Home » Complex Specified Information, Darwinism, News » #4 of 2011 for ID community: “Stylus” Computer Program Aims to Bridge Gap Between Real World and Artificial Evolutionary Simulation.

#4 of 2011 for ID community: “Stylus” Computer Program Aims to Bridge Gap Between Real World and Artificial Evolutionary Simulation.

Every year, Access Research Network publishes a list of the 10 most significant science news stories for the year, for the intelligent design community – in consultation with theorists and writers. For 2011, here’s #4:

Many computer simulations that purport to simulate Darwinian evolution have deficiencies: Pro-ID scientists like William Dembski or Robert Marks have shown how programs smuggle in information such that they are pre-directed to evolve their targets. Doug Axe’s work demonstrates that such programs typically evolve solutions to artificial, rather than real-world problems. A new peer-reviewed paper in BIO-Complexity by Axe, Philip Lu, and Stephanie Flatau explains that the “functions” of the digital organisms in these simulations are often divorced from real-world meaning. According to Axe they designed Stylus to present a more accurate picture of what evolution might be able to accomplish in the real world: “The motivation for Stylus was the recognition that prior models used to study evolutionary innovation did not adequately represent the complex causal connection between genotypes and phenotypes.”

Basic to life is an information conversion, where the information carried in genes (the genotype) is converted into an organism’s observable traits (the phenotype). Those biological structures then perform various functions.

Stylus uses Chinese characters as digital objects as explained in the paper: “These translation products, called vector proteins, are functionless unless they form legible Chinese characters, in which case they serve the real function of writing. This coupling of artificial genetic causation to the real world of language makes evolutionary experimentation possible in a context where innovation can have a richness of variety and a depth of causal complexity that at least hints at what is needed to explain the complexity of bacterial proteomes.” There probably will never be a perfect computer simulation of biological evolution, but the free Stylus software brings new and improved methods to the field of evolutionary modeling. This tool will help those interested in testing the viability of Darwinian claims to assess whether complex features can be created by random mutations at the molecular level.

See also:

#1 of 2011 for ID community: 50th Peer-Reviewed Pro-ID Scientific Paper Published.

#2 of 2011 for ID community: The Design of the Butterfly Continues to Inspire and Amaze.

#3 of 2011 for ID community: Woodpecker Drumming Inspires Shock-Absorbing System.

#5 of 2011 for ID community: Explosive Radiation of Flowering Plants Confirmed

#6 of 2011 for ID community: Golden Orb-Weaver Fossil Spider Provides New Evidence for Stasis.

#7 of 2011 for ID community: Complexity in the Universe Appears Earlier Than Thought.

#8 of 2011 for ID community: An Identity Crisis for Human Ancestors.

#9 of 2011 for ID community: DNA Repair Mechanisms Reveal a Contradiction in Evolutionary Theory.

#10 of 2011 for ID community: Limits to self-organization of life identified

Follow UD News at Twitter!

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

45 Responses to #4 of 2011 for ID community: “Stylus” Computer Program Aims to Bridge Gap Between Real World and Artificial Evolutionary Simulation.

  1. Why does the program have to produce existing words in order to demonstrate evolotion?

    The other day I was thinking about knots and whether they are similar to protein folds. Certainly knots represent specified functional information. There are ways of coding them digitally. Their existence is objective in a way that human languages are not. They do not require a dictionary of targets.

    So why not Model evolution as knot tying?

  2. 2
    EndoplasmicMessenger

    One reason they chose Chinese is because the ratio of the number of Chinese characters to the universe of possible characters is roughly equal to the ratio of the number of actual functional proteins to the number of possible proteins.

    Can you say the same thing about knots?

  3. Really? What is that ratio? What is your source?

    Does Stylus support selection of incremental improvements? Does it have populations with individuals having selectable differences in fitness?

    My question about knots is a question about whether an objectively definable function can occur without being designed. There’s an entire branch of mathematics dealing with knots. I’m sure there’s a way to code them.

  4. Encode them? As information? For what effect, or purpose? Can they be encoded without the use of abstractions, and rules to actualize those abstractions, or are representations and protocols (a formal system) required for the encoding itself?

  5. I’m not sure what you are asking. Knots are definitely functional. They also exhibit a continuum of functionality. They can be described in a way that allows new forms that have never existed before.

    That makes them more like proteins, which might be viewed as a subset of knots.

    Ckinese characters are a limited set, and evolving them is just a variation ova the Weasel program. I find my Itatsi.com program slightly more interesting, because it is not limited todictionary words. It can create new words that are not in any dictionary, but which could be useful to people looking for new tradenames.

    I’ve already seen it produce some existing drug names — words that are not in its dictionary. It can do this because its definition of fitness is based on pronouncability rather than on inclusion in a dictionary.

  6. You’ve repeated the statement that you can record knots as encoded information. I assume you mean to encode these as a means of replication. Among other things, I am asking how you plan to do that without a formal system of representations and protocols.

    If that question trips you up, then perhaps you are in over your head. It happens.

  7. I’m not sure what you mean by a means of replication. Do Chinese characters replicate? The process being modeled is incremental change. In order to have evolution there must exist a landscape having connected regions of function, as defined by some definition of fitness.

    I don’t know about Stylus,so I asked if it supports incremental, selectable change.

    But I also think any model of evolution should have the capacity to create objects not in the dictionary used by the selection function.

    Which is to say, it should support the creation of novelty and novel functionality.

    Arrangements of strings can be modelled in a way that allows determining whether they form a knot, even if the knot takes a form never seen before. One could even classify them and sort them by complexity.

  8. So you say that you want to model evolution by the encoding of knots. (Evolution, obviously being driven by variation in the encoded information used for replication, unless you now wish to claim that evolution can occur without replication). But when asked how you intend to encode the information, you first respond that you don’t understand the question about “encoding”, and now you respond that you don’t understand what is meant by “replication”.

    Did you not say:

    So why not Model evolution as knot tying?

    There are ways of coding them digitally.

    I’m sure there’s a way to code them.

    So the question is rather basic; How do you plan to encode them into information without a formal system of representations and protocols?

    Again, if Step One is a cause for problem, then you are very obviously over your head.

    (Yet, you are the first to claim that there is no evidence of design in nature).

  9. So what is it going to be Petrushka? How are you going to record the state of your knots in information without a formal system of physical representations and physical protocols (just like any other form of encoded information).

    Be upfront about it.

  10. I don’t understand why you are asking the question. There has to be some equivalent of DNA and some procedural language for describing the process of moving the string ends. So what is your point?

    My point is that knots can be abstracted, and so can the movements required to form them. And unlike dictionary lists, there is no arbitrary limit to the number of different knots. And like protein folds, any possible configuration can be classified as belonging or not belonging to the set of knots. One does not need a list, just a few rules.

  11. 11

    I don’t understand why you are asking the question.

    Petrushka, I find that very hard to believe. I have been rather consistent in my dealings with you for the very reason that you have (in turn) been rather consistent in repeatedly taking for granted what must be explained, while at the same time voicing over and over and over again that ‘no evidence whatsoever exist for design’.

    Above, you suggest the idea of modeling evolution by means of encoding knots, and you give your reasons for the suggestion – fine.

    My first question to you is how are you going to encode the information without a formal system of representations and protocols – that is, (as you seem to acknowledge) by abstracting the state of the knots into physical representations which must in turn be coordinated to specific protocols for decoding those representations back in to the state of the knot. In other words, without the rise of the abstractions and rules (which are the critical requirements of any and all recorded information).

    Abstract representations and protocols are a formal system, Petrushka – a formal system. A system which must arise PRIOR to genetic heredity – prior to the onset of evolution by means of mutable genetic information. And it is also a sterling example of irreducibly complex (a physical representation and a physical protocol operating in a precise dynamic relationship, how could it be any other way, either object is useless without the other).

    To any rational person, information is the organizing force behind all living systems. And its very existence is entirely dependent on a formal semiotic system of representations and protocols. Look at your own language that you used… where you talk of abstractions, rules, and procedures. These are the things you simply take for granted, and under the scenario of materialism, these are the very things which must be organically produced (prior to genetic evolution) from nothing but the very simplest of chemical structures.

    Materialism promotes proto-life chemistry as being the source of the observed formalities in biology; creating abstractions and establishing rules.

    Gimme a break. You should seek to be more accurate in your apprasal of the evidence.

  12. Abstract representations and protocols are a formal system, Petrushka – a formal system. A system which must arise PRIOR to genetic heredity – prior to the onset of evolution by means of mutable genetic information.

    You seem to be bringing origin of life into a discussion of evolution. My question is why?

  13. 13

    1) Because its science.

    2) Because its the way we find it.

    3) Because you yourself keep bringing it up (eg “no evidence for design”).

    4) Because you’d rather have your spleen removed with a fork than to acknowledge the evidence in front of you.

    - – - – -

    Now, how are you going to enciode the information in your knots without physical representations and physical protocols acting together in a dynamic semiotic system?

  14. Suppose I can’t. what’s your point?

  15. What evidence for design have you brought to the table?

  16. Answer a simple yes or no question. You are discussing the origin of life: yes or no?

  17. 17

    Knots can be functional, but how many potential functions can they have?

  18. I am not trying to model chemistry.

    I’m addressing what I think is a key defect in Stylus, and that is its finite list of targets. I also suspect Stylus has no way to select incremental steps toward its targets. No one with knowledge of the program has stepped forward to clarify this.

    A virtual knot tying program would not be limited to a predefined list of targets.

    The problem with Weasel type programs is the fixed target or list of targets.

    Evolution is constructive. It builds things that have no previous physical existence. If you are going to model evolution you need to get away from the concept of targets and into the concept of construction. Construction of novel forms.

    The thing I like about knots is that there are an infinite variety of knots, but you can test any configuration to see if it is in fact a knot,and will not unravel when the ends are pulled.

    If you have a population of evolving knots you can select by some mathematical criterion, such as the number of crossovers. You could also build physical knots and test them for strength.

  19. “Why does the program have to produce existing words in order to demonstrate evolotion?”

    The extant words represent functional proteins or organic functions in similar fashion to the way the symbols represent amino acid residues (in the case of the former) or proteins binding together(in the case of the latter.)

    “So why not Model evolution as knot tying?”

    If I’ve understood the purpose of “Stylus” properly, it is to demonstrate the extent to which functional proteins or networks of functional proteins can be assembled by randomly acting (no design) processes. Useful knots are intimately linked to intelligent agents who tie knots for a purpose; thus their “usefulness”. Knots that simply occur at random (most recent example, the strings of Christmas lights I placed in a plastic storage bin last year), likewise have no apparent use or purpose (function). Since the program’s purpose (again, with the caveat I’ve understood it’s purpose correctly) is to demonstrate to what extent random processes can assemble independent parts into a functional whole, knot tying simply doesn’t help in that regard.

  20. Now, how are you going to enciode the information in your knots without physical representations and physical protocols acting together in a dynamic semiotic system?

    I see you are still assuming your consequent.

  21. 21

    I don’t see why knots are a good example. You can add infinitely to their complexity without increasing function. This might be me lacking imagination, but I don’t see what a knot will ever do besides be a knot, especially if that’s the only expectation ever set for it.

    If you wish to use models to demonstrate how variation and selection can result in increased function and innovation, does it not follow that your model must actually do that?

    Evolution is constructive. It builds things that have no previous physical existence. If you are going to model evolution you need to get away from the concept of targets and into the concept of construction. Construction of novel forms.

    Actually the jury is out on that or else you wouldn’t be discussing the use of models to make your case. Right now no one can say that evolution does any such thing.

    But I don’t see how evolving a billion “novel” knots would even be a step in that direction.

  22. 22

    The answer is not “Suppose I can’t”

    The only answer is “I can’t”.

    Pertrushka, as you can see from your own example (if you’ll allow the evidence to appropriately trump your ideology for one moment), the manifest step for evolution is the Onset of Information. Evolution cannot take place without it. The onset of information requires specific physcal entailments to be placed on matter, ie formalities; a formal system of physical representations and physical protocols.

    So evolution cannot exist without information, and information cannot exist without a formal semiotic system.

    “What evidence for design have [I] brought to the table?

    That was an ignorant question.

    ignore:

    1: to refuse to take notice of

    A central prediction of ID is supported by the material evidence.

    Will that do?

  23. 23

    Actually, if you did model the evolution of knots as you propose, it would demonstrate exactly the opposite of what you wish. (At least I think it would.)

    As with real-world evolution and every GA ever written, you would see variation of an existing form with no real innovation.

    If the fitness test were that pulling on the ends didn’t unravel it, what sort of innovations do you expect the knots might evolve to avoid being untied? Camouflage? Evasive maneuvers? Poisonous spikes?

    Yes, I’m being facetious. But I suspect that this, like any other GA, would reinforce both the strengths and limits of darwinian evolution just as every GA or real-world biological example already has.

  24. 24

    Dr Liddle, your unsupported counter claims have run dry. You have sat silent for almost two months while the response to you has been waiting on your blog.

    If you are now going to attempt a new tactic where your others have failed, then have the common descency to support it.

    But if you do, then stop running from the kitchen when it goes against you.

  25. Oh, excellent. Yes, I have been neglecting my blog.

    But, as you well know, I withdrew my claim.

    Doesn’t alter the fact that you are still assuming your consequent.

  26. 26

    Dr Liddle, if after two months time if all you can muster is a hit-and-run assertion, then I am happy to leave it at that. Congratulations, you have now done it twice.

    But if you intend to support your claim that the semiotic argument is invalid (beyond making the mere assertion of such) then I am still awaiting your response to the question:

    “If in one instance we have a thing that actually is a symbolic representation, and in another we have something that just acts like a symbolic representation – then someone can surely look at the physical evidence and point to the distinction between the two.”

  27. UprightBiped:

    Premise 1: the manifest step for evolution is the Onset of Information.

    Restatement of Premise 1: Evolution cannot take place without it.

    Premise 2: The onset of information requires specific physcal entailments to be placed on matter, ie formalities; a formal system of physical representations and physical protocols.

    Conclusion 1: So evolution cannot exist without information,

    Conclusion 2: and information cannot exist without a formal semiotic system.

    As you can see, your Conclusion 1 is identical to Premise 1 and your Conclusion 2 is identical to Premise 2.

  28. What “hit and run assertion”?

    As for your “question”…

    “If in one instance we have a thing that actually is a symbolic representation, and in another we have something that just acts like a symbolic representation – then someone can surely look at the physical evidence and point to the distinction between the two.”

    …where is the question? Looks like, um, an assertion to me :)

  29. 29

    Dr Liddle, you must be kidding.

    The manifest step for a basketball game is having a basketball.

    A basketball game cannot be played without a basketball.

    Study argumentation much?

  30. 30

    I take that as a mealy-mouthed concession. Another one.

  31. But that isn’t the form of your argument. Look at it.

  32. I repeat: What “hit and run assertion” were you referring to?

    And what is your question?

    You cannot take failure to answer a non-question as a “concession” of anything!

    Ask your question, and I will attempt to answer it.

  33. 33

    Petrushka has read my argument Dr Liddle. The above was not my argument – as you very well know.

    These tactics are very reminiscent of your last post to me before you abadoned the conversation. In that instance you started putting words in my mouth even though you knew (without any doubt) that the words/thoughts you attributed to me were not mine, but yours. They served your pupose in the situation where the evidence against your claim had simply accumulated to the point that they were no longer tenable.

    After having to concede twice, I think these pointless skirmishes are beneath you. Either attack the evidence, or lets let sleeping dogs lie.

  34. 34

    Not to carry on, because I know I’m very likely just talking to myself.

    But this demonstrates the low expectations that people have of GAs and modeling evolution. I don’t think it would ever even occur their designers that they might truly innovate like what we observe in nature. The bar is set quite low, the virtual evolutionary equivalent of varying finch beaks.

    We have antennas that lengthen, shorten, or take on particular shapes. In the case of knots, we have a string looped around itself so that it cannot be straightened by pulling on both ends. And the expectation is never set that evolving them would result in any more than a variation of exactly that.

    As UB has pointed out, a great deal of design and assignment of symbolic information must occur to even model such simple things. And yet the common excuse for why models can’t evolve and innovate even as simple cells is that the models are not nearly complex enough.

    It’s just another way of stating the obvious, that a system even hypothetically capable of evolving any complexity requires a massive investment of complexity and information which must obviously arise without evolution.

  35. 35

    Elizabeth, the question highlighted above was given to you no less than half a dozen times (or more) in the context of a conversation that began in May 2011 when you claimed to be able to demonstrate the rise of information in a sim. That conversation ended months later after tens of thousands of words (over numerous threads) when you finally conceded that you could not fulfill the claim.

    You can choose to finally address it here now, or you can’t.

  36. Those were your exact words, Upright BiPed. I put no words in your mouth at all.

    It was the argument you presented in 3.1.2.2.5

  37. What question? Please state it in words. I cannot see a “question highlighted above”. Cut and Paste will do.

    And yes, I conceded that I could not demonstrate the rise of information in a simulation where information was defined as you defined it.

  38. Yes or no? Your argument concerns the origin of life?

  39. The origin and subsequent evolution are directly linked. So whatever concerns the origins also concerns the evolution of.

  40. As far as I can tell, yes. His challenge seems to be: how did the information chicken get into the egg without an information chicken to put it there?

    I was going to have a shot at demonstrating that with a sim, but we got bogged down over the definition of information.

    Eventually I figured that the challenge was either impossible to meet (I wasn’t reassured that if I succeeded with what I proposed that he would accept the result), or indeed possible in the time I had available.

    I did find an interesting sim that did the first part of what I proposed, but I’ve forgotten the name, and I can’t seem to google it up any more. It was a one syllable word with some an s or a z in it somewhere. Schnazz or something.

  41. I don’t have any academic training in philosophy or logic, but your assertion that evolution has certain lofty prerequisites is the subject of much research. Some Nobel prizefighter winning scientists should perhaps abandon their research programs and just ask you.

    At any rate you have hijacked the discussion of Stylus.

  42. 42

    What question?

    Two friends walk out of a pub after a game of pool. The friend who did the driving turns to his buddy and says “My car is GONE!”. His friend then looks at where the car was parked, and says “It’s not gone”. The driver then throws his hands in the air, “look here’s my parking stub with the number on it, and over there is where you poured out your cola when we got here…it’s GONE!”

    “It’s not gone” his buddy says again.

    Exasperated, his arms drop down to his side and he quietly asks “If its not gone, then I suppose you can answer the question”. His friend then turns to him and mutters “What question?”

    This is the level of intellectual pettiness being marshaled against the obvious question being raised. My fear is that it will only go downhill from here. There are no indications otherwise.

    And yes, I conceded that I could not demonstrate the rise of information in a simulation where information was defined as you defined it.

    Allow me to refresh your memory. It was my definiton of information that your argued against for months on end. Only after failing to show it false (and in fact, being forced to agree with it) did you finally concede. Perhaps you’ll remember commenting that my definition was “more defensible” than any others that you’d encountered. The reason for that is simple; it was based on physical evidence.

    Clearly, at this point that definiton has not changed, and neither have you, so I see no profit in rehashing it again. But that is not the only observation to be made here.

    Among people who work in strategy for a living, it is widely known that the losing position will often go for a foul. One can then relish in the foul, in place of dealing with the defeat. A common method of drawing a foul is with pettiness.

    I’m not interested.

    The simple fact remains; you cannot simulate the unguided rise of information, you cannot argue against the definition of the term, and you cannot demonstrate that the semiotic argument is false.

  43. UprightBiPed, you have this disconcerting habit, when I don’t understand you, and ask you for clarification, of giving me a lecture on my faulty integrity instead of attempting to clarify your meaning.

    As I’ve said before: we seem to have a major communication difficulty. I do not blame you, the fault could be entirely on my side (although of course my view is that it probably isn’t).

    But, whatever the reason, it means I could not, and cannot, fulfill your challenge. It also means I do not know what question you asked me to answer just now. Truly.

    If you’d spend a fraction of the time you spend lecturing me on simply phrasing your question clearly, I could at least attempt to address it.

    But, as you will not, I cannot.

    I do not find the question “obvious” and my request is not “petty”. It is exactly what it says on the tin: “please tell me what your question is.”

  44. 44

    Really, the question isn’t obvious? (Just poking at this.)

    If in one instance we have a thing that actually is a symbolic representation, and in another we have something that just acts like a symbolic representation – then someone can surely look at the physical evidence and point to the distinction between the two.

    Hopefully I’m not misstating, but this could easily be rephrased as, “What is the physical distinction between a symbolic representation and something that is not but acts like one?”

  45. I would think that you would be rather tired of being lectured by me, Dr Liddle (I am certainly tired of it). But that situation is no doubt tied to the awkward relationship between our two positions. You are the specialist here, but came into this conversation with a completely upended notion of what information really is, and you fought for that notion even after it was shown to be entirely in error. Just imagine someone thinking they could cause the onset of recorded information by no more than mingling digital entities in a simulation.

    And it is not only my habits which are on display here Dr Liddle. Whenever your attacks fail to hit their mark, you tirelessly return to this rather manufactured position of self-effacing politeness; where you consistently forget what the point was, or where the conversation left off, or don’t know what’s being talked about. Apparently, if you cannot defeat your opponents by argument, then you’ll conceal your failure by becoming lost in a pleasant fog of polite misunderstanding. I have noticed recently that several others at UD have become acquainted with your bit, and have called you on it.

    And finally Dr Liddle, it is not I who brings your integrity into question, you have done that yourself. A perfect example is your “What is the question” program; as if you didn’t know you were being asked to point out the physical distinction between a representation versus something that just acts like one. That question actually began with you and Patti (Mathgirl) from months ago. You knew exactly what was being asked.

    Another perfect example is your remark upthread, where you conceded not being able to fulfill your simulation – given that “information” was being defined as it was. You completely skirted the fact that the term was being properly defined, as evidenced by your eventual acceptance of the definition. In other words, you concede that you can’t play basketball, if “basketball” is being defined as a game where you dribble a round orange ball though defenders in order to get closer to a suspended hoop, which you must put the ball through in order to score. Your constant gaming of legitimate dialogue gets truly tiresome Dr Liddle. Any questions of integrity can be laid at your own feet.

Leave a Reply