Is randomness = indeterminsim?

[Peter Soderqvist]

No. For example, the decimal digits of 1/7 are an unbiased, uniform collection of the digits 1,2,4,5,7, and 8 (all digits appear equally frequently), but they're hardly random. On the other hand, a "fair" roulette wheel has a green number come up less than 10% of the time, so it's strongly biased against green, but is still random.

Soderqvist1: you math is not a physical process so it doesn't apply!
I know close to nothing about roulette wheels! But I think that gamblers around the table thinks that all gamblers have equal chance to win, the game is fair when the outcome is determined by pure chance! It appears to me in my "fuzzy concept of roulette wheels", that it is simply a lottery! Nor do I understand the concept of green in connection with roulette wheels, but since other ways than green is favored because they are strongly biased against probability green, the outcome is not random, since we know that green has not a fair chance in the game, he is more likely to lose than to win! But all other ways than green are random, because they have equal chance to win as far as I can see from your message!

Randomness means that the physical process in principle is not biased in any particular way; all possible ways are equally probable to happen. Mutation is random because there is no correlation between a particular genotype and mutation; in short mutation have no direction! But natural selection is nonrandom, because there are correlations between selection and mutations! Natural selection favor mutations which are compatible with a particular niche, and disfavoring outfit mutations say; green colored bear furs since they are more detectable and so an easier target than white furs in the artic!

 

[BillHoyt]

Randomness means that the physical process in principle is not biased in any particular way; all possible ways are equally probable to happen.

No, it does not. Take a die, 1/6 probability for each face. Random. Now give it the faces 1,1,2,3,4,5. It is still random, except the probabilities for landing on any particular face are different. It is biased toward '1'.

Mutation is random because there is no correlation between a particular genotype and mutation; in short mutation have no direction!

There are most certainly correlations between genotypes and mutation. In addition, the forward mutation rate differs from the reverse mutation rate. The bias in mutations is further complicated by codon code degeneracy.

 

[Peter Soderqvist]

TO BILL HOYT

No, it does not. Take a die, 1/6 probability for each face. Random. Now give it the faces 1,1,2,3,4,5. It is still random, except the probabilities for landing on any particular face are different. It is biased toward '1'.

Soderqvist1: I think we both can agree that the dice's trajectory and its rotation is random, but we disagree about values! Thus, in your language, a random process, with random values, its output is random! A random process, with random values, its output is random but biased! It is inconsistent to have two propositions with the same premises, but with different conclusions! What rationale do you use in order to determine if something is random or not?

In my language, a random process, with random values, its output is random! In my language, a random process, with nonrandom values, its output is nonrandomly biased! Two propositions with different premises with different conclusions are consistent! Non-tautologically, a random process, with nonrandom values, its output is nonrandom! The rationale I am using to determine if something is random or nonrandom is to find out if there is a pattern there or not! Thus a fair dice have values 1-2-3-4-5-6 I cannot see any pattern in it, thus it is random! An unfair dice 1-1-2-3-4-5 the value 1 repeat itself is its pattern, thus that part is nonrandom!

There are most certainly correlations between genotypes and mutation. In addition, the forward mutation rate differs from the reverse mutation rate. The bias in mutations is further complicated by codon code degeneracy.

Soderqvist1: of course there is a correlation between genotypes and mutation in the sense that the genes in the genotype mutates, but there is no correlation between a particular gene and "beneficial mutation" Some is bad, some is neutral, some is beneficial, and there is no pattern there in order to justify teleology that evolution is guided or directed, thus mutations' direction is therefore simply random when correlated with the teleological ladder! But natural selection is nonrandom in the sense that all mutation has not the same survival value! Mutated genes with adaptive fitness in a niche tend to be selected to replicate meanwhile neutral or deleterious is disfavored or ignored by natural selection!

I will be back at Monday!

 

[BillHoyt]

I think we both can agree that the dice's trajectory and its rotation is random, but we disagree about values! Thus, in your language, a random process, with random values, its output is random! A random process, with random values, its output is random but biased! It is inconsistent to have two propositions with the same premises, but with different conclusions! What rationale do you use in order to determine if something is random or not?

In my language, a random process, with random values, its output is random! In my language, a random process, with nonrandom values, its output is nonrandomly biased! Two propositions with different premises with different conclusions are consistent! Non-tautologically, a random process, winonrandom values, its output is nonrandom! The rationale I am using to determine if something is random or nonrandom is to find out if there is a pattern there or not! Thus a fair dice have values 1-2-3-4-5-6 I cannot see any pattern in it, thus it is random! An unfair dice 1-1-2-3-4-5 the value 1 repeat itself is its pattern, thus that part is nonrandom!

I can't comment on your language except to say that if it disagrees with the universal languge of mathematics, it is wrong. An unfair coin is still random. Detection of patterns within randomness does not render it non-random. Random simply means you cannot predict with certainty what is going to happen.

Soderqvist1: of course there is a correlation between genotypes and mutation in the sense that the genes in the genotype mutates, but there is no correlation between a particular gene and "beneficial mutation" Some is bad, some is neutral, some is beneficial, and there is no pattern there in order to justify teleology that evolution is guided or directed, thus mutations' direction is therefore simply random when correlated with the teleological ladder! But natural selection is nonrandom in the sense that all mutation has not the same survival value! Mutated genes with adaptive fitness in a niche tend to be selected to replicate meanwhile neutral or deleterious is disfavored or ignored by natural selection!

You've switched propositions here. You first wrote "Mutation is random because there is no correlation between a particular genotype and mutation; in short mutation have no direction!" This is incorrect. Depending on the starting DNA configuration, some mutations are more probable than others. This has nothing to do with either selection or teleology and has everything to do with biochemistry and the genetic code.

 

[Peter Soderqvist]

TO BILLY HOYT

Detection of patterns within randomness does not render it non-random.

Soderqvist1: Why so when disorder in the main doesn't rule out local order in a thermodynamical system?

I can't comment on your language except to say that if it disagrees with the universal language of mathematics, it is wrong. An unfair coin is still random. Detection of patterns within randomness does not render it non-random. Random simply means you cannot predict with certainty what is going to happen.

Soderqvist1: suppose Alice and Bob takes part in some hypothetical game of dice!
They use fair dices controlled by an independent party, they have each 100 casts and the person with the highest number when all trial is added together is the winner! It is not possible to predict the winner in this game, since it is a game in pure chance! It is therefore random according to your definition!
Now suppose Alice still have her fair dice, but Bob has now the unfair one since it has no number 6 but two 1. I can predict that Alice tend to win because of her number 6 advantage, and because of Bob's two 1 disadvantage, therefore this game is nonrandom! Frankly to hammer home the point; If Alice's Dice have only sixes, and Bob's dice only ones, I can predict with 100 % certainty that Alice will win every time, because the game is not random at all, Alice is a predetermined winner!

You've switched propositions here. You first wrote "Mutation is random because there is no correlation between a particular genotype and mutation; in short mutation have no direction!" This is incorrect. Depending on the starting DNA configuration, some mutations are more probable than others. This has nothing to do with either selection or teleology and has everything to do with biochemistry and the genetic code.

Soderqvist1: I was too short!
I mean what I have said in my post above, but I will also add that harmful or neutral mutations have incomparable many more ways than beneficial ones!

 

[BillHoyt]

Why so when disorder in the main doesn't rule out local order in a thermodynamical system?

Now you're on a different topic. How about we handle the basic randomness issues before tackling systems theory?

suppose Alice and Bob takes part in some hypothetical game of dice!
They use fair dices controlled by an independent party, they have each 100 casts and the person with the highest number when all trial is added together is the winner! It is not possible to predict the winner in this game, since it is a game in pure chance! It is therefore random according to your definition!
Now suppose Alice still have her fair dice, but Bob has now the unfair one since it has no number 6 but two 1. I can predict that Alice tend to win because of her number 6 advantage, and because of Bob's two 1 disadvantage, therefore this game is nonrandom! Frankly to hammer home the point; If Alice's Dice have only sixes, and Bob's dice only ones, I can predict with 100 % certainty that Alice will win every time, because the game is not random at all, Alice is a predetermined winner!

So by your above assertions, the dice used at the craps table cannot be random simply because the house over time is the winner? You're confounding many concepts here. The probability density functions of the fair die and the unfair die are different; that doesn't mean one is random and the other not. When you moved your example to a die of all sixes and one of all ones, you collapsed probability space altogether on both dies. There is no longer randomness.

You're broader confusion revolves around the concept of uniform distribution, which you may want to acquaint yourself with.

I was too short!
I mean what I have said in my post above, but I will also add that harmful or neutral mutations have incomparable many more ways than beneficial ones!

The "fitness" of any particular mutation is a separate matter again to the matter of your original statement. You wrote: "Mutation is random because there is no correlation between a particular genotype and mutation; in short mutation have no direction!" In fact, there is direction in mutation. There is "forward mutation" and "reverse mutation" The rates are different. There is bias to mutation events. They are not uniformly distributed. Yet they are still "random."

You should not confound "direction" in the sense of "forth and back" with "direction" in the sense of "teleology." Just as you should no confound "biased" or "non-uniformly distributed" with "non-random."

 

[Peter Soderqvist]

The regular and the random!

TO BILL HOYT

So by your above assertions, the dice used at the craps table cannot be random simply because the house over time is the winner? You're confounding many concepts here. The probability density functions of the fair die and the unfair die are different; that doesn't mean one is random and the other not. When you moved your example to a die of all sixes and one of all ones, you collapsed probability space altogether on both dies. There is no longer randomness.

Soderqvist1: It seems to me that your language is too binary either, or!
When I say that an unfair dice is not random the unfair dice still have random properties, just as the dice is not a surface but still have surface properties, the dice is a cube, which belong to the class of solids! You also know that when every postulates are equal in two geometries except the fifth postulate, one of these geometries is Euclidian, and the other is non-Euclidian! With the same way to use language, a fair dice is random, and an unfair dice with only sixes is regular, the slightly unfair dice mentioned earlier with two 1, is quasi-random, or quasi-regular but doesn’t fulfill the name random, nor dos it fulfill the word regular! An unfair dice with three 1 is less random, but not yet regular!

 

[BillHoyt]

Re: The regular and the random!

It seems to me that your language is too binary either, or!
When I say that an unfair dice is not random the unfair dice still have random properties, just as the dice is not a surface but still have surface properties, the dice is a cube, which belong to the class of solids! You also know that when every postulates are equal in two geometries except the fifth postulate, one of these geometries is Euclidian, and the other is non-Euclidian! With the same way to use language, a fair dice is random, and an unfair dice with only sixes is regular, the slightly unfair dice mentioned earlier with two 1, is quasi-random, or quasi-regular but doesn’t fulfill the name random, nor dos it fulfill the word regular! An unfair dice with three 1 is less random, but not yet regular!

Peter,

No, my "language" was a response to your "language." You wrote, "I can predict with 100 % certainty that Alice will win every time, because the game is not random at all, Alice is a predetermined winner!" You managed to write that ignoring your previous assertion: "I can predict that Alice tend to win because of her number 6 advantage..." Those statements are contradictory. "Tend to" is a statement of randomness, which your fair die/biased die game clearly still has. Over short runs, Alice can certainly still lose. Even over longer runs, Alice can still lose. The game is still random. The bias doesn't alter that fact a tad.

If we have a die with five 1's and one 6, we still have a random die. It has a 5/6 chance of coming up 1 and a 1/6 chance coming ujp 6. You cannot be serious to state anything else, I'm afraid. What would you say of the classic "ball-and-urn" examples from elementary probability courses? That they are wrong to be called random because the urn contains 12 black balls and only 2 red?

We can't use private definitions when discussing these concepts, Peter. Random means you cannot say with certainty what the outcome of the event will be. That's it, in a marble-covering nutshell.

 

[Peter Soderqvist]

TO BILL HOYT

Peter, No, my "language" was a response to your "language." You wrote, "I can predict with 100 % certainty that Alice will win every time, because the game is not random at all, Alice is a predetermined winner!" You managed to write that ignoring your previous assertion: "I can predict that Alice tend to win because of her number 6 advantage..." Those statements are contradictory. "Tend to" is a statement of randomness, which your fair die/biased die game clearly still has. Over short runs, Alice can certainly still lose. Even over longer runs, Alice can still lose. The game is still random. The bias doesn't alter that fact a tad.

Soderqvist1: my statement is consistent because tend to indicates that the game is not determined purely by chance, since a truly random system have no tendencies at al, or in other words; biased system has a tendency to deviate from a pure chance outcome!

If we have a die with five 1's and one 6, we still have a random die. It has a 5/6 chance of coming up 1 and a 1/6 chance coming up 6. You cannot be serious to state anything else, I'm afraid. What would you say of the classic "ball-and-urn" examples from elementary probability courses? That they are wrong to be called random because the urn contains 12 black balls and only 2 red?

11-12-2004 11:56 AM: An unfair coin is still random. Detection of patterns within randomness does not render it non-random

Wikipedia, the free encyclopedia. Dice
Dice are thrown to provide (supposedly uniformly distributed) random numbers for gambling and other games (and thus are a type of hardware random number generator); however, because the edges of typical dice are often rounded, they do not provide "true" random numbers. Casino dice come the closest to true uniformly distributed random numbers.
http://en.wikipedia.org/wiki/Dice

Soderqvist1: Thus an unfair coin is not a true random number generator because of its bias! I have never seen that urn but I assume that the balls are selected at random because the "human selector" is blindfolded regarding the urn, but the balls color is not uniformly distributed the outcome is therefore; not truly random!

We can't use private definitions when discussing these concepts, Peter. Random means you cannot say with certainty what the outcome of the event will be. That's it, in a marble-covering nutshell.

Soderqvist1: All natural laws a statistical in their nature!
I can predict with statistical certainty that the fittest animals tend to survive, thus I cannot predict that with 100% certainty because of random accidents, yet natural selection is nonrandom according to Dawkins. Take a look below what Richard Dawkins has replied to the Alabama's State board of Education's Question!

 

[Peter Soderqvist]

A MESSAGE FROM THE ALABAMA STATE BOARD OF EDUCATION
[to be pasted in all biology textbooks] "EVOLUTION ALSO REFERS TO THE UNPROVEN BELIEF THAT RANDOM, UNDIRECTED FORCES PRODUCED A WORLD OF LIVING THINGS. "

Dawkins: Where did this ridiculous idea come from that evolution has something to do with randomness? The theory of evolution by natural selection has a random element -- mutation - but by far the most important part of the theory of evolution is non-random: natural selection. Mutation is random. Mutation is the process whereby parent genes are changed, at random. Random in the sense of not directed toward improvement. Improvement comes about through natural selection, through the survival of that minority of genes which are good at helping bodies survive and reproduce. It is the non-random natural selection we are talking about when we talk about the directing force which propels evolution in the direction of increasing complexity, increasing elegance and increasing apparent design.

The statement that "evolution refers to the unproven belief that random undirected forces. . ." is not only unproven itself, it is stupid. No rational person could believe that random forces could produce a world of living things.
http://www.simonyi.ox.ac.uk/dawkins...Work/Articles/alabama/1996-04-01alabama.shtml

 

[BillHoyt]

Soderqvist1: my statement is consistent because tend to indicates that the game is not determined purely by chance, since a truly random system have no tendencies at al, or in other words; biased system has a tendency to deviate from a pure chance outcome!

Peter,

Please take some time to search the internet for "non-uniformly distributed." If you take the time to read the URLs you get from such a search you will find you are confusing "uniformly distributed" with "random." Several people now have attempted to get this point across to you. One example was the urn. Another was the roulette wheel.

"Random" does not mean "unbiased." Neither does it mean having "no tendency." In fact, Peter, a key theorem of statistics is the Central Limit Theorem. That theorem states that it is a tendency of all random distributions, if you gather enough together, to approach the binomial distribution. Look up, please, the binomial distribution to see how it is utterly different from the uniform distributions you seem to be hung up on.

Wikipedia, the free encyclopedia. Dice
Dice are thrown to provide (supposedly uniformly distributed) random numbers for gambling and other games (and thus are a type of hardware random number generator); however, because the edges of typical dice are often rounded, they do not provide "true" random numbers. Casino dice come the closest to true uniformly distributed random numbers.
http://en.wikipedia.org/wiki/Dice

Please read this again, Peter, and observe the scare quotes surrounding "true." They are there because what is actually meant is "true uniformly distributed random numbers." Read it in context, please.

Soderqvist1: Thus an unfair coin is not a true random number generator because of its bias! I have never seen that urn but I assume that the balls are selected at random because the "human selector" is blindfolded regarding the urn, but the balls color is not uniformly distributed the outcome is therefore; not truly random!

Read it, Peter.

Soderqvist1: All natural laws a statistical in their nature!
I can predict with statistical certainty that the fittest animals tend to survive, thus I cannot predict that with 100% certainty because of random accidents, yet natural selection is nonrandom according to Dawkins. Take a look below what Richard Dawkins has replied to the Alabama's State board of Education's Question!

Peter, if you have a language barrier or problem, then it is incumbent upon you to improve your foreign language skills if you wish to engage in a dialogue in the foreign language. I do not disagree with Dawkins's statement at all. Apparently you think I do. Please take out whatever dictionaries you need, and read my statements again about mutation direction, which is real, sir, and teleology which is not.

 

[drkitten]

Peter,


Please read this again, Peter, and observe the scare quotes surrounding "true." They are there because what is actually meant is "true uniformly distributed random numbers." Read it in context, please.

Wikipedia article has been amended to remove the lying scare quotes. It now reads "fair (uniform) random numbers," which I think will be less confusing.

 

[Peter Soderqvist]

I am short of time here, but how do you reconcile that the fittest animals only tend to survive, it is not 100 % certain that they do, tend to is not compatible with nonrandom as you have said earlier, "Tend to" is a statement of randomness" but natural selection is nonrandom according to Dawkins?

 

[drkitten]

I am short of time here, but how do you reconcile that the fittest animals only tend to survive, it is not 100 % certain that they do, tend to is not compatible with nonrandom as you have said earlier, "Tend to" is a statement of randomness" but natural selection is nonrandom according to Dawkins?

I don't. Take it up with Dawkins and see if he finds your uninformed ignorance any more amusing. Alternatively, you could look up the Central Limit Theorem and the concept of expected value, and reconcile it yourself.

 

[T'ai Chi]

In fact, Peter, a key theorem of statistics is the Central Limit Theorem. That theorem states that it is a tendency of all random distributions, if you gather enough together, to approach the binomial distribution.

The CLT refers to a normal distribution, not a binomial distribution, to be precise.

 

[Jorghnassen]

The CLT refers to a normal distribution, not a binomial distribution, to be precise. [/B]

I have to agree, Central Limit Theorems usually imply convergence to a normal distribution. In some rare cases, Cauchy, it depends on the set up. My probability theory is rusty.

 

[qXdc]

uhhh

uhhh http://true.wxcs.com/ig

 

[Peter Soderqvist]

I don't. Take it up with Dawkins and see if he finds your uninformed ignorance any more amusing. Alternatively, you could look up the Central Limit Theorem and the concept of expected value, and reconcile it yourself.

Soderqvist1: I am glad to see that you too are ignorant uninformed about how to reconcile "Tend to" is a statement of randomness" with nonrandom natural selection the fittest tend to survive! But on the other hand your ad hominem is much simpler task to do than to answer that question in layman's terms! If that is what you implicitly mean with scholarship, then I think I prefer my uninformed ignorance!

 

[Peter Soderqvist]

TO BILL HOYT

Soderqvist1: I shall investigate these 4 terms!
"non-uniformly distributed." Central Limit Theorem. binomial distribution. uniform distributions

Peter, Please take some time to search the internet for "non-uniformly distributed." If you take the time to read the URLs you get from such a search you will find you are confusing "uniformly distributed" with "random." Several people now have attempted to get this point across to you. One example was the urn. Another was the roulette wheel.

"Random" does not mean "unbiased." Neither does it mean having "no tendency." In fact, Peter, a key theorem of statistics is the Central Limit Theorem. That theorem states that it is a tendency of all random distributions, if you gather enough together, to approach the binomial distribution. Look up, please, the binomial distribution to see how it is utterly different from the uniform distributions you seem to be hung up on.

Soderqvist1: My proposition; a fair die has no bias or no tendency whatever regarding its numbers, statistically in the long run every number is equal frequent, the die have no significant tendencies to show any particular number more than any other number, that die is truly random! That sequence of digits is not possible to describe in any shorter terms than the sequence it self, therefore that sequence is random! In what sense invalidate the Central Limit Theorem that?

The digits of 1/7 is endless but not random because it is compressible to repeat 1428571 endlessly after 0,
Suppose we have a urn with 50 white balls and, 50 black balls and someone select these blindfolded and puts these in a row, and repeats this row after row for a long while, this sequence of balls is equally random as the earlier die is, because it is not compressible because of its lack of pattern! Now compare that with 99 black ball and 1 white balls row after row and so on, that long sequence is not random because every individual black sequence is compressible (to 1 x the particular number in the sequence), in stead of black, black, black, black, black, black, black, ,,,,,

Please read this again, Peter, and observe the scare quotes surrounding "true." They are there because what is actually meant is "true uniformly distributed random numbers." Read it in context, please.

because the edges of typical dice are often rounded, they do not provide fair (uniform) random numbers. Casino dice come the closest to true uniformly distributed random numbers.
http://en.wikipedia.org/wiki/Dice

Soderqvist1: there is no quotes arund true there, but the word not is there!
Please read the word not in this context!

Soderqvist1: Casino Dice is closest to the true uniformly distributed random numbers. Because a Casino Die is only an approxomation to the idealy fair die!

Peter, if you have a language barrier or problem, then it is incumbent upon you to improve your foreign language skills if you wish to engage in a dialogue in the foreign language. I do not disagree with Dawkins's statement at all. Apparently you think I do. Please take out whatever dictionaries you need, and read my statements again about mutation direction, which is real, sir, and teleology which is not.

Soderqvist1: we both agree that mutation is random, I have not alleged otherwise!
But you have said that "tend to is a statement of randomness" which is not consistently compatible with nonrandom natural selection's tendency to favor survival of the fittest!

 

[Peter Soderqvist]

Random number From Wikipedia
A number itself cannot be random except in the sense of how it was generated. Informally, to generate a random number means that before it was generated, all elements of some set were equally probable as outcomes. In particular, this means that knowledge of earlier numbers generated by this process, or some other process, do not yield any extra information about the next number. http://en.wikipedia.org/wiki/Random_number

Soderqvist1: In the unfair die set, all numbers is not equally probable as outcomes, therefore; every number which this die output is nonrandom, when some number is statistically determined as more frequent than other numbers, then this frequency give us information that the die is biased in that way! Conversely a fair die doesn't give us any information about its further state!