Home » Comp. Sci. / Eng., Complex Specified Information, Intelligent Design, Mathematics, News » Order, Organization, Disorder, Disorganization — the role of specification in perception of design

Order, Organization, Disorder, Disorganization — the role of specification in perception of design

Can I find examples that can fit any of the four following descriptions? Are some even impossible in principle?
[Assume first that the artifact in question has high improbability (like a set of a million coins or as stream of a million bits) or high Shannon entropy. Also assume by "disordered" I mean Kolmogorov simple or algorithmically simple or possessing low algorithmic entropy. Curiously (as far as I know), organization cannot be measured in terms of either Shannon entropy nor algorithmic entropy, it is a transcendent feature that can perceived almost only through subjective specification. Further, thermodynamic entropy isn't even considered in any of these cases (and may not even be applicable)].

1. ordered and organized
2. ordered but disorganized
3. organized but disordered
4. disorganized and disordered

For #1 I would say 1 million fair coins all heads is a set that is both ordered and organized if one admits homogeneity as a form of organization.

For #2, this is hard, it may not even exist in principle, but my best shot at such an example is a car’s parts laid out in alphabetical order in a garage. Clearly the intended organization of the car is not in evidence, so it’s disorganized. But I really won’t press the point if someone contests this as an example of an ordered but disorganized artifact.

For #3, an encrypted ZIP file is an example of an organized but disordered artifact. With the possible exception of some pathological cases, it is highly disordered (perhaps maximally so since it cannot be described much more compactly by a smaller algorithm), yet it is highly organized. If it weren’t organized, decompression of the file would be impossible. A human embryo might be described as a highly compressed representation of an adult human. Thus a human embryo is highly disordered but also highly organized.

For #4 a million coins randomly shaken is both disorganized and disordered

What is curious is a disordered object can be either disorganized or organized. How can we know? I conjecture organization can almost only be perceived through the lens of a subjectively defined specification. For example, without a CODEC and a decoding key, the encrypted ZIP file will look just like gibberish. This also means, one person may see organization in an object while another may not, and in the case of encrypted files, the exclusion of others from perceiving the design through specification is actually the intent!

The beautiful thing about biology is that the specifications are built in, and specification from the world of human engineering also provides specifications to understand biology.

As far as built-in specifications in biology, the most obvious is the ability for biological organisms to function as copy machines of themselves. And not just any copy machine, but a Quine computing copy machine. Hence, one improbable copy (the parent) specifies another improbable copy (the child).

We see also self-specification in biology in as much as DNA specifies proteins. The homochirality in biology serves as a self-specification as well since homochirality is a configuration very far from expectation. As for specification from human engineering, we have many ready-made specifications that serves as design templates for biology such as: coders, decoders, error correction schemes, language processors, sensors, transducers, logic gates, networks, feedback control systems, motors, etc.

Even when anti-ID biologists study the architecture of a biological system, they are studying the organization implicitly through a specification. They use the instances of the explanatory filter without realizing it because biology looks designed. To perceive organization, one needs conceptual templates, and conceptual templates are specifications, and seeing physical objects (like a living creature) that conforms to specifications is what it means to perceive design.

[posted by scordova to assist the News desk with content and commentary through 7/7/13]

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

20 Responses to Order, Organization, Disorder, Disorganization — the role of specification in perception of design

  1. Sal,

    just my point of view:

    First, my definition:

    a) Ordered: any digital sequence (as usual, I will stick to digital information for simplicity) that is highly compressible (IOWs, whose Kolmogorov complexity is much lower than its “raw” complexity).

    b) Organized: any digital sequence for which at least one functional specification can be given. In that sense, “highly organized” means simply that a functional specification can be given which is highly complex (IOWs, a very high number of functional bits is necessary to provide the defined function).

    Keeping in mind those definitions, I would give the following examples:

    1. ordered and organized:

    1 million fair coin tosses all heads is fine with me, if, and only if, it cannot be the result of some algorithm that exists in the system. IOWs, if I get 1 million fair coins all heads in a system where no necessity law can constrain the output to heads, then such a homogeneity can only be the result of design, of conscious volition, aimed to obtain the function of homogeneity. But, as I have argued many times, in the case of highly ordered sequences, necessity explanations should have full priority, because they are well known to have the causal power to generate ordered outputs. In these cases, it can be very difficult to rule put completely any reasonable necessity explanation. However, design is a possible good explanation, if necessity is convincingly ruled out: conscious intervention, indeed, can certainly be a cause of homogeneity and order.

    2. ordered but disorganized

    According to my definitions, this has not much sense. As said, order can always be considered a functional target, so it can very well be considered a specification. But, as said, if the order can be explained by a necessity mechanism, there is no need to see it as organization. So, in that sense, 1 million coin tosses all heads, with a coin that is completely unfair, and can only give heads, would certainly be an example of a sequence “ordered but disorganised”, in the sense that the observed order cannot be considered “organisation” because it is obviously the outcome of a known necessity mechanism.

    3. organized but disordered:

    This is really easy. Almost all functional information is that way. A protein sequence is a good example. A software code. Hamlet.

    Now, I am well aware that all these are in some measure compressible. But that is rather normal, even in completely random sequences. Some of these examples can have higher compressibility than the others, due to the alphabet system, or the basic syntax, that is necessary to express the functional information. But in all cases, the specific function is not in itself obtained by “order”. A functional protein sequence has no special “order” that guarantees the function. The same can be said for a software that does some specific task, or for meaningful, poetic language like Hamlet. While all these examples can certainly be compressed in some measure, none of these is “highly compressible”.

    Let me comment about the zip example. Let’s say that we have a zip file of Hamlet. There is really no substantial difference with the original text. OK, some of the basic order has been compressed, and the file is shorter. And so? The complexity of the compressed file is still huge. OK, we have to know how to unzip it to understand the meaning. And so? In the same way, we have to understand english language. There is no substantial difference. In both cases, the constraints can be easily included in the functional definition. So, for the simple text, the definition will be:

    1a: a sequence of symbols that conveys the exact text of Hamlet, in english language

    And, for the zipped file:

    1b: a sequence of symbols that, when decode by the zip algorithm, conveys the exact text of Hamlet, in english language

    So, for a protein coding gene, the functional definition will be:

    1c: a sequence of nucleotides in a DNA molecule that, when decoded by the genetic code, conveys the aminoacid sequence for a protein with such and such function.

    4. disorganized and disordered:

    This is easy too: I would say, any long sequence truly obtained in a random system. Any sequence of 1 million fair coin tosses, truly empirically obtained by tossing a fair coin, will do.

  2. gpuccio,

    So nice to hear from you! My usage and definitions may not be in agreement with others out there. It may be fair to say few people will agree on proper definitions of these concepts. :-)

    Thanks for you comments!

  3. gpuccio,

    I agree with you that examples of #2 may not be feasible, and with respect to ID even if examples exist, they are irrelevant. The point is moot, but I list it as a hypothetical possibility just for the sake of completeness.

    As far as #1 being regarded organized, whether or not it is a moot point as well if we are dealing with 500 coins or possibly homochirality, this may still indicate design irrespective if we call it organized or not. Examples of #1 are still important because we can bypass the requirement of specification in order to infer design. I dealt with this special case in:
    Siding with Mathgrrl on a point, and offering an alternative to CSI v2.0

    I’m glad we generally agree on #3, that’s a big one.

    We clearly agree on #4

  4. Sal,

    thank you for your comments!

    I think that clear and explicit definitions help a lot.

    I also believe that we essentially agree on all the important points :)

  5. [comment deleted at Alan's request by scordova, local moderator, but not global moderator]

  6. Oops HTML error, I’ll repost

    @ moderator in the ceiling please delete #5

  7. Even when anti-ID biologists study the architecture of a biological system, they are studying the organization implicitly through a specification. They use the instances of the explanatory filter without realizing it…

    Oh really! I’ve wondered for a long time whether anyone could actually demonstrate how to use the “explanatory filter”. Now Sal claims people are doing it without realizing it! Are we back to the trivial process of simple pattern matching as performed at Genetic-ID?

  8. AF: People do use the explanatory filter on a routine, informal basis all the time, e.g. in detecting that comments in UD threads are designed, not credibly lucky noise. Contrary to your assertions in the face of repeated correction, you specifically have been informed repeatedly on how the filter works (e.g. cf here for a 101, especially note the flowchart onlookers). Just, it does not sit comfortably with your ideological agenda. KF

  9. People do use the explanatory filter on a routine, informal basis all the time…

    So KF opts for trivial pattern matching and no scientific utility. Thanks for the confirmation.

  10. LoL! @ Alan Fox

    Alan, The EF is a process that all scientists use if they are following Newton’s four rules of scientific investigation.

    Thanks for continuing to prove that you do not understand science and investigations.

  11. 1. ordered and organized
    2. ordered but disorganized

    As I thought on the matter, there is an important subtlety to keep in mind. Perception of organization is dependent on what set of specifications the observer has to judge organization. One may perceive organization, another may not.

    For example, if my set of specifications were for grammatically correct English language sentences, then the highly ordered set of scrabble letters in alphabetical order:

    aaaaaaaaa….bbbbb……….zzzzz

    could be labeled as disorganized. If one admits alphabetization as a specification, then it looks organized.

    This subjectivity may seem troubling, but it shouldn’t be. At issues whether we are detecting the same subjective patterns that a designer conceived in his mind before making the artifact. We are, in essence trying to match his subjective thought process.

    Hence, in the case of encrypted data, if one is able to break the encryption, one will be able to perceive the designs which others may not be able to. But there is no question, once the encryption is broken, even though the specification is subjectively perceived (via the broken encryption codes), that the object in question (an encrypted file) was indeed the product of a mind or minds….

    Dembski was very smart to perceive that there is huge improbability associated with two minds arriving at the same subjective pattern, hence if one create an object conforming to a pattern and an observer recognized the pattern, design is reasonably inferred.

  12. Hi Alan,

    How do you measure the “triviality” (or lack thereof) of pattern matching?

    Take fingerprinting and DNA matching. Trivial and lacking in any scientific utility?

    By the way, there are numerous resources on pattern-matching. Have you actually ever studied the subject?

  13. Mung,

    Alan uses a dipstick for measuring.

  14. Sal:

    An important point about functional specification which is often misunderstood, is that we can accept any functional specification for a digital sequence. There are no restrictions. The important point is only that the function be explicitly and clearly defined.

    Once a function is explicitly defined, there is no more subjectivity in assessing if it is present in any possible sequence, and in measuring the complexity linked to the function.

    You say:

    “At issues whether we are detecting the same subjective patterns that a designer conceived in his mind before making the artifact. We are, in essence trying to match his subjective thought process.”

    That’s correct, but we must remember that it is only the complexity linked to a defined function that allows us to infer that a mind willed that specification and function in the object we are observing.

    IOWs, we can absolutely define a function in a sequence we read in an object, but that is no guarantee that a mind willed it. If the function we define is simple, it is perfectly possible that such a functional sequence arose by chance. In that case, our mind is the first mind that sees that function in the object/sequence. No other mind “designed” it, and yet the function is there, for us to observe it and define it. A design inference would be unwarranted in such a case.

    But that is true only for simple functions. The whole point of ID is that complex functions have always been observed only as the output of a conscious mind.

    So, the clue to the design inference is only the association, in the same digital sequence, of two things:

    a) Any explicitly definable function

    which

    b) Is complex (can only be provided by a high number of bits of specific information)

    According to our empirical experience, when we have both things, we can safely infer design.

    And I am perfectly confident that we will not be wrong.

  15. How do you measure the “triviality” (or lack thereof) of pattern matching?

    You don’t need to. Synonyms would be obvious, straightforward, simple. You have a fingerprint on file for Joe Bagsnatcher. You find a fingerprint at the scene of a crime and it matches Joe Bagsnatcher’s sample. Trivial process but important result (especially for Joe Bagsnatcher). If Joe has not previously had his fingerprints taken and filed, the print at the scene can’t help other than to eliminate other suspects and/or we suspect Joe Bagsnatcher for other reasons and compare his prints. We can’t identify a criminal from a print without having a comparison. Ditto for DNA.

    Take fingerprinting and DNA matching. Trivial and lacking in any scientific utility?

    You may be able to answer this yourself, now.

    By the way, there are numerous resources on pattern-matching. Have you actually ever studied the subject?

    That doesn’t surprise me, seeing that all sentient beings function by pattern matching. Can you recommend a good layman’s guide?

  16. …it is only the complexity linked to a defined function that allows us to infer that a mind willed that specification and function in the object we are observing.

    Hi gpuccio

    When you say “a mind willed” are you including processes such as a person thinking of something and then acting on it or are you talking of discontinuities, of disembodied “minds” violating the second law? Or both?

  17. Hi, Alan:

    I just mean that some being, capable of subjective representations, outputted the desired form to some material object, through his will and consequent actions.

    In the case of human beings, who are conscious beings with a physical body, the output is obviously realized through their physical actions, which convey the desired form to the material object.

    In the case of biological design, as I have argued many times, different hypotheses can be made.

    My personal scenario is: non physical conscious agents, interacting with biological matter at quantum level.

    In essence, the scenario is not so different from human design, that we witness daily, because I do believe that human consciousness similarly interacts with the human brain at quantum level. So, essentially, the main difference would be that our consciousness can only interact with its physical body, and particularly the brain cells, while the biological designer(s) ‘ consciousness interacts directly with biological matter.

    I never make statements about the second law, because I am not a physicist. But I can certainly state that a mind -matter interaction at quantum level needs not violate any physical law, but can certainly violate probabilistic laws, for example generating dFSCI where no other physical system can do the same.

    So, in a sense, the only law that would be “violated” would be the intrinsic probability of quantum mechanics ‘ “wave function collapse”, which would be “guided”, on occasion, to intelligent results, instead of operating in complete randomness. If that can be considered a violation of the second law, it is not for me to tell.

    The only “discontinuity”, then, is the emergence of dFSCI where no physical system can explain it: a discontinuity of functional information, but not of the cause and effect chain that we observe in the macro-world (because the effect would be modeled at quantum level).

    I hope that is clear enough.

  18. I hope that is clear enough.

    Yes, indeed. Regarding the second law, I found physics the dullest of the sciences at school and never got to grips with entropy. I was taught entropy = disorder and I’m only now beginning to get my head around the idea of micro-states and the like.

    Playing Devil’s advocate for a moment, if I had to explain God’s guiding hand on the cosmos, I would pick apparent randomness. The celestial designer could influence the outcome of apparently stochastic events without (apparently) creating discontinuities. This does not seem a million miles from your quantum collapse.

    I guess where we differ is I see no need to postulate these hypotheses as the simpler one that random events are in reality random seems to work just as well for me. If dFSCI (digital functional specified complex information?) works for you, then fine. I am absolutely sure that you are not going to use it as an excuse for world domination! :)

  19. Alan:

    “This does not seem a million miles from your quantum collapse.”

    It is very much the same concept.

    “I guess where we differ is I see no need to postulate these hypotheses as the simpler one that random events are in reality random seems to work just as well for me.”

    OK, but certainly not for me.

    “If dFSCI (digital functional specified complex information?) works for you, then fine.”

    It does! :)

    “I am absolutely sure that you are not going to use it as an excuse for world domination!”

    Not a priority. At present, it is not even a good tool to get any advantage in my family :)

  20. Alan:

    And yes, dFSCI means: digital Functionally Specified Complex Information.

Leave a Reply