Working Definitions for the Design Detection Game/Tool

I want to thank OMagain in advance for doing the heavy lifting required to make my little tool/game sharable. His efforts will not only speed the process up immeasurably they will lend some much needed bipartisanship  to this endeavor as we move forward. When he is done I believe we can begin to attempt to use the game/tool to do some real testable science in the area of ID . I’m sure all will agree this will be quite an accomplishment.
Moving forward I would ask that in these discussions we take things slowly doing our best to leave out the usual culture warfare template and try to focus on what is actually being said rather than the motives and implications we think we see behind the words.

 

I believe now would be a good time for us to do some preliminary definitional housework. That way when OMagain finishes his work on the gizmo I can lay out some proposed Hypotheses and the real fun can hopefully start immediately.

 

It is always desirable to begin with good operational definitions that are agreeable to everyone and as precise as possible. With that in mind I would like to suggest the following short operational definitions for some terms that will invariably come up in the discussions that follow.

 

1.      Random– exhibiting no discernible pattern , alternatively a numeric string corresponding to the decimal expansion of an irrational number that is unknown to the observer who is evaluating it

2.       Computable function– a function with a finite procedure (an algorithm) telling how to compute the function.

3.       Artifact– a nonrandom object that is described by a representative string that can’t be explained by a computable function that does not reference the representative string

4.      Explanation –a model produced by a alternative method that an observer can’t distinguish from the string being evaluated

5.       Designer– a being capable of producing artifacts

6.       Observer– a being that with feedback can generally and reliably distinguish between artifacts and models that approximate them

Please take some time to review and let me know if these working definitions are acceptable and clear enough for you all. These are works in progress and I fully expect them to change as you give feedback.

Any suggestions for improvement will be welcomed and as always please forgive the spelling and grammar mistakes.

peace

541 thoughts on “Working Definitions for the Design Detection Game/Tool

  1. fifthmonarchyman: Would you say that “natural phenomenons” can produce patterns that are nonrandom and noncomputable?

    can you provide some examples?

    peace

    Honestly, It’s still unclear to me what you mean by non-computable or what does that have to do with any design inference. Can you provide an example for non-computable and explain why it implies design? Maybe an example of a non-computable entity that was designed?

  2. fifthmonarchyman: I would strongly disagree, Market returns are the artifact of millions of discretedecisions by conscious agents. I can’t see how you could possibly think of them as anything but designed

    The individual components of the results may be designed (even that is questionable), but the aggregate certainly is not. Consider a rubbish dump: every object making up the dump is designed and manufactured, but the dump itself is happenstance and not designed (note that I am not talking about the location of the dump, or its external shape, but of its internal organisation).
    ETA : the internal organisation of a rubbish dump would end up at the Law + Chance prong of the EF.

    It doesn’t have to be designed by me.

    The game is simply a nifty means for us to see if a particular real string can be distinguished from fake (randomized or algorithmic) ones.

    I am still not quite clear on what the game does and how it does it. As far as I can tell it is going to demonstrate that humans are rather poor at creating random numbers and rather good at spotting non-random patterns. This has of course been known for a long time.

    Here is a suggestion to build some more scientific rigour into the game: run each designed string and each fake through a random number test and analyse the results after a number of runs. I suspect that you will find that the designed strings are significantly less random than the fakes. That is why you can tell them apart.

    There is an evolutionary explanation for this, by the way.

    fG

  3. dazz: Honestly, It’s still unclear to me what you mean by non-computable or what does that have to do with any design inference. Can you provide an example for non-computable and explain why it implies design? Maybe an example of a non-computable entity that was designed?

    from my definitions

    Quote:

    Computable function– a function with a finite procedure (an algorithm) telling how to compute the function.

    end quote:

    a non-computable entity is an entity that can’t be modeled algorithmicly,

    There is simply no way to produce it by just following a recipe.

    In the Intention, Intelligence and Teleology thread I asked EL if a computer program could produce an object that was indistinquishable from a real snow flake she said

    quote:

    I think it would be possible to model the snowflakes so well that they would be indistinguishable from a photograph of the real thing.

    I didn’t try that hard – I didn’t attempt to simulate anything much like the actual process, although I could have done. But mine differ from real snowflakes in that they just have hexagonal symmatry. The 60 degree angle isn’t coded into the patterns along the arms, only into the fact that there are six arms 60 degrees apart.

    I also picked out the most snowflake ones! And did a bit of post processing as well, to smooth them a bit.

    end quote:

    It’s those last two sentences that are the hallmark of design!!!!!!!!!

    they define and illustrate what a noncomputable process is.

    Do you get this ???

    peace

  4. faded_Glory: The individual components of the results may be designed (even that is questionable), but the aggregate certainly is not.

    Again I would strongly disagree the market is designed to achieve the most fair price for goods and services. Market returns are precisely the artifact of that process

    faded_Glory: Consider a rubbish dump: every object making up the dump is designed and manufactured, but the dump itself is happenstance and not designed

    Market returns are not a garbage dump but the deliberate record of millions of transactions each one the result of a conscious decision by an intelligent agent.

    peace

  5. faded_Glory: I am still not quite clear on what the game does and how it does it. As far as I can tell it is going to demonstrate that humans are rather poor at creating random numbers and rather good at spotting non-random patterns. This has of course been known for a long time.

    Well evidently you don’t have a clear idea of what the game will do, This is understandable since I have not detailed the exact processes involved or the assays that will be conducted at this point.

    Perhaps we will just have to wait. Do you see why I’m so anxious to get this out there?

    Once again I could share it with you directly if you like. I’d love to get feed back

    peace

  6. fifthmonarchyman: a non-computable entity is simply an entity that can’t be modeled algorithmicly

    How do you go about determining whether something can or cannot be modeled algorithmically? You can’t just claim something is non-computable just because there’s no known way to do it. That’s an argument from ignorance (a classic in ID rethoric)

    fifthmonarchyman: a non-computable entity is simply an entity that can’t be modeled algorithmicly

    Please give an example of something designed that is not computable. You need to support your claim that non-computable implies design. We as humans happen to design models all the time. That’s how we go about understanding reality. We model it, and I guess those models are all computable under your definition. Nothing we ever design is non-computable, so all you can do at this point is to beg the question that something else (life?) must be non-computable and designed

  7. dazz: How do you go about determining whether something can or cannot be modeled algorithmically?

    Did you read the paper? The authors presented mathematical proof that integrating functions are non computable.

    The only question is whether or not a particular object was produced by an integrating function. That is the point of the game

    peace

  8. fifthmonarchyman,

    from the paper

    quote:

    In the following section we formally prove that, given the
    Partial Information Decomposition (Williams & Beer, 2010)
    formulation of synergy, the amount of integrated informa-
    tion an information-lossless process produces on statistically
    independent inputs is equivalent to the data compression it acheives

    end quote:

    This has nothing to do with the problem I pointed out with the paper (although that section does have its own set of problems). Here’s the summary of the paper’s argument, again:

    1) Conscious observation must be integrated with previous memories, which generats integrated information.

    2) Memory functions must be vastly non-lossy, otherwise retrieving them repeatedly would cause them to gradually decay.

    3) Lossless information integration cannot be achieved by a computable process.

    4) Therefore we have to abandon either the idea that people enjoy genuinely unitary consciousness or that brain processes can be modelled computationally.

    The part you quoted is used in conjunction with (2) to support (4). Since (2) is demonstrably false, the conclusion in (4) is not supported.

  9. dazz: We as humans happen to design models all the time. That’s how we go about understanding reality. We model it, and I guess those models are all computable under your definition.

    exactly

    dazz: Nothing we ever design is non-computable, so all you can do at this point is to beg the question that something else (life?) must be non-computable and designed

    No we produce lots of things things that can’t be modeled the can only be copied .

    The Mona Lisa is a good example.

    Any “fake” painting will be distinguishable from the Mona Lisa. Unless it specifically targets the original and we make “post processing” adjustments.

    Get it now?

    Peace

  10. fifthmonarchyman,

    In an effort to give you every benefit of the doubt I have reread your “response” to the paper

    Patrick: Remember, they aren’t talking just about cognition, they are talking about “the requirement for a conscious observation to be integrated with previous memories, hence generating integrated information.”

    quick question

    In your opinion what is “just” cognition? How can cognition occur with out integrating an experience with previous memories?

    It doesn’t. That’s the point. As I already quoted from the paper:

    Tononi (2008) explains the foundations of his theory through two thought experiments, which we adapt slightly here. The first thought experiment establishes the requirement for a conscious observation to generate information. The second establishes the requirement for a conscious observation to be integrated with previous memories, hence generating integrated information.

    The integration of memories is essential to the argument being made in the paper. Unfortunately for the authors, their claim about memories:

    While it seems intuitive for the brain to discard irrelevant details from sensory input, it seems undesirable for it to also hemorrhage meaningful content. In particular, memory functions must be vastly non-lossy, otherwise retrieving them repeatedly would cause them to gradually decay.

    has been proven to be incorrect. Memories do change when they are recalled. The “memory function” they talk about is not “vastly non-lossy”.

    This makes the conclusion of the paper unsupported.

  11. Patrick: Since (2) is demonstrably false, the conclusion in (4) is not supported.

    Again 2 is not demonstrably false.

    Each time I recall my phone number the memory does not degrade my impression of it gets stronger.

    Think back to when you memorized stuff for school

    That whole point was to repeatedly recall the information till it became strongly ingrained in your mind.

    If the memory function was lossy each time you recalled a fact it would become more fuzzy in your consciousness not more clear.

    Why can you not see this? It is literally elementary school stuff

    peace

  12. If the memory function was lossy the way to forget something would be to dwell on it as much as possible.

    The whole idea is ludicrous

    peace

  13. fifthmonarchyman: No we produce lots of things things that can’t be modeled the can only be copied .

    The Mona Lisa is a good example.

    Everything physical can be modeled, if we don’t require that the model be an exact replica.

    Nothing physical can be modeled, if we do require that the model be an exact replica.

    That is to say, it is all a nonsense argument that allows you to find a gap where you can put your “god of the gaps” wherever you feel like putting it. It is all about finding ways of deceiving yourself.

  14. Neil Rickert: Everything physical can be modeled, if we don’t require that the model be an exact replica.

    Nothing physical can be modeled, if we do require that the model be an exact replica.

    Again we are not talking about identical but indistinguishable.

    An observer will always be able to tell the difference between the fake and the real thing.

    that is what separates true observers from artificial ones

    It is not a gap it is the definition of noncomputable

    peace

  15. fifthmonarchyman,

    If the memory function was lossy each time you recalled a fact it would become more fuzzy in your consciousness not more clear.

    You are attempting to counter scientific evidence with anecdote and emotion. The fact is that memory function is not “vastly non-lossy” as claimed and required by the paper you’re touting.

  16. fifthmonarchyman: Again I would strongly disagree the market is designed to achieve the most fair price for goods and services. Market returns are precisely the artifact of that process

    Market returns are not a garbage dump but the deliberate record of millions of transactions each one the result of a conscious decision by an intelligent agent.

    peace

    Well we have to agree to disagree. I think it is obvious that nobody designs the shape of the Dow Jones index over time. That does not mean it is random, of course. It will be influenced by many factors. Some of those are by design, many are simply happenstance. What the experiment shows is that people can distinguish more random graphs from less random ones. Interesting, but hardly news.

    fG

  17. Patrick: You are attempting to counter scientific evidence with anecdote and emotion.

    Your “scientific evidence” is about reminiscence not about data compression and is therefore laughably irrelevant to the subject at hand.

    Why you can’t see this is beyond me

    peace

  18. faded_Glory: It will be influenced by many factors. Some of those are by design, many are simply happenstance.

    I agree,

    Teasing out the happenstance (random) and the externally constrained from the design is what modeling is all about.

    peace

  19. fifthmonarchyman,

    You are attempting to counter scientific evidence with anecdote and emotion.

    Your “scientific evidence” is about reminiscence not about data compression and is therefore laughably irrelevant to the subject at hand.

    No, the scientific evidence is about how memory works and it directly counters the claim made in the paper, upon which the paper’s conclusions depend.

    If you aren’t willing to recognize when one of your hypotheses is disconfirmed, you don’t really want to talk about science.

  20. What are the rules of the game. Could anyone provide the rules for generating the test strings?

  21. fifthmonarchyman: The Mona Lisa is a good example.

    Any “fake” painting will be distinguishable from the Mona Lisa. Unless it specifically targets the original and we make “post processing” adjustments

    Sorry, this is silly. As Neil pointed out, models are never going to be perfect representations of the object modeled, but you seem to go even further by essentially claiming that whatever the model is, it can at best be a copy of the original. I mean if we could one day replicate the Mona Lisa, molecule by molecule, that still would be a copy right? If you do that, then essentially everything is non computable. One could apply the same reasoning to any other thing, including all sort of naturally produced entities. After all if snowflakes can only be “copied” or modeled, then they’re non-computable and we could therefore infer design don’t you think?

  22. fifthmonarchyman: Again we are not talking about identical but indistinguishable.

    When we are talking about physical things, “identical” and “indistinguishable” are indistinguishable.

    An observer will always be able to tell the difference between the fake and the real thing.

    that is what separates true observers from artificial ones

    The “no true observer Scotsman” argument.

  23. Neil Rickert: The “no true observer Scotsman” argument.

    actually it is a testable hypothesis. That Patrick said he would falsify with a simple little two week hack.

    That was several months ago

    peace

  24. dazz: I mean if we could one day replicate the Mona Lisa, molecule by molecule, that still would be a copy right?

    QM insures that this will never happen

    dazz: After all if snowflakes can only be “copied” or modeled, then they’re non-computable and we could therefore infer design don’t you think?

    Snowflakes are designed

    peace

  25. fifthmonarchyman: exactly

    No we produce lots of things things that can’t be modeled the can only be copied .

    The Mona Lisa is a good example.

    Any “fake” painting will be distinguishable from the Mona Lisa. Unless it specifically targets the original and we make “post processing” adjustments.

    Get it now?

    Peace

    Are you aware of the progress that has been made by computerised music composing? There are now many compositions that listeners cannot tell apart from ones composed by humans. Some have even been performed by top orchestras like the LSO.

    Have a read and listen here.

    Interestingly, this particular program, Iamus, uses evolutionary algorithms to compose its music!

    fG

  26. fifthmonarchyman: Your “scientific evidence” is about reminiscence not about data compression and is therefore laughably irrelevant to the subject at hand.

    Non-lossy compression: it is possible to extract the original (identically).

    Lossy compression: it is possible to extract something which the human mind cannot easily distinguish from the original.

    The human mind is pretty much the paradigm example of lossy.

  27. petrushka:
    Did anyone answer my question about what makes fake different from real?

    The way I understand it is that a fake is a perturbed copy of the original. I suspect that in the process it becomes somewhat more random, which is why observers can tell the difference.
    But I may be wrong. I think it is about time for fmm to stop being coy and show us their work.

    fG

  28. I don’t understand the observer thing. There are lots of claims being made about the abilities of observers and fakes and such that to my simple mind appear to be untrue.

    Lots of things have been counterfeited.

    I would like some clarification here that isn’t simply assertion.

    Simple case in point. A number of consumer testing labs have tried to do blind taste and smell tests on brands of vanilla. Some have rated artificial vanilla higher than natural, even though instruments can detect that the artificial is simpler chemically. The trained human nose can’t reliably distinguish them.

  29. fifthmonarchyman: QM insures that this will never happen

    Snowflakes are designed

    peace

    QM apply to the subatomic level, not the molecular level

    …and snowflakes designed? WTF? LOL

    And you know that because they’re non-computable, and you know they’re non-computable because that means they’re designed, right?

  30. dazz: And you know that because they’re non-computable, and you know they’re non-computable because that means they’re designed, right?

    No I know they are noncomputable because they pass the test of the game.

    I know they are designed because their ability to pass the test of the game makes them an artifact according to the definitions we are using here.

    peace

  31. Guys,

    I’d like to point out that EL claims that she can algorithmicly model a fake that is indistinguishable from an artifact and Patrick claims that he can build software that will be able to generally and reliably distinguish fakes from artifacts.

    I contend that they will not be able to do what they claim they can do.

    regardless of anything else this is a profound development

    We have at least two ID Hypotheses and a means to test them and folks with motivation to do so

    I love science

    peace

  32. fifthmonarchyman: I know they are designed because their ability to pass the test of the game makes them an artifact according to the definitions we are using here.

    Why go through so much trouble then, when you can simply define whatever you want to be designed?

    Can’t you see this is useless and proves absolutely nothing?

  33. faded_Glory: Are you aware of the progress that has been made by computerised music composing? There are now many compositions that listeners cannot tell apart from ones composed by humans. Some have even been performed by top orchestras like the LSO.

    Yes.

    It is my hypothesis that as cool and interesting and these models are they will never be indistinguishable from the original that they are modeling unless the algroythym targets the actual specific information the artifact contains

    peace

  34. fifthmvonarchyman,

    Snowflakes are designed

    Do you have any objective, empirical evidence to support that claim?

    Yes. That is what the game does

    How exactly does your game provide objective, empirical evidence that snowflakes are designed?

  35. dazz: Can’t you see this is useless and proves absolutely nothing?

    Science can not prove anything it can only falsify a hypothesis

    dazz: Why go through so much trouble then, when you can simply define whatever you want to be designed?

    because I don’t want to “define what ever I want to be designed” I want a tool to use to help me objectively determine if an object is designed.

    I believe I might have one

    time will tell

    peace

  36. Patrick: How exactly does your game provide objective, empirical evidence that snowflakes are designed?

    It provides objective empirical evidence that a snowflake is nonrandom and noncomputable.

    That according to our definition is what an artifact is

    peace

  37. fifthmonarchyman,

    How exactly does your game provide objective, empirical evidence that snowflakes are designed?

    It provides objective empirical evidence that a snowflake is nonrandom and noncomputable.

    You’ve left out a lot of steps between the observation that “snowflakes exist”, some as yet poorly defined game, and the conclusion that snowflakes are designed. Please fill in the blanks with a rational argument supported by objective, empirical evidence.

  38. Neil Rickert: Non-lossy compression: it is possible to extract the original (identically).

    Lossy compression: it is possible to extract something which the human mind cannot easily distinguish from the original.

    The human mind is pretty much the paradigm example of lossy.

    oh contraire

    The original in this case is the image in the human mind not the actual phyisical thing itself.

    We can easily extract that or things like computer passwords and phone numbers would be impossible

    peace

  39. Patrick: Please fill in the blanks with a rational argument supported by objective, empirical evidence.

    Be patient or jump in and give OMagain a hand

    peace

  40. fifthmonarchyman: oh contraire

    The original in this case is the image in the human mind not the actual phyisical thing itself.

    peace

    So the “original” is dynamic, constantly changing and fading. You are using a moving target.

  41. Flint: So the “original” is dynamic, constantly changing and fading. You are using a moving target.

    No the original is static, I’ve had the same phone number for decades

    peace

  42. fifthmonarchyman: Yes.

    It is my hypothesis that as cool and interesting and these models are they will never be indistinguishable from the original that they are modeling unless the algroythym targets the actual specific information the artifact contains

    In fact some of this music is already indistinguishable from the original (qua style) – the software does not aim to produce exact copies of already existing music but rather new music in a certain style. People familiar with that style of music can in many cases not distinguish the copy from an original.

    You now seem to add a new requirement: unless the algroythym targets the actual specific information the artifact contains

    If music is composed using EA’s it will be constrained by fitness functions which will presumably reflect the rules and conventions of the style of music it is aiming at. This is necessary because music is not just random sound patterns. Likewise, a human composer will be constrained in similar ways when she produces an original. I don’t see a reason why such a requirement would be a problem for artefacts and not for originals. Mona Lisa has been painted using rules and information from the environment as well.

    fG

  43. petrushka: What are the rules of the game. Could anyone provide the rules for generating the test strings?

    1) a “fake” model must be algorithmic and it must not target the actual specific digits of the real string
    2) a “fake” random string can be produced by any of the generally accepted randomization processes
    3) a “real” string is simply a numeric representation of some object or phenomena

    peace

  44. unless the algroythym targets the actual specific information…

    A perfect illustration of the lossiness of fifth’s memory. No matter how many times we correct him, he continues to misspell ‘algorithm.’

Leave a Reply