Recently I described a possible alternative Turing test that looks for certain non-person like behavior. I’d like to try it out and see if it is robust and has any value.
Here is some data represented in an ordinary control chart.
It’s easy to see there is a recognizable pattern here. The line follows a downward trajectory till about observation 16 then it meanders around till about observation 31. At which point it begins an upward track that lasts till almost the end of the chart.
The data in question is real and public but obsolete. I won’t say what it’s source is right now to avoid any bias in your attempts see if we can infer design. I can provide the actual numbers if you like.
The question before the house is. Is the clear overall pattern we see here a record of intention or does it have a “non-mental” cause?
Using the criteria discussed in my “poker” thread I would suggest looking for the following behaviors and excluding a mental cause if they are present.
1) large random spikes in the data
2) sudden changes in the overall pattern of the data that appear to be random
3) long periods of monotony
4) unexplained disjunction in the pattern.
I have some other tests we can look at as well .
What do you say design or not? Are you willing to venture a conjecture?
peace
Every scientific paper defines the terms. Failure to define the terms = not science. Turing’s 1950 article was a proposal, not a report of experiments, and even so it speaks about definitions already in the second sentence because that’s really how science is done. And he had defined the terms in the work that led up to the article. Please wake up and grow up at least to the level of a fifth-grader.
Or continue in your denial while calling everybody else to be open-minded. Your choice.
That’s not how you state it. We are able to recognize some behaviors as human because we know the capabilities and limitations of humans.
It a good thing I’m not leaping from there then
That’s exactly what you’re doing when you talk about minds in the abstract rather than actual human beings. It’s very similar to the IDCists trying to use evidence of human design in support of their omnipotent designer.
Like I said I’m not all that interested in whether the test is “science” or not. That is your hang up
I’m looking for useful.
peace
Uh what do you think I’m doing?
peace
I’d say we are able to recognize behaviors as person-like because we have experience with persons.
Humans have minds it’s what makes us human and not zombies
peace
Actually the spelling errors are just a symptom of my general problem with detail. I’m more interested in the big picture
I sometimes post ideas here in hope that I will get some perspective from more detail orientated folks.
with varying rates of success
😉
peace
This qualifies as useful, if you are in total denial about the feedback you are getting. And of course, in denial you are.
You are doing sheer nonsense. This was objectively detected in the first comments already.
Now you’re just repeating your unsupported assertion again. That doesn’t make it better supported.
Suggesting that others are not participating in good faith is against the rules.
In addition to being rule violating and rude, you are also wrong. Operational definitions are required for any objective test.
Your “test” has not been demonstrated to have any value and the fact that you refuse to operationalize your definitions explains why it will continue to be useless. You are writing literal nonsense.
All available evidence supports the conclusion that humans have minds because we have complex brains and minds are what brains do.
I am not at all convinced that ‘having minds’ or not is a binary proposition. Why can’t ‘minds’ come in types and sizes, different for different types of living beings? And if we don’t think of ‘mind’ as a thing, but as a process (or a set of processes), why can’t other living beings be capable to run some or all of the same processes that humans do, even if to a different degree? Just look at that long list of mental capabilities observed in animals.
Animal cognition
and
Theory of Mind in animals
Only if we define ‘mind’ strictly on the basis of human capabilities, carefully honed so that it excludes every other living being, can you say that ‘having minds is what makes us human’. All this does is erect a wall between ourselves and other species, or machines for that matter. And as we all know, the point of walls is to stop progress, not facilitate it.
We agreed that there is a huge grey zone where graph inspection isn’t going to work. Surely that in itself should tell you something.
There’s that define word again. Why can’t you have an open mind? It’s amazing at what lengths normally rational people will go to avoid even thinking about things they don’t like.
Let’s simply agree that FMM’s project is useful and practical and call it a day. FMM, you win. Everybody else lose.
That’s surely not right.
All mammals can be said to have minds. But that does not make them human.
And, of course, there’s the big question of what does it mean to “have a mind”.
Some people would say that computers have minds, yet are zombies.
There’s currently a series of posts at “The Brains Blog” on whether plants have minds.
Are minds even important? Yes, they are important to humans. But it doesn’t follow that they are important to other organisms.
Like whether you listen a little to others, or not at all.
Anyway, you can’t get the big picture without the details. More importantly, meaningful big pictures come from the details, the facts, rather than from your presuppositions. That’s why you just churn away insisting that you’re right rather than making any kind of progress.
Glen Davidson
Your whole problem here, you think that if you claim something it’s true unless it’s shown to be false. And it can’t be shown to be false per se, in most cases.
It’s your entire approach that is completely wrong, as there are endless propositions that can’t be shown to be false that remain useless and without any kind of evidentiary support (can the “how do you know?” nonsense until you begin to understand how people do know what you don’t). Your religious presuppositions, for instance. Come up with the first reason why we should believe that you’re right on any significant matter and you might come up with something not useless–for once.
Glen Davidson
WTF? If a test is not scientifically valid, it is not a useful test.
I have just been looking at it as providing feedback to fmm. I’m inclined to think that his project is misguided. But that’s not my problem to deal with.
In the end, I expect that he will conclude that he has reached the conclusion that he expected. But he will find the evidence for that is so mushy that he cannot persuade anybody.
fifth,
As Glen notes, you won’t get the “big picture” right if you don’t deal properly with the details.
That would be the most efficient response, I suppose.
Useful is not a synonym for scientifically valid
History tests are useful and not scientifically valid
Piano auditions are useful and not scientifically valid
Romantic compatibility tests are useful and not scientifically valid
Even spelling tests can be useful but not scientifically valid.
My test being useful is what is important to me. I could care less if it fits some arbitrary label.
peace
I know it’s useful and practical. I already use it after all.
I’m trying to see if it can me made universal.
I’m working on a new OP to give you a better idea of how I use the test right now so you can see what I’m talking about.
stay tuned
peace
The problem is detection of mind / non-mind though graphical representations might not be one of those things it is useful for.
Then again maybe it is, either way I don’t think one or two examples is enough to know.
Of course they are. If it is not scientifically valid (ie., does not accurately measure a person’s basic knowledge of the subject) then it is useless.
Of course they are.
Again, of course they are.
No they can’t.
There is nothing arbitrary about the requirement for a test to be scientifically valid for it to be useful. For any test to be usefull it must be both reliable (ie., repeatable) and valid (ie., accurately measure what the test is supposed to measure). Since your test does not meet either of these criteria, it is no more useful than phrenology, numerology, palm reading or reading tea leaves. Just because you refuse to accept the truth when it is shown to you does not make your test any more usefull.
Perhaps FMM should try writing a book, or claiming that his method does not work because of persecution by the darwinist cabal. It’s what everyone else always seems to end up doing.
Subjects test don’t measure a persons “basic knowledge of a subject” they measure an entity’s ability to give the preferred answer to questions that are asked.
A computer can score 100% on a spelling test and not know anything all.
I have no problem with these criteria but they do not equate to scientific validity. That is unless we redefine “science”.
who said my test does not meet these criteria? On what basis do you make the claim?
I’m more than happy to accept the truth when it’s shown to me.
When exactly did you do that when it comes to my test being not repeatable or valid? Be specific please
peace
You need to be able to spell before you can write a book. 😉
I think I’ll stick to playing around with Turing tests
peace
Says the guy who just told us that computers are aces at spelling.
Seriously, fifth — have you never wondered about that red squiggly line that appears when you type “algroythymic”?
I don’t spend a lot of time worrying about red squiggly lines.
I do from time to time wonder about people who would waste time focusing on such things 😉
quote:
“I was terrible in English. I couldn’t stand the subject. It seemed to me ridiculous to worry about whether you spelled something wrong or not, because English spelling is just a human convention–it has nothing to do with anything real, anything from nature. Any word can be spelled just as well a different way.”
end quote:
Richard Feynman
peace
quote:
I don’t see any use in having a uniform and arbitrary way of spelling words. We might as well make all clothes alike and cook all dishes alike. Sameness is tiresome; variety is pleasing.
end quote:
Mark Twain
Since you don’t really need to know any thing to write a book computers should be pretty good at it
😉
peace
If a subjects test does not test basic knowledge of the subject, then it is useless.
For something to be valid it is, by definition, scientifically valid. If not, it is conjecture, rumour or bullshit.
On the basis that you have not demonstrated that they can measure what you are intending to measure (not valid), and are not repeatable (not reliable). Feel free to demonstrate that I am wrong.
The fact that your test is not repeatable or valid is about as specific as anyone can get. The burden of proof is on you to prove that your test is repeatable and valid. When you can do this, you will have a usefull test. Until then, it is useless.
Let us know when you start doing this. I would be interested.
That is your unsupported opinion of the situation.
The simple fact is that subject tests don’t test basic knowledge of anything. They test the ability to answer questions in a way that we choose to associate with basic knowledge of the subject.
If that is the case then the word “science” is simply a redundancy. Why not just say valid and be done with it? Why the need for an additional word?
One what basis do you equate “not demonstrated to be X” with “is not X”.
I’ve made no attempt to demonstrate anything here I simply asked for input on a test I’m thinking about.
Perhaps if I was interested in making some kind of claim right now. However what I was doing is looking for some constructive criticism and some folks willing to make a guess instead of being afraid.
peace
And you wonder why people hold you in contempt? “No, the fact is that the sky is not blue. It simply presents to us a color which we choose to SAY is blue.” By this reasoning, any effort to know anything is a waste of time, because knowledge is not possible. In which case, how could you possibly get constructive criticism?
Fortunately, contempt is not fear. Even if you choose to SAY it’s fear.
If that has been your experience with subject tests then you went to the wrong school. but that probably explains a lot.
Then why were you insistent that you were trying to develop a test that was useful, not scientifically valid. The fact still remains that for a test to be useful, it must be valid.
It has repeatedly been pointed out to you why a graphical Turing test to identify design cannot work unless you provide so much context that the design is implicit at the outset. Yet, you keep asking for constructive advice. This is me giving you constructive advice.
You have begin making a claim from the beginning of this thread that a graphical Turing test to identify design (mind, intention, whatever) should be possible.
And constructive criticism has been given by several people. Telling you that you should stop beating a dead horse is constructive criticism.
keiths:
fifth:
We’ve had this conversation before. Have you forgotten already?
You quoted Feynman, and I replied:
Do you honestly think Turing tests are a dead horse? And they call ID a science stopper
I have done no such thing. I’ve only asked if a Turing test that is not constrained to language based communication is possible.
Turing tests identify person-like behavior we often associate such behavior with mind, intention. whatever
I think I’ve made it abundantly clear that I think it’s impossible to definitively say that anything has a mind/intention.
The problem is that you have not specified how much context is necessary in standard a Turing tests so I have no idea if your “criticism” is hypocritical or not.
Once you establish the amount of context you require in a standard Turing test we can work on duplicating that in an alternative one.
If fact my next OP will supply additional context that I think goes beyond what you find in a typical standard Turing test. Stay tuned
peace
I agree, and you are no Leighton 😉
quote:
“What Do You Care What Other People Think?”
end quote:
Richard Feynman
peace
Acartia:
fifth:
So when you titled this thread A Practical Exercise in Design Detection, you weren’t claiming that it would in fact be a practical exercise in design detection?
I see.
Fmm:
Turing tests identify person-like behavior we often associate such behavior with mind, intention. whatever
Not according to Turing himself, they don’t. It appears that you still haven’t read his paper.
Read it.
Note especially his discussion about the possible objection based on the ‘Argument from Consciousness’. He finishes with:
I do not wish to give the impression that I think there is no mystery about consciousness. There is, for instance, something of a paradox connected with any attempt to localise it. But I do not think these mysteries necessarily need to be solved before we can answer the question with which we are concerned in this paper.
In other words, his paper is not about detecting consciousness (or ‘mind, intention, whatever’). It is simply about the possibility of computers performing so like humans in a question-and-answers game that it becomes impossible to tell which is which. (Interestingly, the inverse proposition that it is impossible for a human to behave in such a way that they appear to be a computer is never seen as a sign that something is missing in humans!).
The word ‘person-like’ is never used in his paper.
Also of interest is section 8, The Argument from Informality of Behaviour, because Turing actually discusses whether computers could mimic human behaviour. He skewers that particular objection by pointing out that we don’t actually know if human behaviour is law-like or not:
“…we cannot so easily convince ourselves of the absence of complete laws of behaviour as of complete rules of conduct. The only way we know of for finding such laws is scientific observation, and we certainly know of no circumstances under which we could say, “We have searched enough. There are no such laws.”
In other words, he takes no position on duality and free will – as becomes a scientist.
You, on the other hand, very clearly are a dualist and this leads you to conclude that there is something in your game before you have even tested it properly. Instead of starting with a question and looking for the answer, you think you already have the answer and are only looking for ways to substantiate it.
No it is a practical exercise in design detection.
I wanted to see if you could recognize that the process that was represented by the graph showed evidence of being the result of intention.
If the German navy could have done that they certainly would have changed the way they were conducting themselves.
That would have been highly practical, That does not mean I’m claiming anything
peace
Exactly what Ive been saying.
I have repeatedly (contra Patrick and others) said that there is no way to definitively say even that any other minds exist at all.
Turing tests are about identifying behaviors we associate with minds not minds themselves
Just as I’m doing
I’m not even sure what it means to be a dualist the term has so much baggage.
If you are saying that I believe that other Minds exist I would agree but that belief has no bearing on the test. It would work just as well either way.
That is the beauty of it
peace
I would say that because you very clearly reject the idea of mind in things like the the cause of evolution it leads you to resist the idea of a non standard Turing test before you have even explored the idea.
What I’m trying to convey to you is that it’s OK to explore this stuff.
It does not have to be so scary for you.
You can relax
peace
Exactly so.
Correct Turing speaks of humans instead of persons.
I’m not so restrictive for me any ole person will do it does not have to be a human being.
peace
You’re going round in circles. If I had to say ‘human’ or ‘bot’, I’d say ‘bot’ by now.
I haven’t read every word of this post and its comments, but superficially it seems to be just a rehash of the older thread about FMM’s game.
That thread accrued about 900 comments, and there was never any explanation of what was being detected. By explanation, I mean an operation definition of the classes of datasets to be discriminated.
My rather biased reading of FMM is that he prefers defining the classes in ways that proscribe operational definition. It is equivalent to to the explanatory filter, which requires one to know that a process cannot be the result of evolution before running it through the filter that is supposed to discriminate evolved products from designed products.
I’m doing no such thing I’ve been consistent from the beginning of this thread.
I would say that you are reading things into what I say that are not there.
It’s that sort of thing that will make you see circles where none exist.
I’ve always said there is nothing to compel you to ever infer that any other mind exists.
peace
Could you point to the comment where I said that Turing tests were dead horses? If not, I don’t see the point of your response.
The context necessary for the standard Turing test have been clearly documented in numerous publications, going all the way back to Turing. Why is it necessary to spell it out again?
”
OK. You want advice? For a graphical Turing test to be valid (and it has to be valid for it to be usefull) you have to inform the test subjects that the graph could represent a specific intentional behaviour (eg., voting numbers by year) or a specific non-intentional set of data (eg., diurnal temperatures).
but this would still not be a Turing test. A Turing test does not detect the difference between intentional and non-intentional, or design and non-design. It detects the difference between one human design and another human design. There is nothing non-designed about the output of a computer.
Any example of any old person who is not a human?