Home » Informatics » The MathGrrl files: Reestablishing what we know

The MathGrrl files: Reestablishing what we know

MathGrrl’s friends have been discussing her recent post( here), on measuring complex specified information, which garnered 324 comments and counting.*

Not being a mathie, I couldn’t follow most of the discussion here, but certain turns in the discussion reminded me of something I’d heard before:

If design is a part of nature, then the design is embedded in life as information. But many people are not used to thinking in terms of an immaterial quantity like information. As G.C. Williams writes: “Information doesn’t have mass or charge or length in millimeters. Likewise, matter doesn’t have bytes. You can’t measure so much gold in so many bytes. It doesn’t have redundancy, or fidelity, or any of the other descriptors we apply to information. This dearth of shared descriptors makes matter and information two separate domains of existence, which have to be discussed separately, in their own terms.”- quoted in By Design or by Chance?. P. 234.

The reason the materialist doesn’t see how information can’t just “appear” is that materialism, which undergirds everything else he believes, never took information into account. So he can treat it as magic, as something that “just happens.”

Meanwhile, perhaps junior ID theorists cannot formulate a single definition of complex specified information at present. But it really wouldn’t matter if they could. The materialist would just blink and say, “I’m sorry. This is so confusing.”

I’ve been through this with enough different issues to know that that would be the outcome, for sure.

So, for ID theorists, the goal is not convincing such people or reaching an impasse with them, but formulating definitions that actually lead to new discoveries or clearer understanding of current ones.

A similar thing happened in Isaac Newton’s day, when Newton’s equations for gravity were rejected because they involved action at a distance. And that wasn’t allowed. His “laws” were accepted anyway by working scientists and engineers because they enabled accurate calculations. There is no other way it could have happened.
* And counting: Our practice is to close comments after thirty days, so there’s still time to make a contribution.

Next week, I hope to present an interview with Jonathan Wells on junk DNA.

  • Delicious
  • Facebook
  • Reddit
  • StumbleUpon
  • Twitter
  • RSS Feed

85 Responses to The MathGrrl files: Reestablishing what we know

  1. 1

    O’Leary

    His “laws” were accepted anyway by working scientists and engineers because they enabled accurate calculations. There is no other way it could have happened.

    This is exactly the opposite situation to which CSI finds itself.

    If CSI becomes something that provides results that can be obtained no other way and the results of which can be verified then it will revolutionize the science of design detection.

    Yet that simply is not the case. The mechanisms presented so far for CSI do not seem to enable accurate calculations.

    Meanwhile, perhaps junior ID theorists cannot formulate a single definition of complex specified information at present.

    What, you mean people like Salvador Cordova who has been a poster at UD for many years and who has many posts and claims regarding CSI to his name? I’d hardly call him “Junior” and his contribution to the thread was simply a passing comment. Perhaps you could drop a note to Dr Dembski and ask him to formulate (and show how such is determined) the values of CSI for the 4 examples given?

    The point is O’Leary that the calculation of CSI has been claimed by the ID camp to be something that not only can be done but which unambiguously indicates intelligence in the design of the object being assessed and which can easily be done.

    Yet we’re now seeing something different. There are as many versions or ways to determine CSI as there are ID proponents. And only one of them has even attempted to address the scenarios presented and the outcome of that was that CSI can be generated by naturalistic processes such as gene duplication. Hardly a victory for ID.

    So, for ID theorists, the goal is not convincing such people or reaching an impasse with them, but formulating definitions that actually lead to new discoveries or clearer understanding of current ones.

    Exactly so. Mathematicians are convinced via mathematical means.

    The best way to convince that CSI has real meaning and really indicates design is for it’s calculation for the examples given to be demonstrated in a open, clear and repeatable manner.

  2. Hello, JemimaRacktouey at 1,

    Sal Cordova would doubtless be thrilled by your assessment of his status. This was the first I’d heard, and it is not any reflection on his arguments to say so.

    I am told that many scientists did reject Newton’s description of gravity, because it involved action at a distance, and they rejected that in principle, which coloured their thinking.

    Actually, I can’t prove 2+2=4 where political or other sentiment is against it. I know, I’ve tried.

  3. Meanwhile, perhaps junior ID theorists cannot formulate a single definition of complex specified information at present. But it really wouldn’t matter if they could. The materialist would just blink and say, “I’m sorry. This is so confusing.”

    Perhaps so, but my concern is that, on the other blog post, that the ID scientists seem to be using the skepticism they are facing as a reason to not do anything at all vis-à-vis coming to a single definition.

    The acceptance of an idea seems to be proportional to it’s utility in practice. And if the ID scientists aren’t willing to clearly define CSI and show it in action, there is at least some warrant to skepticism.

  4. The ID crowd need to do better than this. Can we at least answer the following questions:

    1. If CSI is a function F, what is it’s domain and codomain? Sets of bit strings?

    2. What describes the mapping from the domain to the codomain? How does F tell us about intelligence?

    3. Don’t discussions about genetics obscure the issue at hand, which is how to establish whether a sequence is the product of an intelligence?

    4. Can I post a bit sequence and have somebody tell me whether it evinces design? Am I asking the right question?

    I am not antagonistic to ID’s aims or claims… and I always remain open to persuasion. But I will not be persuaded until these very basic issues are addressed—at least not on the CSI front.

    Lengthy, rhetorical posts that refer to other work and attack ID sceptics in withering tones do nothing to shore up my confidence in ID.

    At least answer question 1. In one line if you can :-)

    (Let me close by remarking that I am pleased this forum exists, having followed it for a few years, and I wish all those who post on it the all the best.)

  5. 5

    O’Leary

    I am told that many scientists did reject Newton’s description of gravity, because it involved action at a distance, and they rejected that in principle, which coloured their thinking.

    Of course, but ultimately it won out because Newton’s ideas could be expressed in a way that everybody could apply themselves and generate the same results from the same input. If the idea had been rejected because of “action at a distance” and because it did not generate useful results then they would have stayed rejected.

    It remains to be seen if CSI will be placed in the same category. I don’t think the mainstream rejection of CSI is on principle, I think it’s rather that it cannot as yet be shown to consistently generate useful repeatable results and be applicable in a range of situations.

    Newton’s laws are applicable in a vast range of circumstances. CSI, no so much, not yet anyway.

  6. For me the interesting point in the MathGrrl thread is that they are even squabbling over the precise value of CSI in different situations in the first place, for I remember the time not so long ago that many armchair Darwinists, and even professional Darwinists, refused to even admit that ‘information’ resided in the cell. So I guess it is great progress for Darwinists to be forced to argue that the Information we find in cells does not really matter since a precise definition of CSI for different situations is so difficult to ascertain.,,,

    Now that darwinists have at least admitted information is in the cell, Perhaps we can now work on getting Darwinists to admit that there is ‘quantum information’ in the cell which cannot possibly be reduced to the materialistic framework of Neo-Darwinism:

    Confirmation of quantum information in proteins
    http://www.uncommondescent.com.....ent-375209

    and in DNA
    http://www.uncommondescent.com.....ent-374898

  7. CSI- Complex Specified Information.

    Information- see Shannon, Claude

    Specified Information is Shannon Information with meaning/ function

    Complex Specified Information is 500 bits or more of specified information

    MathGrrl wants a mathematically rigorous definition of CSI and I say that is like asking for a mathematically rigorous definition of a computer program (which contains CSI).

    The mathematical rigor went into calculating the probabilities that got us to 500 bits of SI = CSI.

  8. Cause and effect relationships-

    Everytime we have observed CSI and knew the cause it has always been via designing agency. We have never observed nor do we have any experience with Mother nature producing CSI.

    Biological specification always refers to function. An organism is a functional system comprising many functional subsystems. In virtue of their function, these systems embody patterns that are objectively given and can be identified independently of the systems that embody them. Hence these systems are specified in the same sense required by the complexity-specification criterion (see sections 1.3 and 2.5). The specification of organisms can be crashed out in any number of ways. Arno Wouters cashes it out globally in terms of the viability of whole organisms. Michael Behe cashes it out in terms of minimal function of biochemical systems.- Wm. Dembski page 148 of NFL

    Biological functionality is specified information.

    So what do we have to do to see if it contains CSI? Count the bits and figure out the variation tolerance because if any sequence can produce the same result then specified information disappears.

    And again, CSI is all about origins…

  9. Great, Joseph. How many bits are in a Jackson Pollock painting? Or a snowflake?

  10. Jemima,

    No. CSI is easy to recognize. It may (for sake of argument) not be amenable to calculation yet. But as I’ve said before, light was recognizable before it was calculable (with precision).

    It does not take math to recognize light and it does not take much math or reasoning at all to recognize specified complexity. It’s a rather simple concept.
    Please read Dembski’s work. For a primer: http://creationwiki.org/Specified_complexity
    You can follow the link below for further information.

  11. 11

    Collin,

    No. CSI is easy to recognize. It may (for sake of argument) not be amenable to calculation yet. But as I’ve said before, light was recognizable before it was calculable (with precision).

    Hmm. Sounds very much like obscenity then. You can’t say exactly what it is but you know it when you see or hear it.

    nd it does not take much math or reasoning at all to recognize specified complexity

    Yes yes, but the point of the thread(s) is to calculate CSI, as per claims of ID supporters over many many years, for a couple of examples it should be amenable to were it actually as developed as the many and varied claims over the years have made it out to be.

    And it’s interesting to see how that’s actually turned out.

    MathGrrl, if you are looking for another topic after the CSI thread plays out (as noted by O’Leary it has a limited lifespan) then perhaps I could suggest the Explanatory Filter?

    http://www.conservapedia.com/Explanatory_filter

    Like CSI many claims over the years have been made for it but there are no real examples of it in action.

    Perhaps the EF would generate a similar level of interest as the recent topic of CSI has done?

    I’ve learn alot about ID from that thread.

  12. Jemima Racktouey at 11: I doubt you learned anything you didn’t already think.

  13. 13

    O’Leary,
    On the contrary. I’ve learned a great deal.

    For example, the next time somebody makes a claim involving CSI I’ve learned that I can simply link them to that thread and ask them if they can apply their claimed knowledge of CSI to the examples in the OP.

  14. Muramasa:

    Great, Joseph. How many bits are in a Jackson Pollock painting? Or a snowflake?

    I covered that in the thread that this thread is about.

    CSI is only good in places were the observed is readily converted into bits.

  15. I have no idea if MathGrrl knows what she’s talking about. I do hope she’ll offer some summary comments.

    As the discussion winds down, however, MathGrrl’s opponents don’t fill me with confidence. None of them sounds like a scientist. Now, as an amateur observer myself, I have nothing against amateurs. But the discussants are flailing around, reading papers for the first time, questioning motivations, and throwing up any number of distractions. The level of disagreement among ID supporters about basic concepts is remarkable.

  16. JR:

    For example, the next time somebody makes a claim involving CSI I’ve learned that I can simply link them to that thread and ask them if they can apply their claimed knowledge of CSI to the examples in the OP.

    The examples in the OP are bogus for the reasons provided in that thread.

  17. Joseph,

    So CSI (or specified complexity), upon which Dembski has premised his Law of Conservation of Information is not universally applicable? Huh.

  18. uramasa:

    So CSI (or specified complexity), upon which Dembski has premised his Law of Conservation of Information is not universally applicable?

    It could be. You would have to ask him.

    If the point of CSI is to indicate design (which it is) I can do that with a painting without having to know anything about CSI.

    That is all I m saying…

  19. JemimaRacktouey -

    I agree that CSI is difficult to calculate. A lot of work in the last decade of ID research has been into coming up with more concrete ways of calculating functional information. Durston, for instance, has published his method – http://www.tbiomed.com/content.....2-4-47.pdf

    The way that I think that is best is Dembski/Marks’ Active Information concept, which basically measures the distance between the expected result according to NFL theorems and the actual result obtained by experiment.

    I’ve done the Active Information calculation for the somatic hypermutation system for antibodies, and I am working on a more general method for performing the calculation, which can be used to see if hypermutable states in bacterial adaptive mutagenesis relies solely on mutation rate, or if it also relies on mutational specificity.

    You can see my work on somatic hypermutation here:

    Abstract

    Slides

    Poster

  20. Jemima, IT IS DEFINED! Why don’t you at least read Dembski?

  21. Jemima,

    Sorry for shouting and being rude. That was unnecessary and I can’t delete my comments like on Facebook.

    I think that VJTorley gave calculations of CSI. What was wrong with those?

    I specifically stated that I concede that CSI may not be calculable. So what’s the deal? I was just making a side point that it is recognizable and that it should lead to a common sense inference (not proof, necessarily) of design. That is why Dembski calls it the Design Inference not the Design Proof.

  22. 23

    Collin

    Jemima, IT IS DEFINED! Why don’t you at least read Dembski?

    Sure, but all those other people on the thread have read Dembski too and they cannot calculate (or most of them) CSI for the examples.

    I think that VJTorley gave calculations of CSI. What was wrong with those?

    Nothing at all. And he apparently showed that CSI can increase from gene duplication events. Which is nice.

    But I’d guess given all the different opinions we really need a couple of people to have a go at a couple of the examples to get a really good idea of how to determine CSI for any arbitrary system.

    I specifically stated that I concede that CSI may not be calculable.

    That’s nice. But others insist it is. And it’s those others who should really have a go at it.

  23. Johnnyb,

    I think that what makes CSI difficult to calculate is that it is supposed to be an indicator of something that is poorly defined: intelligence. Scientists, philosophers, psychologists etc have been redefining intelligence for hundreds of years and there is still no exact consensus. Is it reasoning ability? Do you factor in “emotional intelligence?” What about people with low IQ that can tell you what day of the week it will be on January 3 2099? Or can play the piano very very well. I think that intelligence is something we can recognize intuitively. And so CSI is also something that we can intuit.

  24. Collin @22:

    The discussion here is about mathematics. One doesn’t brandish mathematics without working out a proof.

  25. 26

    Another thing to keep in mind is that these things take a long time. ID is only 20 years old. In that other thread people are saying things like real scientists spend their entire careers pursuing something. Is twenty years even enough for an entire career?

  26. Denise,

    You should let johnnyb do a post on this research he has done. I think it addresses some of what Mathgrrl was talking about.

    http://www.blythinstitute.org/....._final.pdf

  27. 28

    Joseph

    Specified Information is Shannon Information with meaning/ function

    Complex Specified Information is 500 bits or more of specified information

    For example let’s say that I have A and B, each 500 bits of Shannon Information.

    A is Specified Information because it is the compressed instructions for an antidote to a poison.

    B is random noise.

    In fact A is Complex Specified Information as it has 500 bits in it!

    So I get poisoned. I make the antidote. But, oh noes, I die!

    It turns out that B was the antidote after all.

    Yet when I look back at A and B it seems to me the only difference between them is the additional meaning I have to assign to them as I know in advance that one is the antidote. It seems “Specified Information” is “knowing more about the information in advance”, in this example anyway.

    Your “CSI” only works if you know in advance if something is designed, at least according to your examples ID Guy.

    The information in A and B did not change, only our knowledge of what it could be used for.

    You “metric” of Specified Information tells us nothing in advance. Both either have 500 bits of Specified Information or they don’t. Until you know which one is random noise neither do and both do.

    Very quantum. Very woo. Very you Joe, very confused.

  28. JR:

    Sure, but all those other people on the thread have read Dembski too and they cannot calculate (or most of them) CSI for the examples.

    The examples are still bogus for the reasons provided.

  29. 30

    johnnyb,
    That’s all great stuff, thanks. But best to address MathGrrls specific examples I think if you want CSI specifically to gain traction, or rather even just to live up to the claims previously made for it.

  30. JR:

    For example let’s say that I have A and B, each 500 bits of Shannon Information.

    OK- can nature, operating freely produce either one?

    JR:

    A is Specified Information because it is the compressed instructions for an antidote to a poison.

    B is random noise.

    How do you know? Are we supposed t take your word for it- you hve already proven to be a strawman maker.

    So here we have JR, a known sock puppet and strawman maker, making yet another strawman.

    JR:

    Your “CSI” only works if you know in advance if something is designed,

    No.

    at least according to your examples

    What examples?

    The information in A and B did not change, only our knowledge of what it could be used for.

    All you have done is make up a nonsensical story, just as Monty Python did here.

    I am sure you have yourself fooled into thinking you have refuted something though…

  31. JR:
    But best to address MathGrrls specific examples

    I have. MathGrrl refuses to understand that CI is about origins, even to the point of going out of her way to misunderstand her quote-mine of Dembski.

    And all you can do is act like her parrot. And that is pathetic…

  32. Jemima,

    Please read VJTorley’s comments in the last thread. I think he stated that the information did not even approach the Upper Probability Bound.

  33. 34

    Joe,

    OK- can nature, operating freely produce either one?

    I created both. One as the antidote, one as a red herring.

    How do you know? Are we supposed t take your word for it- you hve already proven to be a strawman maker.

    It’s my example. So by definition you have to “take my word for it”. Do you really not know how these things work?

    In fact both A and B are random noise. It just so happens one set of random noise creates an antidote when fed into my antidote machine.

  34. 35

    Collin

    I think he stated that the information did not even approach the Upper Probability Bound.

    I’m not sure I quite understand that, but logically if a gene duplication event can create N amount of CSI then Y such events can create YxN amounts of CSI. So therefore there is no limit really is there and the 500 bit limit (which is what I think you really mean as the UPB is a very big number indeed) can easily be surpassed if not all at once then in many such events.

  35. JR:

    I created both. One as the antidote, one as a red herring.

    You created a strawman because you are tired of blow-up dolls….

  36. JR:

    I’m not sure I quite understand that, but logically if a gene duplication event…

    Show me gene duplication events during the origin of life.

    You don’t get to start with that which needs explaining in the first place.

  37. Collin -

    I actually have posting privileges here myself, I just don’t make use of them often. You might be interested in some of my older posts.

    I’ll think about putting together a post on active information sometime this week. I’d actually prefer to do a more research-oriented paper first, and then put a summary on UD. Anyway, I’ll think on it.

    JemimaRacktouey -

    I don’t care much any given way for CSI. If someone can make good use of it, then great! If not, then great, too! It’s good to find out if something has a flaw early on.

    In the case of CSI, I think there is both the *measurement* and the *idea*. One can agree with the idea generally, and be in disagreement with the method of measurement.

    Think of all of the different ways that evolutionists have put together phylogenetic trees. Does the fact that there is more than one method mean that they are in disagreement over whether the tree exists? Of course not.

    For my own research, I have found Active Information to be easier to use with the types of data available. If someone else’s research uses a different metric better, then great!

  38. For those of you who know a lot of math and a lot about CSI, please tell me if what I’m about to say is incorrect.

    Calculating CSI in some instances is fairly easy to do. Calculating it for other instances is very difficult although possible in principle. Mathgrrl’s examples are an example of the latter and no one here is willing to go through the immense amount of work to do it. That does not mean that it’s not possible.

  39. Joseph at 35,

    I really don’t think that is appropriate language for a website that is popular among Christian youths.

  40. I have to agree with jon specter. Joseph’s language has crossed a barrier that has gotten others banned.

  41. Denyse:

    Meanwhile, perhaps junior ID theorists cannot formulate a single definition of complex specified information at present. But it really wouldn’t matter if they could. The materialist would just blink and say, “I’m sorry. This is so confusing.”

    This is exactly the case. That’s why I say, if MathGrrl is so interested, then let her do the work. And if she can prove that computer programs have randomly produced CSI, then we’ll be happy to debate. But I’m not doing her homework for her.

    Let’s all please note that she has asked for a “rigorous mathematical definition” for 4 programs.

    If you are trying to understand a concept, you only need 1 example; not 4. But she’s trying to make a point. She’s not trying to understand. And I, for one, don’t care to help her out other than to attempt some answers and to direct her to the sources she should check with AND ‘wrestle’ with.

    Specifications have patterns. Let her determine what the patterns are and how they’re generated. But as I’ve demonstrated so easily and so quickly (once I had Schneider’s paper), ev doesn’t rise to the level of CSI. So why should I, or anybody else, bother to define what CAN’T be defined. If the information is less than 500 bits, then it’s NOT CSI. Simple as that.

    Tell me, MathGrrl—point blank question: the ev program output: does it, or does it not, involve 500 bits of information?

    If you can’t (or won’t) answer this question, then there is very little to discuss.

  42. PaV, you may be right: I hope so. But I don’t think the long-term ID advocates in this discussion have acquitted themselves very well. With the notable exception of vjtorley, — and Denyse — they have been confusing, contradictory, goal-shifting, and altogether unscientific. Nobody can read this discussion and conclude that there’s any settled understanding of CSI or specification in ID theory. It’s the Keystone Kops around here.

  43. jon specter:

    I really don’t think that is appropriate language for a website that is popular among Christian youths.

    My apologies to any offended Christian youth. I should be more careful when taking out the trash.

  44. QuiteID, though I don’t pretend to understand the math side of the CSI thing for ID, but I do have a few measures that I use as benchmarks to see if evolution has generated functional complexity (i.e. functional information) over and above what was already present, that have served me well;

    the ‘fitness test’;

    Is Antibiotic Resistance evidence for evolution? – ‘The Fitness Test’ – video
    http://www.metacafe.com/watch/3995248

    Testing the Biological Fitness of Antibiotic Resistant Bacteria – 2008
    http://www.answersingenesis.or.....-drugstore

    Thank Goodness the NCSE Is Wrong: Fitness Costs Are Important to Evolutionary Microbiology
    Excerpt: it (an antibiotic resistant bacterium) reproduces slower than it did before it was changed. This effect is widely recognized, and is called the fitness cost of antibiotic resistance. It is the existence of these costs and other examples of the limits of evolution that call into question the neo-Darwinian story of macroevolution.
    http://www.evolutionnews.org/2.....s_wro.html
    QuiteID since this simple test has never been passed by Darwinists (not even by modern bacteria that was compared to 40 mya ancient bacteria that was revived), I see no reason to presuppose evolution has demonstrated even the most rudimentary plausibility of gaining functional information over and above what was already present!

    Shoot QuiteID, I don’t even know of a single example where a single novel functional protein was generated by Darwinian means (including Szostak’s ATP binding proteins since they were ‘pathetic’)

    Stephen Meyer – Functional Proteins And Information For Body Plans – video
    http://www.metacafe.com/watch/4050681

    Estimating the prevalence of protein sequences adopting functional enzyme folds: Doug Axe:
    Excerpt: Combined with the estimated prevalence of plausible hydropathic patterns (for any fold) and of relevant folds for particular functions, this implies the overall prevalence of sequences performing a specific function by any domain-sized fold may be as low as 1 in 10^77, adding to the body of evidence that functional folds require highly extraordinary sequences. http://www.ncbi.nlm.nih.gov/pubmed/15321723

    And measuring functional sequence complexity (information) for a single protein is a lot easier, than measuring the CSI for an entire system;

    Measuring the functional sequence complexity of proteins – Kirk K Durston, David KY Chiu, David L Abel and Jack T Trevors – 2007
    Excerpt: We have extended Shannon uncertainty by incorporating the data variable with a functionality variable. The resulting measured unit, which we call Functional bit (Fit), is calculated from the sequence data jointly with the defined functionality variable. To demonstrate the relevance to functional bioinformatics, a method to measure functional sequence complexity was developed and applied to 35 protein families.,,,
    http://www.tbiomed.com/content/4/1/47

    Mathematically Defining Functional Information In Molecular Biology – Kirk Durston – short video
    http://www.metacafe.com/watch/3995236

    Functional information and the emergence of bio-complexity:
    Robert M. Hazen, Patrick L. Griffin, James M. Carothers, and Jack W. Szostak:
    Abstract: Complex emergent systems of many interacting components, including complex biological systems, have the potential to perform quantifiable functions. Accordingly, we define ‘functional information,’ I(Ex), as a measure of system complexity. For a given system and function, x (e.g., a folded RNA sequence that binds to GTP), and degree of function, Ex (e.g., the RNA-GTP binding energy), I(Ex)= -log2 [F(Ex)], where F(Ex) is the fraction of all possible configurations of the system that possess a degree of function > Ex. Functional information, which we illustrate with letter sequences, artificial life, and biopolymers, thus represents the probability that an arbitrary configuration of a system will achieve a specific function to a specified degree. In each case we observe evidence for several distinct solutions with different maximum degrees of function, features that lead to steps in plots of information versus degree of functions.
    http://genetics.mgh.harvard.ed.....S_2007.pdf

    QuiteID not to mention, besides failing to demonstrate the origination of a ‘non-trivial’ functional protein, Darwinists have not even shown how a molecular machine could have originated by their ‘cobbled together process’;

    And of course QuiteID, quantum information is now found in proteins, Quantum Information that is ‘protein specific’ for each unique protein, which of course totally eliminates the probability argument, since quantum information is not reducible to a material basis! i.e. a ‘transcendent’ cause must be supplied for each instance of ‘unique quantum information’ that is beyond the capabilities of the material particles themselves to supply for each instance.

    http://www.uncommondescent.com.....ent-375283

  45. My two cents.

    One of the things Mathgrl wants seems to me to be an equation, something like e=mc squared. CSI is not an equation it is an acronym for Complex Specified Information (500 bits). Do we require equations for acronyms?

    My understanding is that once a specification and pattern is identified then that specification and pattern is subjected to probability analysis to see if it is more than 500 bits. What am I missing here?

    Vivid

  46. 47

    QID, I doubt you worry overmuch what the ID proponents here look like, but in any case, its not to worry. Darwin said “gradualism” in 1859 and everyone said “Hell Yeah”. Then (virutally) the entire scientific community stared directly at a non-gradual fossil record for the intervening 113 years until Gould and Eldridge finally said “Guess What?” in 1972. Even to this day, the excuses are like flies.

    Dembski (almost a one man band) began trying to formulize the completely undeniable existence of CSI, what – 10 years ago?

  47. One question that no one seems to be asking is what is the rigorous mathematical definition of natural selection and how do you use that definition to calculate whether or not a species will survive (or evolve) or not.

  48. One question that no one seems to be asking is what is the rigorous mathematical definition of natural selection and how do you use that definition to calculate whether or not a species will survive (or evolve) or not.

    Excellent point. It is also interesting that the Darwinists have given themselves until the end of time to fill in the gaps in their theory (which has yet to perform a single probability calculation for the arising of new, tightly-coordinated molecular machinery), but ID is expected to have all the answers right now, expressible to the ninth decimal place.

  49. Collin:

    No. CSI is easy to recognize. It may (for sake of argument) not be amenable to calculation yet. But as I’ve said before, light was recognizable before it was calculable (with precision).

    But even light could be roughly quantified. Situations with ‘no light’ could be distinguished from situations with ‘lots of light’. Questions like “If I double the amount of lit candles of the same sort, does that roughly double the amount of light?” could be easily answered.

    So far, CSI doesn’t seem to have even gotten that far.

  50. Joseph

    I should be more careful when taking out the trash.

    You are determined to be as uncivil as possible, aren’t you?

  51. No Jon, that is entirely the realm of the anti-IDists. IOW your bias is showing. Gee just look at the crap I was responding to…

  52. WRT Joseph at 51:

    QED

  53. LMAOF @ Jon 52

  54. johnnyb, I read, with interest, your emphasis on active information (positive, negative and neutral) being key to understanding the Conservation Of Information in complex systems, and was also interested in your desire to do a post on it. In regards to all this I was wondering if you could also clarify the importance of ‘prescriptive’ information, and its relation to active information:

    Prescriptive Information (PI)- Abel
    Semantic (meaningful) information has two subsets: Descriptive and Prescriptive. Prescriptive Information (PI) instructs or directly produces nontrivial formal function (Abel, 2009a). Merely describing a computer chip does not prescribe or produce that chip. Thus mere description needs to be dichotomized from prescription. Computationally halting cybernetic programs and linguistic instructions are examples of Prescriptive Information. “Prescriptive Information (PI) either tells us what choices to make, or it is a recordation of wise choices already made.” (Abel, 2009a)
    http://www.scitopics.com/Presc.....on_PI.html

    ,,,, Functional Information (FI) has two subsets of information, Descriptive Information (DI) and Prescriptive Information (PI),

    Abel, D.L.; Trevors, J.T. 2005, Three subsets of sequence complexity and their relevance to biopolymeric information., Theoretical Biology and Medical Modeling, 2, Open access at http://www.tbiomed.com/content/2/1/29

  55. Collin:
    Calculating CSI in some instances is fairly easy to do. Calculating it for other instances is very difficult although possible in principle. Mathgrrl’s examples are an example of the latter and no one here is willing to go through the immense amount of work to do it. That does not mean that it’s not possible.

    That’s correct. One has to identify the pattern associated with an event. And there is then a chance hypothesis that is associated with the pattern. Once all of that is established, then the math is easy.

    In the case of Schneider’s ev program, the “pattern” is not what MathGrrl claims it to be. The specification should be along the lines of: “a binary representation of an enzymatic binding site that is 16 nucleotides long”. Rather than go through the mumbo jumbo of establishing the chance hypothesis—which then entails looking through the computer program and fitness functions and such—I simply noted the binary length, which is equal, roughly, to 260 bits. But CSI, as defined in NFL, must contain at least 500 bits of information as defined there (which is a simple enough calculation). Thus, the output of the ev program, on the face of it, does not rise to the level of CSI.

    AND, MathGrrl should have understood that!

    So, bottom line, it is impossible to write a “rigorous mathematical definition” of CSI for the “specification” that actually exists in the program. MathGrrl, incorrectly wrote: ““A nucleotide that binds to exactly N sites within the genome.”

    Well, if this were the true “specification” of the ev program, then if N > 50 (there’s 10 binary digits per nucleotide in the program), then it could, conceivably become CSI. But we know N = 16.

    So why did MathGrrl provide an erroneous “specification”?
    Was she up to something?

    Further, since I read Dembski’s “Specification” paper six years ago, and his NFL two-and-half years ago, I’m more familiar with NFL than the Specification paper. So I didn’t notice at first, but it turns out that you won’t find the letters CSI anywhere in that paper.

    So, then, how is it that MathGrrl asks for a “rigorous mathematical defintion” of CSI based on the paper? Is she just ignorant of the differences? Is she playing games? Does she not care to wrestle with Dembski writes there so she looks for someone else to do it for her?

    I’m a little amazed, actually.

  56. PaV @ 55: You say:

    “So I didn’t notice at first, but it turns out that you won’t find the letters CSI anywhere in that paper.”

    Which is correct… but… the paper does mention “Specified Complexity” many times and even has a section with that as the tile. Isn’t the term “Specified Complexity”(SC) interchangeable with the term “Complex Specified Information”(CSI)?

  57. PS: The term ? (chi) also appears multiple times in the paper and in the glossary of terms on this site and identifies it as the metric for CSI.

  58. Quiet ID [43]:

    Nobody can read this discussion and conclude that there’s any settled understanding of CSI or specification in ID theory. It’s the Keystone Kops around here.

    Well, I think this is principally due to several reasons:

    (1) MathGrrl’s request was not only unreasonable, but unachievable in the case of one, and likely in all four cases. That is, CSI, as defined in NFL, does not exist in the ev program; so how does one define what does not exist?

    (2) MathGrrl set up an impossible scenario: she asks for a CSI calculation for these programs, and then INSISTS, and I mean INSISTS, that it be based on Dembski’s “Specification” paper. Well, there’s this little bitty problem: Dembski, in his “Specification” paper, no longer talks about CSI, but, instead, “specified complexity”, and he also talks about “semiotic agents” and such.

    What happened was that she was being given proper answers from Dembski’s paper by a number of people here at UD, but she didn’t seem to understand the answers, and questions, given her. (Or else, she simply refused to answer the questions.)

    (3) For my part, I’m more familiar with CSI as Dembski presents it in NFL since I went through and digested it a couple of years ago. But it’s been six years since I read the “Specification” paper, and I wasn’t sure where that sat anymore. But, because of this putative “discussion”, I’ve had to reread it. While profitable, it has been an expenditure of time and energy foisted upon me simply in an effort to try and understand what MathGrrl was trying to get at.

    However, now have reread most of the paper, I now see that Dembski has reformulated CSI because of some of the (trivial) criticisms leveled against it over the years. Because of the added rigor needed to refute the rather trivial criticisms directed at CSI, Dembski chose to reformulate it mathematically, and now calls it “specified complexity”. In the Addendum to the paper, you can read where and how CSI and “specified complexity” match up. The net effect of these trivial criticisms is that mathematically things have become even a little harder to grasp, and CERTAINLY, would require more time and effort (simply because terms have been introduced into the equation that need to be determined), to DEVELOP a “rigorous mathematical definition” for the indicated scenarios. This would involve serious time, effort and work. Now why should we, here at UD, do this hard labor for MathGrrl. Let her do the work.

    My advice to MathGrrl is that she reread both NFL and the “Specification” paper until she fully understands it, and then tackle one of the remaining three programs she’s listed. I say three, because, patently, the ev program does not produce CSI, or, if you will, “specified complexity”.

    My further advice to her is: don’t waste your time, because you’re not going to be able to prove that either CSI—as defined in NFL—or “specified complexity”—as defined and developed in “Specification”.

    As to saying that there is no “settled understanding of CSI or specification”, this appears so because NFL is different from the “Specification” paper. The simpler concept is CSI. So MathGrrl should stick to NFL and try and prove that CSI, as defined by NFL, exists in one of those programs. That ought to keep her busy for a while.

    If there is no other point you take away from all of this, please understand that a specification and a chance hypothesis is DIFFERENT for every instance of supposed CSI (“specified complexity”). Notice she gave four instances. Why? Because they are different for each one—while the definition of CSI always remains the SAME! There are criteria that have to be met in all valid instances of CSI/specified complexity: these stay the same and form the “DEFINITION” of CSI. What the actual pattern is varies. What the actual chance hypothesis corresponding to the appearance of the ‘pattern’ also varies, etc., etc.

    One final point:

    I’ve maintained all along that MathGrrl’s demand is excessive. The proof of that is that she has failed to provide anything like it. The hard work is in developing the chance hypothesis. She’s failed to do this. And she hasn’t told us WHY she’s failed to do this. And, of course, her “specifications” are in error, or so they seem to me. Maybe she would like to begin by explaining WHY she thinks they represent the specifications inherent in the programs.

  59. utdifian:

    Isn’t the term “Specified Complexity”(SC) interchangeable with the term “Complex Specified Information”(CSI)?

    Not really. CSI is given in “bits”. Specified complexity is a “number”.

  60. utidijian:

    PS: The term ? (chi) also appears multiple times in the paper and in the glossary of terms on this site and identifies it as the metric for CSI.

    There are several, you might say, “co-factors” that have been added to the original CSI equation, and I think chi is part of the reformulation.

    CSI and “specified complexity” are intimately related, but not the same.

    But, bottom-line, if anyone wanted to check for “specified complexity”, the easiest way to check would be to check for CSI. If you find CSI, then you could move on to check for SC.

  61. So chi is in “bits” or a “number”?

  62. 63

    PaV @ 60.

    That all seems non-straight forward.

    But, bottom-line, if anyone wanted to check for “specified complexity”, the easiest way to check would be to check for CSI. If you find CSI, then you could move on to check for SC.

    Once you’ve found CSI how exactly do you “check for SC”?

    Could you give an example of such? It’s not clear to me how that would be applied in a real scenario.

  63. Mrs. O’Leary,

    I’m unexpectedly off the net for a weekend and come back to find yet another thread with my name in it! You’re quite good for my ego.

    I would like to clarify one point you raised, though:

    Meanwhile, perhaps junior ID theorists cannot formulate a single definition of complex specified information at present. But it really wouldn’t matter if they could. The materialist would just blink and say, “I’m sorry. This is so confusing.”

    I’ve been through this with enough different issues to know that that would be the outcome, for sure.

    I firmly agree that some ID opponents would respond exactly that way, just as some ID proponents dismiss empirical evidence that doesn’t support their views. We’re all humans here, and we don’t always respond rationally.

    That being said, I’m an optimist on most days so I think that there will be enough people of good faith and good nature to fairly evaluate all the evidence. You put your finger on a way to contribute to that:

    So, for ID theorists, the goal is not convincing such people or reaching an impasse with them, but formulating definitions that actually lead to new discoveries or clearer understanding of current ones.

    This is exactly what I hoped would happen in the CSI thread, but we’re near 400 comments without any resolution. If anyone in this thread can help out, please join us in that one.

  64. Joseph,

    For example, the next time somebody makes a claim involving CSI I’ve learned that I can simply link them to that thread and ask them if they can apply their claimed knowledge of CSI to the examples in the OP.

    The examples in the OP are bogus for the reasons provided in that thread.

    You didn’t provide any reasons that are supported by any of Dembski’s books or papers on CSI, you simply declared my scenarios “bogus” out of hand.

  65. PaV,

    MathGrrl’s request was not only unreasonable, but unachievable in the case of one, and likely in all four cases. That is, CSI, as defined in NFL, does not exist in the ev program; so how does one define what does not exist?

    You’re still confusing the simulator with what is being simulated. I am asking for the CSI in the bit strings that represent the genomes of the digital organisms.

    MathGrrl set up an impossible scenario: she asks for a CSI calculation for these programs, and then INSISTS, and I mean INSISTS, that it be based on Dembski’s “Specification” paper. Well, there’s this little bitty problem: Dembski, in his “Specification” paper, no longer talks about CSI, but, instead, “specified complexity”, and he also talks about “semiotic agents” and such.

    If my understanding is correct, specified complexity beyond a certain number of bits constitutes CSI, according to Dembski.

    I’ve maintained all along that MathGrrl’s demand is excessive. The proof of that is that she has failed to provide anything like it.

    I’m asking for clarification of a core ID concept. Why would it be incumbent on me to provide “anything like it”?

    ID proponents make very strong claims that depend on CSI being an objective metric. It is not “excessive” to request a rigorous mathematical defintion of that metric and some detailed examples of how to calculate it.

  66. MahGrrl:

    I am asking for the CSI in the bit strings that represent the genomes of the digital organisms.

    How my bits are the for each digital organism?

    Also, as I said in that long thread, CSI is a yes or no thing- either it is present or it isn’t.

    If it is then that indicates design.

  67. MathGrrl:

    You didn’t provide any reasons that are supported by any of Dembski’s books or papers on CSI, you simply declared my scenarios “bogus” out of hand.

    CSI pertains to origins and your quote-mine doesn’t change that fact.

    And you still haven’t read “No Free Lunch” so I say you don’t have anything to say about CSI.

    BTW I am sill waiting for that matmatically rigorous definition of a computer program. You seem to be avoiding that at all costs.

    I say that is because it exposes your srawman.

  68. Also, as I said in that long thread, CSI is a yes or no thing- either it is present or it isn’t.

    I thought it was measured as a number of bits?

  69. Yes it is and either it is present or it isn’t.

    So how many bits in those digital organisms MG is talking about?

    I take you haven’t read “No Free Lunch” either.

  70. You know. Joseph, your continued hostility towards just about everyone is rather tiresome.

  71. So now pointing out the obvious is now “hostility”?

    Dealing with weasels is rather tiresome yet here we are.

  72. 73

    Joseph,

    CSI is a yes or no thing- either it is present or it isn’t.

    And

    Also, as I said in that long thread, CSI is a yes or no thing- either it is present or it isn’t.

    Yet to the best of my knowledge that’s simply not true. For example Dr Dembski said:

    If we now define CSI as any specified information whose complexity exceeds 500 bits of information, it follows immediately that chance cannot generate CSI. Henceforth we take the “C” in “CSI” to denote at least 500 bits of information.

    Do you accept you are wrong in your claim?

  73. Well, it seems like the complexity part is defined by bits but the specificity is not.

  74. 75

    Moderators,
    I followed the link on the poster’s name for “Joseph” to his blog and found the following comment he left:

    Why are you in such a rush to have your boyfriend hospitalized?

    I’m surprised you’ll allow a link on every post of his on this site to a blog containing such homophobic threats!

    But your moderation policies are not mine to question so I defer to whatever your standard policy is on such issues.

  75. Maybe I’m finally starting to get it. Somebody tell me if I’m wrong. Specificity can be present in any information even if its only 1 or 2 bits. But complexity is defined by Dembski as at least having 500 bits of information. If that information is all specified, then it is CSI and an indicator of design. Is this an accurate portrayal of the idea?

  76. JR-

    There isn’t anything in your Dembski quote-mine that contradicts what I posted about CSI.

    And my daughter’s god-father is homosexual.

    But thanks for stooping to new lows to try to stop your opponents.

  77. So let me guess- JemimaRacktouey is either blipey the clown or oldmanintheskydidit/ OM.

    My guess is the sockpuppet has been banned from here and has come back to incite hostilities.

    My apologies for taking the bait even though I knew it was fishing.

  78. Collin- that is it. 500 bits of si is the upb. but I doubt the limit needs to be that high.

  79. 80

    Joseph

    And my daughter’s god-father is homosexual.

    Then what’s with all the homophobia and threats of violence against of all things a clown?

    You are not scared of clowns are you?

  80. 81

    Collin

    If that information is all specified, then it is CSI and an indicator of design.

    More or less. However that calculation now needs to be performed for the examples given (Josephs objections notwithstanding) on the other thread. Can you? Can Joseph?

  81. blipey the clown:

    Then what’s with all the homophobia and threats of violence against of all things a clown?

    It’s all your twisted imagination.

    You are not scared of clowns are you?

    Erik you are the one sending me emails.

  82. Still waiting for evidence for gene duplicaions and the origin of life.

    And we are till waiting for tht mathemaically rigorous definition for a computer program.

    Geez you would think that the evos drooling over CSI would step up instead of avoiding the issues.

  83. Jemima,

    No I cannot. I would refer you to VJTorley’s most recent thread, though.

  84. My thanks to everyone for what has been a fascinating discussion.

    To Joseph @ 68…

    I’m not a computer scientist, but wouldn’t a Turing Machine count as a mathematically rigorous definition of a computer programme? I understand Dr. Hava Siegelmann of Technion claims to have found an even more powerful and mathematically rigorous class of computing machines.

    Googling for “evidence gene duplication” also turns up may interesting results.

    None of that of course rules out Intelligent Design as a concept.

Leave a Reply