Moderated Coin Flipper

[psionl0]

The set includes 8182 integers greater than 127.

Where did you get that number from? Did Leumas drop the file somewhere for you to examine?

I would have imagined that whenever a new trial is run, a new file of random numbers has to be loaded up or you won't get random results.

You could get more out of the file if you just examined each bit instead of whole bytes at a time.

 

[jt512]

Where did you get that number from? Did Leumas drop the file somewhere for you to examine?

Here: http://godisadeadbeatdad.com/CoinFlipper4/scripts/worker_TRNG.js

I would have imagined that whenever a new trial is run, a new file of random numbers has to be loaded up or you won't get random results.

I had imagined the same. I was surprised to see that it was hard coded in the script.

 

[jsfisher]

Why are you so very much concerned about a bloated nothing but window dressing pointless app that contributed nothing to your knowledge...

You started the thread. Did you not expect people to participate? My questions are real, by the way.

...and that you slandered me as lying about???

When you admit that you slandered me in error and you apologize for doing so I might consider your concern as of interest to me.... until then... I am not interested in your concerns.

You are still misusing that word. You mean libel, not slander.

At any rate, I did not libel you. I did accuse you of lying. That was because, contrary to your claim about modeling the "edge case", your code did not model the "edge case" in any way that could be described as legitimately actually modeling the "edge case". Your claim was false, and you would not discuss nor clarify nor amend it, so it was false and deliberate. I.e., a lie.

Still, I'll ask again: Can we engage in a simple civil conversation?

 

[Leumas]

At any rate, I did not libel you. I did accuse you of lying. That was because, contrary to your claim about modeling the "edge case", your code did not model the "edge case" in any way that could be described as legitimately actually modeling the "edge case". Your claim was false, and you would not discuss nor clarify nor amend it, so it was false and deliberate. I.e., a lie.

Still, I'll ask again: Can we engage in a simple civil conversation?

Your lack of understanding of my code and then accusing me of not accounting for the Edge case because you did not understand how the code is accounting for and edge case... is your problem... but it is NOT CIVIL to accuse people of lying because you do not understand how their code works.

A code that accounts for the edge case is a code that accounts for it.... your lack of understanding of it does not negate it is accounting for it... neither does your OPINION about it make it a lie.

The SLANDER is to say because you do not understand or not like how it accounts for the edge case then IT DOES NOT... and that I am lying to say so.

That is SLANDER... deliberate and malicious SLANDER... and IT IS NOT CIVIL.

In any case... you keep doubling and now quadrupling on your slander.

So until you retract your slander and admit your error and apologize for your now QUDRUPED SLANDER... I am not interested whatsoever in your CONCERNS.

Note: your lack of understanding of something or your dislike for it does not make it nonextant.


.

 

[Dr.Sid]

This guy

 

[Leumas]

Where did you get that number from? Did Leumas drop the file somewhere for you to examine?

His number is wrong... the file is 16,384 Bytes.

I would have imagined that whenever a new trial is run, a new file of random numbers has to be loaded up or you won't get random results.

Why... if the bytes are random... and I shuffle the array before using it... then any draw of numbers from the array of random numbers is a random number.... no?

There is no need to get a new file every time... no?

A draw from a deck of cards that is shuffled is a random draw... and now add that the deck of cards is itself random.

All you need to keep getting random draws is to shuffle the cards.

In any case the site you gave me has a FREE API for programmatically downloading the numbers... but you can do only 1,000 downloads a day... for more you have to pay monthly payments.

If I am going to use the API to download a file for every flip and my app does 10,000 flips then I will be meeting the limit in ONE RUN of one usage of the API.

Not to mention the time it takes to communicate and download the 16,384 bytes every time I want to do a flip for 10,000 times.

You could get more out of the file if you just examined each bit instead of whole bytes at a time.

Why? I think the arrangement of bits in a byte that makes a random Unsigned Integer which is randomly positioned in an array from which a random chunk of numbers is drawn... is as random as anything to be expected from an actual real world source of randomness... no?

 

[jt512]

His number is wrong... the file is 16,384 Bytes.

That's what I said:

The version of your app that uses the TRNG appears to be biased in favor of Tails. If I understand your code, your script includes a hard-coded set `randomData` of 16,384 integers between 0 and 255 that you had downloaded from http://www.random.org/bytes. The script shuffles this set, picks a random starting point, and draws a sequence of N integers n from the set, where N is the number of flips set by the user (up to 10,000). For each flip that does not result in an edge, the script classifies the flip as a head if n > 127, and a tail otherwise.

The set includes 8182 integers greater than 127. Thus, conditional on the flip not being an edge, the probability of a head is 8182/16,384=0.4994, and a tail 1–0.4994=0.5006. Thus, your app is biased in favor of tails.

It seems strange that you are somehow sure that I misinterpreted your code, when you failed to comprehend the post in which I explained my understanding of it.

 

[jsfisher]

Your lack of understanding of my code and then accusing me of not accounting for the Edge case because you did not understand how the code is accounting for and edge case... is your problem...

Oh, but I do understand your code. It is not in any way complicated. Moreover, you have made a bizarre claim about the code regarding the probability of zero occurring as the sum of four 32-bit integers.

The available evidence suggests (strongly) that it is you who does not understand the code.

...but it is NOT CIVIL to accuse people of lying because you do not understand how their code works.

That might have been correct if it were not for the fact I do understand your code.

A code that accounts for the edge case is a code that accounts for it.... your lack of understanding of it does not negate it is accounting for it... neither does your OPINION about it make it a lie.

Then how about we discuss the code? Let's lay off the snipes in either direction. That would be much more productive than you belittling my knowledge of computer programming and me accusing you of lying.

The SLANDER is to say because you do not understand or not like how it accounts for the edge case then IT DOES NOT... and that I am lying to say so.

That is SLANDER... deliberate and vicious SLANDER... and IT IS NOT CIVIL.

It is not SLANDER. You are using the word incorrectly. Why won't you correct such a simple misstatement???

If you believe my posts to be uncivil, please report them as such, but I fear your understanding for Rule 0 is at odds with many others here, including the moderator staff.

 

[Leumas]

That's what I said:
It seems strange that you are somehow sure that I misinterpreted your code, when you failed to comprehend the post in which I explained my understanding of it.

Why are you so incessantly concerned about this... as you said... worthless useless app that added nothing to your knowledge?

And are you ever going to apologize for SLANDERING me and admit your error in calling me a liar?

Until then I am not at all interested in reading anything you have to say... especially when it is clearly nothing but red herrings.

 

[jsfisher]

...
The script shuffles this set, picks a random starting point
...

I found this part intriguing. Given that the dataset has been pseudo-randomly reordered, what is the point of then picking a pseudo-random starting point (on the interval [0,200)) for the sequence?

I could ask Leumas, but I know what to expect.

 

[Leumas]

Where did you get that number from? Did Leumas drop the file somewhere for you to examine?...

Can you now see the extent of CONCERN RED HERRINGS?

• First concern about not being a real coin inside the computer
• Then concern about RNG's quality
• Then concern about Edge landings... and slandering me because it is not what they want
• Then concern about BIAS towards Tails
• Then concern about the bytes in the TRNG's data
• Then back to more concern about the PRNG

 

[jt512]

Leumas,

Can you now see the extent of CONCERN RED HERRINGS?

• First concern about not being a real coin inside the computer
• Then concern about RNG's quality
• Then concern about Edge landings... and slandering me because it is not what they want
• Then concern about BIAS towards Tails
• Then concern about the bytes in the TRNG's data
• Then back to more concern about the PRNG

You're not concerned about whether an app you programmed to simulate coin flipping is biased or not?! You consider such a concern to be a red herring? If your app doesn't produce unbiased results, it isn't performing the very function you intended it to do? That's what you consider a red herring?

 

[jt512]

Why are you so incessantly concerned about this...

More importantly, how can you not be concerned about whether your app is biased or not?

 

[theprestige]

Can you now see the extent of CONCERN RED HERRINGS?

• First concern about not being a real coin inside the computer
• Then concern about RNG's quality
• Then concern about Edge landings... and slandering me because it is not what they want
• Then concern about BIAS towards Tails
• Then concern about the bytes in the TRNG's data
• Then back to more concern about the PRNG

The interest in the quality of your RNG is directly related to your attempt to relate the output of your program to the nature of randomness in the real world.

The interest in the nature of your data is directly related to the desire to better understand your program and its output. So is the interest in possible bias.

I agree that the business about the edge case is a red herring, and I'm surprised that you've put so much effort into implementing it.

You seem upset about the interest in your code and its output. What did you desire, when you started this thread, if not interest in your code and its output?

ETA: And of course there is a legitimate concern about whether code that doesn't simulate a real coin can tell us anything useful or interesting about real coins.

 

[Leumas]

You're not concerned about whether an app you programmed to simulate coin flipping is biased or not?! You consider such a concern to be a red herring? If your app doesn't produce unbiased results, it isn't performing the very function you intended it to do? That's what you consider a red herring?

Your mistaken and erroneous assertions about my app... that you consider worthless and pointless that you can do better by using an entry into a spreadsheet app... means nothing.

And your continued attempts to be CONCERNED about it are nothing but red herrings.... since you already decreed it worthless to all (including me) and nothing but window dressing that contributes nothing to your knowledge... or anybody else's including mine.

... The fact is your app is trivial. Not only does it do nothing more than illustrate a statisitical fact that has been well understood for hundreds of years, it merely implements a rudimentary random number function, which in R (the programming language I know best) requires exactly one line of code.

Your app contributes exactly nothing to our knowledge of anything. And everybody here—including you—knows it.

 

[Leumas]

The interest in the quality of your RNG is directly related to your attempt to relate the output of your program to the nature of randomness in the real world.

The interest in the nature of your data is directly related to the desire to better understand your program and its output. So is the interest in possible bias.

Aha... thanks for admitting that your concern is to defend no randomness.

I agree that the business about the edge case is a red herring, and I'm surprised that you've put so much effort into implementing it.

Thanks for admitting it... and yes you are right I am completely naïve for having fallen for the dissimulation of being genuinely CONCERNED.

 

[Leumas]

More importantly, how can you not be concerned about whether your app is biased or not?

It is not.

Your CONCERN to incessantly assert that it is... is just another red herring like that you did for the edge case. [imgw=60]http://www.internationalskeptics.com/forums/imagehosting/5128262c750038bf97.jpg[/imgw]

 

[theprestige]

Wait wait wait. If Leumas is using a set of pre-generated values to choose from, how do we know the values are evenly distributed on either side of the "> 277" boundary?

 

[Leumas]

Wait wait wait. If Leumas is using a set of pre-generated values to choose from, how do we know the values are evenly distributed on either side of the "> 277" boundary?

Hahahaha.... amazing!!!

Do you understand what randomness means?

Or is this as you already admitted just another red herring? [imgw=60]http://www.internationalskeptics.com/forums/imagehosting/5128262c750038bf97.jpg[/imgw]

I agree that the business about the edge case bias in data is a red herring, and I'm surprised that you've put so much effort into implementing it.

 

[theprestige]

Thanks for admitting it... and yes you are right I am completely naïve for having fallen for the dissimulation of being genuinely CONCERNED.

Admit? I told you in plain language on the first page of this thread. That's why I'm so surprised you went ahead with it anyway.

And I wouldn't say you're completely naive. I just think you've got some pretty large blind spots.

A good programmer would welcome third party examination of their code as an opportunity to find bugs and make improvements. But you rush off to chase red herrings (both programmatic and social) rather than focus on the things that really matter: The quality of your data set and the risk of bias in your logic.