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. First edit

    quote:

    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

    end quote:

    should be

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

    peace

  2. One problem that I see, is that “random” is observer-dependent, so in some sense subjective. Whether a pattern is discernable depends on how you look at it.

  3. Neil Rickert: One problem that I see, is that “random” is observer-dependent, so in some sense subjective. Whether a pattern is discernable depends on how you look at it.

    Why is this a problem?

    I would say true randomness does not exist. That is of course a philosophical position on my part. What we see as random is only apparent randomness. Something might appear random to me but not to someone else with more information.

    For example the string 15926535 might appear to be random unless it is seen in the following context 3.14159265359. Which is of course Pi

    Once we know the context the string is not random at all but in absence of context it might as well be random.

    What we need is a definition that will work for folks who hold to determinism as well as those who don’t.

    Would you be satisfied if we changed random to apparently random?

    peace

  4. I don’t think the recommended definition of random is usable. I would prefer a definition based on the method of generation. Preferably one useable for encryption.

  5. petrushka: I don’t think the recommended definition is usable. I would prefer a definition based on the method of generation. Preferably one useable for encryption.

    care to share one you feel would be more appropriate?

  6. I don’t see how the definition of observer works. The ability to distinguish artifacts is the issue in dispute.

  7. fifthmonarchyman: care to share one you feel would be more appropriate?

    There are encryption certified algorithms for generating pseudo random strings, or one could use strings generated by radioactive decay. A truly random string would contain every possible finite pattern. So lack of pattern is not a useful definition of random.

  8. petrushka: I don’t see how can’t be explained” can be operationally defined.

    if you define “explanation” as a model produced by a alternative method that an observer can’t distinguish from the string being evaluated then “can’t be explained” simple means that no close nondistinguishable model is available.

    This is all related to definition of integrating function.

    i.e. the knowledge of m (z)does not help to describe m(z′),when z and z′are close.

    found here

    http://arxiv.org/pdf/1405.0126v1.pdf

    You might check out this paper it will be relevant as we move forward

    peace

  9. Neil Rickert: If random is observer dependent, and if “artifact” and “designer” are defined in terms of random, then artifact and designer are also observer dependent.

    Why would that be a problem?

    Like I have always said in the end the debate will always boil down to the problem of other minds. In the mean time there is lots of room for productive work.

    IMO the design inference can be seen as simply a transfer of information between the designer and an observer. It’s simply communication. Communication is at least a little subjective by it’s very nature.

    One feature of the game is that it minimizes the subjective nature of the inference by introducing multiple observers with different conflicting biases and no direct knowledge the particular assay being performed so they won’t be able to consciously cheat.

    peace

  10. petrushka: How do you go about determining that a model cannot be made?

    It’s not that a model can’t be made. it’s that in the case of artifacts modeling is not a computable function.

    There is of course at least one algroythym that can produce a string that is indistinguishable from the original string (the string that produced it in the first place).

    What is at question is whether or not getting to that model is a computable function. If it’s an artifact the answer would be no.

    peace

  11. fifthmonarchyman: It’s not that a model can’t be made. it’s that in the case of artifacts modeling is not a computablefunction.
    There is of course at least one algroythym that can produce a string that is indistinguishable from the original string (the string that produced it in the first place).
    What is at question is whether or not getting to that model is a computable function. If it’s an artifact the answer would be no.
    peace

    I’m sorry, but that makes no sense to me.

  12. petrushka: I’m sorry, but that makes no sense to me.

    have you read the paper?

    petrushka: In your Scheme is biological evolution a designer?

    If it can produce artifacts it’s a designer.
    My hypotheses is that it can’t we shall see
    peace

  13. petrushka: There are encryption certified algorithms for generating pseudo random strings, or one could use strings generated by radioactive decay.

    In the game the random strings could be produced by either of those methods it does not matter.

    It’s just important to say that I am not committing to the existence of true randomness only that there are strings that appear random from the perspective of the observer

    peace

  14. Neil Rickert: I can’t say if it would be, until I know more about where you are going with this idea.

    You will be able to see for yourself when OMagain gets done, right now I’d like to make sure the working definitions are acceptable. Lets not put the cart before the horse.

    We are in bad shape if you need to know the entire argument before you can agree on definitions.

    Neil Rickert: Artifacts are finite, and finite functions are computable.

    I think you are confusing the algroythym that produces an artifact with the function that specifies the algorithm. It might help to think “search for the search”

    Did you read the paper?

    Lets not get ahead of ourselves. There will be plenty of time to discuss the implications of all this later on. If we could I’d like focus on definitions fro now.

    peace

  15. fifthmonarchyman,

    It’s just important to say that I am not committing to the existence of true randomness only that there are strings that appear random from the perspective of the observer.

    Any finite string can appear random, because any finite string can be a substring of any random string. But more importantly, there is no way to prove that a given string isn’t the result of a computation.

  16. petrushka: there is no way to prove that a given string isn’t the result of a computation.

    you are confusing the algorithm that produces the string with the function that selects it

    peace

  17. We can discuss your definitions, but it would be easier to reach a meeting of the minds if you provided operational definitions.

    As it stands you are assuming the existence of things that cannot exist.

  18. petrushka: Any finite string can appear random, because any finite string can be a substring of any random string.

    I would disagree

    The string 123456789 does not appear random according to the definition I’m using

    I see a definite pattern in the string.
    I can distinguish between that string and any other possible string.
    I know the algroythym that produced it

    peace

  19. No. You cannot say it is not a substring of a random string. So why not operationalize your definition of random? Specify the procedure for generating the string.

  20. fifthmonarchyman: I would disagree

    The string 123456789 does not appear random according to the definition I’m using

    I see a definite pattern in the string.
    I can distinguish between that string and any other possible string.

    peace

    I don’t understand this. The string you produced is no more or less likely than any other string produced at random, and it’s possible to find patterns in nearly anything. The idea is, flip a coin enough times, recording the results, and after some large number of flips you can extract substrings which appear highly nonrandom.

    Yes, we are all familiar with argument by cherry-picking.

  21. Your “observer” as defined, cannot exist. If such an oracle existed, there would be no need for this discussion.

    You are also assuming it is possible to prove the nonexistence of models.

  22. Just operationalize your definitions. Please. Before we have another 600 posts that go nowhere.

  23. petrushka: You cannot say it is not a substring of a random string.

    According to my philosophical starting point I can. I begin from the starting point that there are no truly random strings so by definition there are no substrings of random strings.

    I realize that is a philosophical position and not a scientific one but it is not a uncommon belief to hold.

    What we need is a definition that will work regardless of whether or not determinism is true. I think mine does that but I am welcome to any suggestions you may have to improve it.

    Why not post your proposed definition?

    petrushka: So why not operationalize your definition of random? Specify the procedure for generating the string.

    In the game I use a simple random number generator but flipping a coin works just as well. For a string to be effectively random all that is important is that the observer sees no pattern.

    peace

  24. Flint: it’s possible to find patterns in nearly anything.

    I will venture that this is untrue

    This is a hypothesis that we will be able to test together as soon as Omagain is finished.

    peace

  25. fifthmonarchyman,

    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

    Neil already addressed the problem that any finite string is computable, which may be the answer to my question, but I’d like to be sure.

    I’m confused about what you mean both by “a nonrandom object” and “can’t be explained by a computable function…” Can you give an example of an artifact and a non-artifact to demonstrate the difference?

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

    When you say “an observer can’t distinguish” about a string, are you just saying that an “explanation” is an identical string produced by a different method than the first string?

    5. Designer– a being capable of producing artifacts

    Does a designer have to be a “being”? How about a process?

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

    What do you mean by “approximate them”? If an artifact is a string and a model is a string, either the strings are identical or they can be distinguished from each other, right?

    Perhaps if you provided a step-by-step example of you playing your game it would be more clear.

  26. Flint: The string you produced is no more or less likely than any other string produced at random

    Interesting comment.

    Lets say that I wanted to produce a string that was indistinguishable from 123456789 at random with out targeting that particular string. Do you think such a thing is likely?

    peace

  27. fifthmonarchyman,

    For a string to be effectively random all that is important is that the observer sees no pattern.

    That’s a subjective definition of random. As a couple of people have pointed out, you can create an operational definition of random that focuses on how the string is generated rather than what it looks like. There can be patterns in random strings.

    It depends, of course, on the argument you’re making.

  28. Patrick: I’m confused about what you mean both by “a nonrandom object” and “can’t be explained by a computable function…”

    have you read the paper?

    Patrick: Can you give an example of an artifact and a non-artifact to demonstrate the difference?

    that is what the game will do

    Patrick: When you say “an observer can’t distinguish” about a string, are you just saying that an “explanation” is an identical string produced by a different method than the first string?

    no, indistinguishable is not the same thing as identical. It just means that the second string will have the same pattern as the first one for example

    123123123…… is indistinguishable from 023123123….. if the observer views from the second digit on.

    Patrick: Does a designer have to be a “being”? How about a process?

    A process is a “being” if it can produce an artifact according to my definition. I’d be happy to use another word if you have one you like better

    Patrick: What do you mean by “approximate them”? If an artifact is a string and a model is a string, either the strings are identical or they can be distinguished from each other, right?

    see my answer above

    peace

  29. fifthmonarchyman,

    Lets say that I wanted to produce a string that was indistinguishable from 123456789 at random with out targeting that particular string. Do you think such a thing is likely?

    Getting any specific string of nine digits at random is equally likely.

    This raises another question about your definition of “explanation”:

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

    You don’t use the word “random” in that definition, but you ask about finding a particular string at random in your response to Flint. Is there an implicit assumption I’m missing?

  30. Patrick: That’s a subjective definition of random.

    If determinism is true can there be an objective definition of random?

  31. Patrick: Getting any specific string of nine digits at random is equally likely.

    I did not ask about if it was equally likely to any specific string
    I asked if it was likely

    Patrick: You don’t use the word “random” in that definition, but you ask about finding a particular string at random in your response to Flint. Is there an implicit assumption I’m missing?

    I don’t think so. My answer to Flint was more of just a case of curiosity on my part

    peace

  32. fifthmonarchyman: Interesting comment.

    Lets say that I wanted to produce a string that was indistinguishable from 123456789 at random with out targeting that particular string. Do you think such a thing is likely?

    peace

    It has the exact same likelihood as any other 9-digit string.

  33. fifthmonarchyman,

    I’m confused about what you mean both by “a nonrandom object” and “can’t be explained by a computable function…”

    have you read the paper?

    Yes, several times. Is your game just the same game they describe there? If so, why not just quote their description?

    Can you give an example of an artifact and a non-artifact to demonstrate the difference?

    that is what the game will do

    Can you not provide an example now to show what you mean? It would help in understanding your definitions.

    When you say “an observer can’t distinguish” about a string, are you just saying that an “explanation” is an identical string produced by a different method than the first string?

    no, indistinguishable is not the same thing as identical. It just means that the second string will have the same pattern as the first one for example

    123123123…… is indistinguishable from 023123123….. if the observer views from the second digit on.

    You need some more definitions, then. I can distinguish those two strings from each other easily.

    Does a designer have to be a “being”? How about a process?

    A process is a “being” if it can produce an artifact according to my definition. I’d be happy to use another word if you have one you like better

    As long as we’re clear that you’re not assuming anything particular about a “being”, it’s only an issue because of the risk of equivocation.

    What do you mean by “approximate them”? If an artifact is a string and a model is a string, either the strings are identical or they can be distinguished from each other, right?

    see my answer above

    I think you need to be more explicit about what “distinguishable” means in this context.

  34. Patrick: Yes, several times. Is your game just the same game they describe there? If so, why not just quote their description?

    It’s a different paper.

    You are thinking of the one that provided the inspiration for the game, The paper I’m talking about provides mathematical proof that integrating functions are not computable

    Patrick: Can you not provide an example now to show what you mean? It would help in understanding your definitions.

    I could but it will not make sense to you till you actually see the artifact and nonartifact through the “lens” of the game.

    I struggled with posting definitions before the game was completed but I decided to give it a go now mostly because I was anxious to get started. Perhaps that was a mistake.

    It will be sad if you need to know the whole argument before you can agree on simple definitions

    Patrick: I think you need to be more explicit about what “distinguishable” means in this context.

    suppose I had two brand new silver dollars to the unaided observer they are for all intents and purposes indistinguishable but they are not identical.

    Does that help?

    peace

  35. Hmm, I feel that these “definitions” might be a little self-serving, as in “I’ve got something in mind and I’d like to define some terms that will help me prove just the thing that I’ve got in mind”. For example:

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

    I read that 10 times (OK it might have been somewhere between 5 and 10) and I still don’t know what it means (generally and reliably?). How about using a recognized dictionary definition of “observer” such as:

    A person who watches or notices something.

  36. fifthmonarchyman: Interesting comment.

    Lets say that I wanted to produce a string that was indistinguishable from 123456789 at random with out targeting that particular string. Do you think such a thing is likely?

    peace

    Huh? If you wish to produce any given string, that string becomes your target. And, perhaps like Barry, I subscribe to the Euclidian axiom that two strings identical to the same string are identical to each other.

    So the answer to your question becomes, the probability of hitting any target string at random (of the same length, using the same character set) is the same as the probability of hitting any other.

    Assuming each character drawn removes that character from the deck, consuming the entire 9-character set, the probability of drawing any specified string is 1 in 9! (there are 9! possible strings). If each character can be thrown back for the next draw, then the odds are 1 in 9^9.

  37. fifthmonarchyman:
    suppose I had two brand new silver dollars to the unaided observer they are for all intents and purposes indistinguishable but they are not identical.

    Does that help?

    peace

    Yes, that helps a lot. I read once about someone who collected baseball memorabilia. He spent a substantial sum for a baseball that had been hit for a pennant-winning home run. And one day, the neighbor’s kid came in with another instance of the same ball, purchased for $5 at the sports store. Somehow the kid mixed the balls up, and there was no way to distinguish the expensive famous ball from the garden variety ball.

    But that meant the solution to the problem was simple — just take either ball, put it back in the case, and carry on.

  38. Flint: If each character can be thrown back for the next draw, then the odds are 1 in 9^9.

    Ok so at least we are talking about the same thing, 😉

    peace

  39. fifthmonarchyman: The string 123456789 does not appear random according to the definition I’m using

    I see a definite pattern in the string.

    I see a string of 9 symbols with no repeats.

    The only pattern that I see, is that there are 9 symbols with no repeats. If you are seeing a pattern other than that, then it is probably a pattern that depends on other ways that those symbols are used in the culture.

  40. Norm Olsen: How about using a recognized dictionary definition of “observer” such as:

    A person who watches or notices something.

    I like dictionary definitions and I have no problem with that the problem is that when we get to using the game things like “person” and “watches” will become problematic and up for debate.

    If a computer program can reliably distinguish between artifacts and models it would be an observer acourding to my definition but it might not be a “person”.

    Hope that helps

    peace

Leave a Reply