate a functional string of any arbitrary length.
The simplest demonstration would be the traveling salesman problem, for which the solution is long, and for which there is no good alternative method of solution.
The solution is functional in a most basic sense of the word: it is worth money.
There are many other commercial uses of GAs. If the solution were programmed into them, it would be simpler and cheaper simply to use the solution, rather than run the program
What gpuccio does not want to admit is that evolution is that evolution is an efficient designer in many situations.
I would encourage Mung to write a GA. It’s an interesting programming task, and may be instructive in terms of gaining an understanding of what faithful representations of evolutionary mechanisms do to populations of replicating strings. The relevance of it to biological considerations may (or may not) become apparent as part of the process.
If you don’t or won’t understand the evolutionary process, you are in a position neither to critique it as an explanatory mechanism, nor to declare it incapable of functioning as a ‘necessity mechanism’ alternative to active string design.
Start with strings of zero length, and keep tabs on descent from a collection of such null-strings numbered 1 to n. This is NOT an analogue of a DNA string of zero length, nor the OoL, it is an analogue of a replicator that can do no more than merely replicate, no better or worse than any other. Copy and kill them at random … and one of them will become the ancestor of all, guaranteed.
Then introduce methods that add and change bits, recombination, and internal duplication of string segments and a fitness function operating on these now non-null bits of ‘extra’ string. You are randomly patterning your replicators, introducing non-critical function which nonetheless affects fitness differentially. Your strings adapt to the prevailing conditions, as if designed to fit.
Gpuccio referred me to Kirk Durston’s paper as supporting evidence of his ideas on dFSCI though it only talks about FSC (functional sequence complexity). I note Wesley Elsberry, Larry Moran,PZ Myers and Jeff Shallit have all been less than complimentary about Durston but I can’t find much on the web specifically about FSC. The cites nearly all involve David Able and the paper seems to have been otherwise largely ignored.
I see PZ has a go at FSC here.
ETA
Here is Hazen’s paper tha PZ mentions.
It is not falsifiable due to gpuccio’s method of determining “dFSCI”.
1) Determine probabilities above UPB.
2) Determine the origin of the information.
3) Withhold “dFSCI” positive for those specific origins gpuccio asserts cannot generate “dFSCI”.
So if gpuccio does not accept “dFSCI” as a possibility of specific processes he has excluded, the string does not have “dFSCI”, regardless of the actual “digital content” of the string.
“dFSCI” must not be dependent on its origin, period.
Toronto
3) Withhold “dFSCI” positive for those specific origins gpuccio asserts cannot generate “dFSCI”.
I think it’s more like this: withhold the designation “has dFCSI” from those cases where, based on gpuccio’s current knowledge, the high functionality is judged to possibly have arisen by “deterministic” origins that include natural selection.
That does mean that dFCSI is then not simply a property of the string that can be judged without inquiring into its possible origins.
(However, gpuccio was firm that the assessment was based on current knowledge and that if it later was discovered to gpuccio’s satisfaction, even in one case, that “deterministic” forces such as natural selection could account for the string’s high functionality, then gpuccio’s whole method would collapse).
Of course there is no theory that renders this impossible. So there is no reason to believe that the whole method works.
gpuccio is very melodramatic. He’s flounced out of the discussion before, only to return later. It may happen again.
(Note: I wrote most of this comment, including the part above, earlier in the day. Gpuccio has since flounced back in.)
On the other hand, he may finally be realizing that the argument from dFSCI just isn’t defensible.
dFSCI is a hunch dressed up like a scientific concept. It has many problems:
1. Gpuccio’s hopes for dFSCI seem to have been much like Behe’s hopes for irreducible complexity. Both were supposed to be objective indicators that something could not have evolved, and that it therefore must have been designed.
2. Not only was dFSCI supposed to be an objective indicator of design, but gpuccio chose to follow Dembski’s example and measure it in bits. That made it sound quantitative and sciencey.
3. Someone who hasn’t looked under the hood might actually get the impression that dFSCI is useful: that you can assess it objectively, and that the result tells you whether a sequence was designed. No more arguing about whether something could have evolved. You just assess the dFSCI of the sequence and the question is settled.
4. That’s the fantasy. The reality is much less impressive. You start off with a sequence — the hemoglobin gene, for example. You know that it didn’t arise through pure random variation without selection, and you know that it has a function. But could it have evolved? You think so, but gpuccio doesn’t.
5. You decide to apply gpuccio’s dFSCI argument to the question. Gpuccio first asks if the gene has a function. You answer in the affirmative, and you and gpuccio come to an agreement on what the function is (a procedure that is fraught with ambiguity — but it doesn’t really matter, as we’ll see later).
6. You and gpuccio somehow magically determine the size of the target space relative to the space of all possibilities. This is also fraught with ambiguity and technical difficulty, but again, it doesn’t matter.
7. Gpuccio takes the negative log base 2 of the ratio of the target size to the total possibility space, and gets a dFSI number (in bits). He compares it to the threshold and finds that it is greater. That means that the gene could not have been produced by pure random variation without selection within the time available.
8. So far all of the work you have done is useless. You already knew that the gene had a function and that it didn’t arise by pure RV. The dFSI number you came up with just confirms this obvious fact. You throw away the number, because you’ll never refer to it again. So much for the quantitative and sciencey part. It tells you nothing that you didn’t already know.
9. Now you get to the third criterion of dFSCI: can the gene be explained by a ‘necessity mechanism’, such as natural selection? You say it can, but gpuccio says it can’t. You go with his assessment, since he is the dFSCI “expert”. Gpuccio now excitedly tells you that this means that the gene has dFSCI, and that it must have been designed.
What does this all boil down to? What does it mean to say that the hemoglobin gene exhibits dFSCI? It means that it has a function, that it couldn’t have arisen via pure random variation without selection, and that gpuccio thinks it couldn’t have evolved.
What do you know now that you didn’t know before the analysis? Nothing. You already knew that the hemoglobin gene has a function. You already knew that it couldn’t have arisen via pure RV. You already knew that gpuccio didn’t think it could have evolved.
The entire exercise was a waste of time. dFSCI is useless, and you still don’t have an answer to the real question, which is “Could unguided evolution produce the hemoglobin gene?” All you have is gpuccio’s opinion, plus his insistence that he is always right when judging these questions.
As I said, dFSCI is a hunch masquerading as a scientific concept. It’s fascinating that gpuccio either can’t see this or won’t admit it.
Not sure this is entirely fair. There are two stages to the ID argument:
1) Evolution can’t explain life.
2) Therefore life was designed.
Gpuccio clearly does not accept (1). But as I see it the dFSCI argument is all about (2) – if evolution were unable to explain life then it is reasonable to deduce design. It is a pseudo-mathematical attempt to justify the argument:
If there is no “known” explanation for functional digital strings in life then there is good reason to believe it is designed
This is supported by suggesting that outside the domain of life when we come across “functional” digital strings and there is no “known” explanation then it always turns out they are designed.
This is made true by being so vague about the meaning of “known” and “functional” so it becomes circular.
Gpuccio 592
You raise an interesting point which I think is not a replay of previous barren discussions (I am sure we will both drop it quickly if it develops that way).
1) An explanation is more than just knowing physics, biochemistry, probability and so on. You can be an Einstein in all of these and fail to realise how they apply to a specific situation. A world’s best geologist may not understand why a particular earthquake happen in a specific pattern. Someone less expert may be aware of a particular fault line that is the cause or have a flash of inspiration as to how the pattern might have occurred. This is nothing to do with failing to apply the dFSCI procedure properly. It is simply one person knowing there is a necessity mechanism and another not.
2) You want to extrapolate the results of applying dFSCI from the world of man-made digital strings such as computer software and written text to molecular strings in life. This is a massive jump. That is why it is relevant to consider whether the procedure would apply in other widely different situations where we know the origin e.g. thousands of years ago.
Joe,
I cannot believe, and probably this goes for all the lurkers reading both sites, that you have backed away from the “dFSCI”/origins relationship.
Unless I am mistaken this is the way the process should work:
1) Assess the string for digital content that represents specific functionality and complexity.
2) If above a threshold measured in bits, assert that the string has “dFSCI”.
Is this what should be done?
Does gpuccio withhold an assessment of “dFSCI” depending on its origin?
gpuccio responded to me:
Will not play. gpuccio does not get to be the one who adjudicates what is a model of natural selection and what isn’t. I’ve been modeling natural selection for about 50 years, and have written many papers. I’ve known, studied under, or disputed with many of the giants in theoretical population genetics (Jim Crow, Dick Lewontin, Sewall Wright, Motoo Kimura, Masatoshi Nei, Tomoko Ohta, Sam Karlin, Alan Robertson, etc etc). I published the only bibliography of theoretical population genetics ever published, in 1981 (at that time the field was 7,982 scientific papers — it will be considerably larger by now). I have taught courses regularly on theoretical population genetics since 1969.
I have a free online e-book which is the text for my course. Perhaps gpuccio can read chapter II (Selection) and tell me where the equations go wrong and what would be better ones.
I have some history with genetic algorithms too:
If gpuccio makes constructive suggestions about how to model natural selection, or how to improve genetic algorithms, fine, we can discuss those. But if gpuccio airily dismisses an entire 100-year-old field with over 10,000 theoretical papers (and even more experimental ones), then I don’t know of any reason to take that dismissal seriously.
I have posted at TSZ an example of a model in which it can be shown that Dembski’s SI can arise by natural selection. I will not get into an endless loop where I post more models, or run more GAs, and each time gpuccio dismisses them as not natural enough selection, or not really what gpuccio meant by dFCSI.
No, the ball is in gpuccio’s court.
I am waiting for some theoretical framework (equations, simulations) showing that dFCSI cannot arise by natural selection. Lacking that, I don’t find dFCSI of any use.
gpuccio,
Does Joe’s string contain “dFSCI”?
Clearly, it is not a product of an intelligent mechanism and yet something is responsible for its origin.
I am as guilty as anyone, but I propose we try hard to resist the temptation to make personal comments – even about Joe! That is what distinguishes TSZ from AtBC and UD. I know it is hard when you receive a steady stream of verbal abuse but anyone browsing this forum will probably not even know about that. There is always AtBC to express our feelings.
But we’ve been waiting, for years and years, yet all we get are different versions of, “We don’t understand how that happened so God must have done it”.
Well said and noted.
At least all the lurkers on both sides can see their failed attempts at debunking evolution.
But your #2 is just another of gpuccio’s hunches: the hunch that if gpuccio doesn’t know of a ‘necessity explanation’ for X, then X must be designed.
The following two arguments are logically identical:
Argument I:
Argument II:
Argument I sounds objective and sciencey, but the logically identical argument II shows that when you take a closer look, all you find are two unjustified hunches leading to an unjustified conclusion.
At forums like UD you see repeated assertions that if we find CSI (or something), that there is no way that this could have been put into the genome by natural selection.
It’s like Gertrude Stein’s description of Oakland, CA: When you get there, there’s no “there” there. The UD commenters have not realized that since Dembski’s Law of Conservation of Complex Specified Information collapsed, they are standing on thin air.
Actually, its simply a replication with some duplication or removal of “HAHAHA”s of a comment I made about one of his musings a long time ago…
What amazes me is the continued absence of Dr Dembski from UD and the continued absence of any question as to why from any UDite.
Sure, he pops by to plug something once in a while but what his opinion on dFSCI is I really don’t know.
I doubt he’s the patience to find “the” definition.
Unlike some here! Good job people, the veil of obscurity has been pierced and surprise surprise there’s nothing there at all.
Since GP rejects all models of evolution due to their inevitably being non-‘natural’, it is hard to see how anything other than a very long wait could meet his stringent criteria. Even then, I think the hope would be that we’d waited so long we’d all forgotten what we were looking for.
dFSCI is only a slightly-modified version of Dembski’s CSI, which he seems to have abandoned in favor of “active information” and the “search for a search”. I doubt that Dembski has much interest in the invention of new acronyms and minor tweaks for an already-discredited concept.
Just to be sure, are you agreeing with Joe and myself, that “dFSCI” is not dependent on its origin?
You should be more humble Mung, considering that the last time someone on this side posted source code, I had to explain it to you.
You should also be more humble when it comes to anything you’re not really equipped to discuss.
This includes biology and computer hardware.
That is what lost your side a Dover victory, the belief that you could simply wave away those things that evolutionary scientists obviously didn’t understand.
Who needs an argument based on biology when you can simply count bits?
Mung,
There is the confusion.
If a), b) and c) are all true, you withhold asserting “dFSCI”, until you examine how that string was generated according to gpuccio.
So Joe will accept “dFSCI” with no consideration of origin and gpuccio won’t.
Let me answer Mung’s unusual comment which responds to me:
Mung: I’m impressed Joe, really. Post your source code.
It’s in some listings in a drawer in my office. In FORTRAN II. I can’t see what purpose there would be in posting it.
Mung: Post the source code. Let’s see how close your GA was to the model.
The source code for a program my thesis advisor wrote and which he described in a 1959 paper? Five years before I worked with him? And what GA of mine was supposed to simulate that model? Mung’s request makes no sense at all.
Mung: Fantastic! Post the source code.
Mung now wants me to post Nils Aall Barricelli’s source code for his 1954 evolution simulation on the Institute for Advanced Studies computer. Guess what? Back then there wasn’t any source code — programming was done punching digits into cards. And of course I don’t have any of those programs, do I? I met Barricelli 14 years later — that doesn’t mean he gave me his early programs.
Mung: Post your unpublished paper that shows how your GA employed an actual model of evolution.
What unpublished paper? What does Mung mean by “actual model of evolution”? (I have a feeling the answer won’t be illuminating).
Mung: Congratulations. What do you know of artificial life? Did you present a paper? Was it published?
I was there to talk about modeling nonartificial life. I described a model of long-term evolution I had earlier published in 1978 in American Naturalist. They must have thought that this would be relevant to artificial life. I think I’ll leave that judgement to them rather than to Mung.
Mung: In your teaching program, what is it that natural selection GENERATES?
Beats me. Darned if I know what Mung means by “generates”. Mung’s term, not mine. My program has generations, so I guess it generates them.
Mung: YIKES! Arbitrary overload!
I thought you said natural selection was deterministic.
Could you please, in your spare time, explain the meaning of arbitrary to onlooker?
Yes, of course. It is a teaching program, to teach people what the outcome of various possible evolutionary forces is. So it allows the user to put in different values of fitnesses, of mutation rates, of migration rates, and of population sizes. These are arbitrary because the user gets to pick them and put in whatever values they want to try.
Next Mung will demand my source code. Actually in that previous comment I gave a link to that program’s web page. You can download executables, and there is a link there to another page which lets you download source code.
I don’t think I’m going to pay much attention if Mung goes around declaring that my programs, or anyone else’s don’t model “actual evolution”. Models never describe actual reality. They make useful simplifications of it, something which the commenters at UD refuse to accept.
I’d be more impressed if Mung could tell us how many times Mung has modeled evolution, either in simulations or by making mathematical models of it.
Perhaps Mung or some other ID advocate would explain how the UPB or dFSCI or the search for a search models evolution.
I think Mung is indulging in the popular internet pastime known as ‘trolling’. The programs don’t generate actual organisms? Oh no! Likewise, epidemiology models don’t generate actual illness, meterological models don’t produce actual weather …
I don’t think anyone, least of all a mathematician, makes that claim. NS denotes a tendency; a bias in the frequency distribution compared to a truly neutral trait due to some consistent differential between the trait and its alternatives on mean survival and reproduction. Positively selected traits win more often; deleterious traits lose more often, but don’t offer a guarantee. Apart from lethal mutations, perhaps.
gpuccio confirms his dFSI, dFSCI are the same as Durston’s (and presumably Hazen’s) FSC (functional sequence complexity). So why do we need dFSCI as a term if we already have FSC?
In the simple models of evolution that population genetics uses, natural selection is a deterministic force. But so is migration, even mutation. Genetic drift is the only random force. If the population size is infinite, then even mutation is deterministic at the level of population gene frequencies, because every possible mutation (and even every possible combination of mutations) will occur at exactly its expected frequency. (Of course that would take a population size bigger than the number of particles in the universe).
Of course at the level of individuals these events are all modeled by population genetics as random events with probabilities.
And there are more sophisticated models where selection is allowed vary randomly from generation to generation even when we calculate its effects on gene frequencies.
It was gpuccio, not me, who insisted in this discussion that natural selection is to be thought of as a deterministic force. I went along with it for the sake of the discussion.
Well, it can certainly be modelled deterministically as a simple exponent, but surely that doesn’t mean that’s what it is? Even with a static nonzero fitness differential, there is drift, and the two contributions to frequency change can’t be separated, except in the infinite-but still-randomly-assorting case or the ‘truly’ neutral one, in which case it’s either all one or all the other. The models you linked, for example, showed repeat runs distributed around a central trend.
I’d see a kind of ‘pseudo-determinism’ in operation, akin to convergence on LLN expected values as sample size increases. As time progresses, there is increasing likelihood that the random fluctuations due to sample error with typical small s values cancel out and leave genomes enriched in beneficial alleles – an adaptive arrow. Any individual allele can go against the expectation, but the generality of repeat runs generates adaptation, even from s values effectively neutral on current population size – evolution is random but biased.
In the past when I’ve asked how to calculate dFSCI I’ve been told that Durston has done it already so go read his work.
I suspect the reason for the invention of further terms is related to the conclusion:
Nothing there about “Therefore Design”. So they have to invent it.
gpuccio,
Where have I lied here?
That means you cannot logically make the following conclusion.
The following are your words gpuccio.
“2″ must be assessed before “1c” if “dFSCI” is not dependent on its origin.
“dFSCI” cannot be qualified on the basis of its origin.
This what frustrates me, that I can get you to agree that “dFSCI” is not dependent on its origin, but as soon as you use it as a tool, you tell me that you check the origin first.
Joe and I both believe “dFSCI” is independent of origin and it must be in order for the string to qualify as “testable for design”.
Where is the lie?
Perhaps for some people it’s easier to call somebody else a liar then reconsider your own position critically.
For some strange reason, you have managed to call not only myself but others on this side liars when we reflect back to you our understanding of your communications to us.
Go back and read all these communications very carefully.
No one has lied, regardless of your emotional interpretation of our clear statements in trying to make sense of your “dFSCI” mechanism.
You, ..gpuccio…, will not assert “dFSCI”, …meaning you will withhold your “dFSCI” label, from a string, not because the string is not functional or complex enough, but because of its origin, i.e how the string “came to be”.
There is no lie there, it is simply my understanding reflected back to you, of cumulative statements made by you.
Toronto,
Gpuccio is drawing a distinction between the actual origin of a string and possible explanations for that string.
Consider the following string:
The actual origin of that string is design. I thought of it and typed it in. However, the string can also be explained by a simple “necessity mechanism”: a clocked D flip-flop whose Q output is connected to its D input via an inverter.
In assessing the dFSCI of that sequence, gpuccio does not look at its actual origin, which in this case is design. Instead, he looks for possible explanations, and finds the one I suggested (or some other comparable “necessity mechanism”).
This means that according to his definition, the string does not exhibit dFSCI, even though we know it was designed. The origin doesn’t matter.
I have no idea why gpuccio sputters and casts aspersions rather than calmly correcting your misunderstanding. He seems pretty volatile.