Said poop, never.
Printable View
Said poop, never.
That was a good point. Almost.
You can't randomize your actions internally. People are notoriously shitty at generating randomness. Like if you ask someone to pick two random numbers between 1 and 10 they'll pick two that seem an equal distance like 3 and 7. That's the opposite of random. They're also horrible at recognizing randomness, because they expect it to be equally dispersed in the space, which it isn't. Our problem with doing these things is we're wired to see patterns, not the lack of patterns.
You can, however, randomize your actions using external cues, like the second hand on your watch, a dice, a coin, or whatever.
If I type "You have a right to defend yourself against tyranny" and the responses are post after post accusing me of saying the word "defeat" or "safe," then quite frankly, I don't think it's my fault that I'm being misunderstood.
If I say it is an inalienable human right to the means of lethal violence and the responses involve, "but guns kill." Well OK.
Lots of things kill. Yes, we're talking about guns, but if your position is to put a double standard on the lethality of one tool over another lethal tool, then I think it's fair to talk about that as pertains to the strength of your conviction.
Me questioning you about the depth of that double standard seems perfectly on topic of me determining how strongly you, personally, adhere to this moral you've asserted. Barely at all, it seems... OK. So that signifies to me that you're not making arguments based on any deep conviction, but on something else. If you can't or wont address that something else, that's your choice. However, the argument certainly appears to be a double standard the way you've presented it.
If I say that innocent until proven guilty is a solid moral foundation for jurisprudence, and you don't counter that, but persist in treating people like criminals when they have not done anything to warrant such, then again... I point to that dichotomy and ask what's really going on?
I look around the world and the universe and I see violence everywhere. From bottom to top. Any assertion that we're in this place of violence, but we have no right to our own means of violence is just such brazen and unvarnished wrongness that I can't understand the assertion as honest and in good faith.
When we can look around the world together and agree that humans, despite our strengths and greatness, are still brutal, savage, unpredictable creatures in certain places, at certain times... then how can you possibly assert that we've advanced as a species to the point where we can in any form of good faith say that we're any different that the rest of this violent place?
The tone in the rest of your post isn't really worth addressing.
Meteorites? What? Cages?
The relevant argument would be "I've never been hit by a meteorite, so no one should have the right to protect themselves from meteorites."
Whether or not you and I personally think a meteorite cage is silly, that shouldn't matter to anyone who doesn't.
The response I've given is you have no practical defense against tyranny today. Sure you have a right to defend yourself against it. My point is a gun does not allow you to do that in any meaningful way. And you seem to acknowledge that.Quote:
Originally Posted by MadMojoMonkey
My response has been to question your use of the term "inalienable human right" as if owning a gun is on the same moral level as not being forced into slavery. I just don't agree there's anything near an equivalence there.Quote:
Originally Posted by MadMojoMonkey
There's no double standard because I'm not advocating for the promotion of cancer or heroin or anything else that kills people. If I had said 'ban guns, but cancer fuck yeah!' you might have a point. All I said was 'ban guns'. I'm not addressing these other evils you keep bringing up here because that - for the last time - is not what the argument is about.Quote:
Originally Posted by MadMojoMonkey
That signifies to me that you're not actually addressing the points and still trying to make the argument about something else, which now seems to revolve around questioning my motives, which are, in fact irrelevant to the question of gun control. And in any case, I have no ulterior motive here in discussing gun control; it's strictly a question of what is best for society.Quote:
Originally Posted by MadMojoMonkey
No, it's a way you use to avoid the topic. I'm not advocating we be soft on cancer.Quote:
Originally Posted by MadMojoMonkey
I've already answered this. I don't consider taking a gun away from someone treating them like a criminal, because that presumes they had a right to own a gun in the first place. The way I see it, it's no more a mistreatment of law abiding indviduals to make gun possession illegal than it is making heroin possession illegal. Rather, I consider it doing them a favour.Quote:
Originally Posted by MadMojoMonkey
It's kind of hard to get my head around what your point is here. Something about 'we need guns to deal with a violent, dangerous world.' Presumably (and I don't want to put words in your mouth), this is to protect us from said world, or make us safer? But I think Oskar has already shown you that isn't what happens.Quote:
Originally Posted by MadMojoMonkey
fypQuote:
Originally Posted by MadMojoMonkey
This was my point.Quote:
Originally Posted by Poopadoop
Well, this wasn't the point I was making, but I agree. If you ask me for a random number between 1 and 10, then me simply choosing a number isn't random. I need to bring in an outside influence. One way I could get a random number is to count how long it takes until a red car goes past, and then use the last number.Quote:
You can, however, randomize your actions using external cues, like the second hand on your watch, a dice, a coin, or whatever.
I know. I just explained why, as opposed to just posting my gut feeling.Quote:
Originally Posted by OngBonga
Nope, you weigh your distribution towards smaller numbers that way. Not random. Semi-random at best.Quote:
Originally Posted by OngBonga
Not really. Is there a higher probability that there will be a red car after 14 seconds compared to 9 and 19?Quote:
Originally Posted by Poopadoop
It depends how busy the road is, I guess. If it's a three-lane motorway, then I agree we'll skew our answers towards lower numbers. But if it's a quiet ish road, say an average of one car every ten seconds, then this problem is eliminated and it is random.
You didn't explain why. You explained your gut feeling.Quote:
Originally Posted by Poopadoop
You need a uniform distribution - i.e., equal probability of every outcome.Quote:
Originally Posted by OngBonga
Picking the first event starting at 0 ensures that won't happen.
E.g., Assume you're trying to randomize between 1 and 10. You do your Ong red car thing. A car comes at 1 with some probability (p). Every time a car comes at 1 you stop and figure that's your random number. But what that means is you will have more 1s than 2-10s because another car could have come at 2, 3, 4, etc., but you're only counting the 1. And if no car comes at 1, but a car comes at 2 you stop, so you also have more 2s than 3-10s. And so on.
You have randomized the event but you haven't randomized how it gets assigned a value. You need to have an equal chance of starting at any number by, e.g., looking at the second hand on your watch as soon as a red car goes by and choosing the last digit - e.g., 48 seconds after 11.30 means your random number is 8.
But if you're going to do that you may as well dispense with the car and just look at your watch.
No, there's actual research that's been done on how bad people are at randomizing, and how bad they are at recognizing randomness.Quote:
Originally Posted by OngBonga
Look it up.
You can reduce the problem to what is probably negligible quantities but you'll never eliminate it. And you'll need a very low amount of cars to do that, like one every hour, at which point it'll probably be dark before you see a red one.Quote:
Originally Posted by OngBonga
Maybe you can start a semi-randomness thread if you want to pursue this probabilistic folly.
I don't need to. I agree. People aren't just "bad" at randomising, they are incapable of it. We just discussed how outside influences are necessary. You can't just "think" of a random number.Quote:
Originally Posted by Poopadoop
What a waste of research. Who asks for a random number from a person and expects an actual random number? People who doesn't understand randomness.
Nailing down the "obvious" with data is one of the purposes of research.
Well yeah, but they're studying randomness by observing things that are clearly not random.Quote:
Originally Posted by MadMojoMonkey
And yes, human behaviour is clearly not random. A lifeform that behaves randomly will not be favoured by evolution.
Genetic evolution appears to happen randomly in random offspring. For all apparent purposes, even if we can't act randomly (upon which I wholeheartedly agree), the way we became what we are is through a vehicle primarily steered by randomness.
It's an interesting dynamic, at least.
Ever play a sport? Sometimes you move left, other times you fake left and then go right? That's randomizing.Quote:
Originally Posted by OngBonga
We can do it in our behaviour, just semi-randomly rather than being fully randomized.
Being able to be somewhat unpredictable is pretty much fundamental to our survival. This requires a degree of randomization.
If every time something that wanted to eat us saw us do X, they knew that Y and Z were going to come next, we'd never have got out of the savannah.
In fact there is an experiment that shows that people try to minimize their randomization when they're cooperating with a partner, and maximize it when they're competing against an opponent, a la Game Theory.
I've seen a guy bet air like AA. Is that random enough?Quote:
Originally Posted by MadMojoMonkey
I should add "in a random hand".Quote:
Originally Posted by Jack Sawyer
MMM, I promise I'm not trying to strawman you, and I don't try to evade questions but I need to condense it down to get to the core of what you're saying because you say a lot of things.
You seem to think there should be almost no restrictions on the sort of firearms you can legally own.
- Are you ok with the restrictions that are currently in place? Do you want less restrictions? Why, or why not.
You say it is an inalienable right. I don't think you have addressed why you think it is one. You say it's so you can defend yourself against tyranny, but not successfully? So you acknowledge that your chances of fighting a potential tyrannical US government with or without guns are simply different shades of zero, correct?
Here's my problem: you understand that children are dying as a direct result of lax gun laws, but you say it's worth it in the end because [...] <- I really don't know what that part is in your mind.
You're playing a trolley problem game on a national scale where on one track you have thousands of children and on the other track you have [...], and you jump on the lever to save [...], but what is it?
You also seem to take any suggestion of more restrictive gun laws as entering a slippery slope. I think of the current gun laws in the US as simply a point on a sliding scale. They are very restrictive if you look at the totality of available weapons. We see that countries that are a little more restrictive, but not totally so have managed to decrease their gun related deaths by a lot. You are saying that this increment on the sliding scale is worth a couple thousand children because of the potential benefits.
It 100% is. Very common now, and it even was a thing way back when.Quote:
Originally Posted by MadMojoMonkey
This isn't random behaviour.Quote:
Ever play a sport? Sometimes you move left, other times you fake left and then go right? That's randomizing.
There is no "semi-random". Something is either random or it isn't. It's binary as fuck.Quote:
We can do it in our behaviour, just semi-randomly rather than being fully randomized.
You're assuming "random" and "unpredictable" are synonyms. In a sense they can be, but only if by "unpredictable" you mean "impossible to predict", and not "very difficult to predict". There's the binary nature of randomness. Something that is difficult to predict is not random.Quote:
In fact there is an experiment that shows that people try to minimize their randomization when they're cooperating with a partner, and maximize it when they're competing against an opponent, a la Game Theory.
Try again Euclid.Quote:
Originally Posted by OngBonga
https://web.mit.edu/16.unified/www/F...es/node56.html
The set [1,1,1,1,1,1,1,1] has zero entropy
The set [1,2,4,6,12,22,34] has low entropy
The set [372, 18879, 53, 22903, 4190013] has high entropy
...because they are.Quote:
Originally Posted by OngBonga
There are varying degrees of unpredictability that depend on the degree of randomness in a monolithic fashionQuote:
Originally Posted by OngBonga
fypQuote:
Originally Posted by OngBonga
Fun puzzle: If you tied a rope tight around the Equator, then added a single metre of slack, would anybody notice it was longer?
I would. I added the slack, after all.
That response had moderate entropy.
How about if someone else added the slack - would you notice?
It'd be like 6" of the ground, anyway, right?
2 pi r -> 2 pi r + 1 m = 2 pi ( r + 1 m / (2 pi) )
which corresponds to a circle with the radius of Earth (in meters) + 0.159 m (about 6.25 inches)
You're going to invoke entropy as evidence of semi-randomness? Nice try, but nope. Entropy is a measure of thermal energy differential, the available thermal work in a system.Quote:
Originally Posted by Poopadoop
Randomness in thermodynamics emerges when we don't know all the variables. Physics treats events as random, with a distinct probability of happening, because it's convenient, for the same reason it's convenient to consider poker as random and play the probabilities, even though it is not random, it is instead based on the dealer's shuffle and the decisions of the players... non-random events. It's random in the sense it is impossible to predict... unless we know all the initial conditions and observe all relevant influences. In which case it's not random.
Randomness is not a matter of frame of reference. If I know the initial conditions and you don't, if I know precisely how the deck was shuffled and you don't, it's not a random event for you but not for me. It's a non-random event for us both. It just seems random to you because you lack information.
If those numbers represent the temperature of atoms in a closed system, then yes, you're right. What does this have to do with randomness? Are you suggesting that [1, 1, 1] is less random than [1, 2, 3]?Quote:
The set [1,1,1,1,1,1,1,1] has zero entropy
Sort of. But it's loose. Are you happy with "lacking information" as a definition of "random"? I'm not. A dice isn't random, it just seems so because we don't have all the information needed to predict the outcome with certainty. All we can do is predict the probability of the outcome, giving rise to the illusion of randomness.Quote:
...because they are.
tl;dr... randomness is an illusion.
Everything happens for a reason, nothing happens for no reason.
There are more forms of entropy than simply thermodynamic entropy
It comes up in information theory. FYI, Shannon Entropy
Alright, fair enough, but that doesn't detract from the point about randomness.
It emerges because of a lack of information. Nothing is truly random.
Isn't radioactive decay considered to be random?
Let's get real. There's a process that governs radioactive decay, we just aren't aware of it yet.
The only thing we have going for us really, is that we can mathematically define what a "random variable" is and work with that. Then we can say if we observe something that behaves like the model, that it, too is a random variable. Then we note that we have error bars on everything and codify all of that... and we can then have a test with a confidence interval that says whether or not something behaves in a random way.
Ong's point that the die roll isn't actually random is probably correct. If we actually knew the initial conditions and material properties of everything in play to arbitrary precision, then we could predict the final state of the die roll ahead of time.
The only thing that can intervene with this is QM, which is truly, statistically random, as defined above. I.e. we've defined what random looks like, and QM looks just right after bazillions of tests. Very large N.
Quantum effects don't manifest on the scale of a macroscopic die roll. So no randomness from physics applies.
QM is only as random as the dice.
The data are absolutely against this conclusion.Quote:
Originally Posted by OngBonga
The reality is that any new physics must fit pre-existing data. The pre-existing data look exactly like perfect mathematical randomness. Whatever new physics illuminates the underlying process, it cannot change that the process manifests as randomness.
I mean I'm drifting into faith here. I don't think QM is a random realm, I just think there are a mind boggling quantity of variables, the vast majority of which we cannot even begin to measure, that it's the closest thing we have to randomness. Statistical analysis is the only way we can hope to understand it, for the same reason statistical analysis is a winning poker strategy. But if we *could* know all the variables, and *could* process them in a timeframe that doesn't involve us dying before we're done, then we can predict QM events.
But to say QM is random is also an act of faith. The only evidence is based on a lack of information.
In QM, the lack of information is quantifiable. We can rigorously account for the randomness associated with our lack of information.Quote:
Originally Posted by OngBonga
When we do so, we find that our predictions are accurate to dozens of decimal places. That is true.
However... what we find is that the lack of information is baked into the physics. It's not that "we" don't know. It's that there is no mechanism in the universe by which that could be "known."
Electrons are indistinguishable. There is no physical meaning of saying "this electron" only "an electron" since you can't tell whether or not "this" electron was swapped for another one when you weren't looking. You can't tell if it decomposed then recomposed in the blink of an eye. You only know it's physical properties, which are exactly identical to any other electron.
So the physics has to reflect that "the universe doesn't know the difference between electrons," not just us. It's baked in as a principle... and it works to astounding precision.
And again... any new physics must be in line with pre-existing data, and the data shows that it isn't possible to know certain things in particle physics. That's not even invoking fourier Transform pairs or the position-momentum uncertainty.
There are many forms of quantifiable randomness in QM
Are you familiar with the "one electron" theory of the universe? Something along the lines of... every electron in the universe is the same electron, travelling through time. As crazy as it sounds, I'm under the impression it makes mathematical sense.
Anything with a googolplex of variables will look like "perfect mathematical randomness" to our puny minds.Quote:
The pre-existing data look exactly like perfect mathematical randomness.
I'm not talking about whether randomness exists in a metaphysical sense like 'God rolling dice' or whatever. I'm talking about randomness in terms of statistical entropy. This is not the same as thermodynamic entropy, it just borrows the term as a familiar word that means 'lack of order'.Quote:
Originally Posted by OngBonga
Clearly such a quantity can be applied to things other than thermodynamics, and is a continuous variable (i.e. not 'on/off' as you suggest).Quote:
We thus look for a single quantity, which is a function of the p_i , that gives an appropriate measure of the randomness of a system.
If you flip a coin the result is slightly random inasmuch as you have p(.5) of guessing the outcome in advance - statistically this can be viewed as a system with low entropy.
If you roll a 6-sided die the result is more random as you have a p(.1667) of guessing the outcome - higher entropy.
And so on.
So yes, certain events can be "more random" than others statistically speaking.
In terms of behaviour, if you ask people to press a button with as little randomness as possible, they are pretty good. The variance of their timings is fairly low.
If you ask them to press a button in a random pattern, they increase the variance of their timings.
So yes, on some level, people can generate randomness, just nowhere near complete randomness.
Because it effectively is perfect randomness (assuming each variable is itself random). Or so close to it that it hardly matters.Quote:
Originally Posted by OngBonga
The state of each individual variable might be a binary 0 or 1 (or it could have several levels, or it could be a continuous value between -infinity and + infinity). Adding those up is complex enough, but the number and magnitude of higher-order interactions (how variable A's outcome affects variable B, or how the combination of A and B outcomes affects C, or the combination of ABC on D, and so on up the chain, is like a googleplex raised to the googleplex worth of complexity. Imagine poking your nose into those data! I'm actually getting an erection just thinking about it.
Quote:
Originally Posted by MadMojoMonkey
So, don't electrons have identities then? Can you not theoretically freeze a moment in time and say ' here is electron A at x,y,z coordinates such and such, and here is a different electron B at these different coordinates?'
This is where we disagree, on a philosophical level. It does matter. To accept randomness is to accept our intellectual limitations. To reject randomness is to strive for a better understanding, or at least to acknowledge that a better understanding exists.Quote:
Or so close to it that it hardly matters.
I can't believe randomness truly exists. That's the same as saying something happens for no reason, there is no process governing that event. If the universe is governed by randomness, how can it be stable?
These events aren't "more random", they are "more unlikely". Something would be "more random" in the context you're using that term if it is harder to predict. Both the coin and the die are easy to predict, in a statistical sense, assuming they are fair. An unbalanced dice is "more random" than a balanced one, despite both having the same number of potential outcomes.Quote:
So yes, certain events can be "more random" than others statistically speaking.
But I prefer "harder to predict", because randomness is an illusion.
In fact the balanced die is probably "more random" than the unbalanced one. Depends how you want to look at it. What is a successful prediction? One that accurately estimates the probability? Or one that is correct more often? If it's the former, then a balanced dice is the less random, but if it's the latter, an unbalanced dice will favour one outcome the most, and once we figure out what that outcome is, we can correctly predict the outcome more often, so that would end up being less random.
But it's not random, so it's not the right word.
I can believe it exists, at least on a quork level or whatever. True randomness is self-correcting, meaning it can easily be part of a larger, stable system because of regression to the mean. If you flip enough coins, you'll approach something extremely close to a 50% heads/tails split with an extremely high probability.Quote:
Originally Posted by OngBonga
The sum of a number of highly random events can have very low randomness.
Maybe. Hope not though. I like games.Quote:
Originally Posted by OngBonga
The dieroll in this case has six possible outcomes, the coinflip only two. Thus one has more randomness than the other. I don't think that's controversial.Quote:
Originally Posted by OngBonga
Au contraire, the unbalanced coin is less random. That's what makes it unfair mon frere.Quote:
Originally Posted by OngBonga
I've heard of it, but I don't think anyone really thinks that's what's happening.Quote:
Originally Posted by OngBonga
An electron's identity is the collection of physical properties which define it. All electrons are identical. In physics we say electrons are indistinguishable. (All fundamental particles are indistinguishable, in fact.)Quote:
Originally Posted by Poopadoop
If we freeze time, we can say, those are 2 electrons are "far" apart, so we're justified to consider them "isolated" wave functions (this is an approximation, if not obvious). For that frozen moment in time, we can tell which is which. However, since we know their positions to arbitrary precision, we know nothing about their momenta. So if we allow time to advance any iota of a second, and we look again, we still see a universe with exactly 2 electrons, but we have no idea which one we originally labeled as A and which was originally labeled as B. We only know there are 2.
EDIT: it's not about what "we" know. That's misleading. It's about what "can be" known. Since the freezing of time rigorously defines positions in your opening statement, it means all momentum information is destroyed.
That's what Einstein thought, too. So... I guess you're in good company by holding this skepticism.Quote:
Originally Posted by OngBonga
However, the data is clear. There is randomness within rigorously defined bounds. The wave model of particle physics is ridiculously powerful. The wave is simply a postulate, a guess, but works real good. Then we apply probability theory to the wave, literally calculate the EV of its position, momentum, energy, spin, etc. with the exact same math we use to calculate EV in poker (well, poker's discrete and some of those are continuous properties, but otherwise the same).
QM is built on a lot of lucky guesses. One of them is that electrons in atoms can be described as waves of discreet (quantized) allowed energies. Simply do that, calculate the EV of position for the different energies, and you can solve for the Hydrogen atom and all those pretty electron orbital flowers.
You literally just guess that A) wave equation will work if you B) quantize the allowed energy states, then C) apply probability theory treating this wave as a probability distribution of a random variable. Then you can calculate the EV of whatever property you like. Note: C was a bit tricky until we figured out all the operators to use, but that was all done decades ago.
...
Or whatever underlies the randomness, manifests as perfect mathematical randomness, so ... uhh... that's it's role? It can't affect anything if it's whole purpose results in what looks perfectly random, unless that underlying mechanism is some kind of white noise filter preventing or obscuring any underlying motives.
Lol.
Like in some past world, Gods were able to affect the universe, but then this shield got there somehow and it wipes away all external influence.
I'd play that video game, if nothing else.
I agree with this.Quote:
Originally Posted by Poopadoop
I don't know how you can agree with it, it's self contradictory. On the one hand, he says that something is "more random" if it simply has more outcomes, while on the other he's saying something is less random if it's unbalanced, despite having the same number of outcomes. Surely if randomness is the number of potential outcomes, an unbalanced coin and a balanced coin are equally random.
This exposes the problem with such a loose definition of "random". When it's ill defined, it can mean different things,
Randomness can't be both a measure of potential outcomes, and a measure of the probability of those outcomes. It's one or the other.
Quote:
Originally Posted by OngBonga
It's perfectly consistent, you're just thinking about it wrongly.
More outcomes = more disorder = more randomness.
From the pov of probability theory:
A fair coin has exactly two outcomes.
The biased coin has two possible outcomes, BUT
A biased coin is closer to having a single outcome than a fair coin. To take an extreme example, if there were a coin that was so biased that if you flipped it 100 million times, it only came up heads once, you would accept I think that such a coin is mathematically closer to a nonrandom outcome (1 outcome) than it is to a random outcome (2 outcomes).
Similarly, you could make a die that was so biased as to be less random than a (fair) coin. If p(direoll 6) = .9999999, this die is mathematically closer to having one outcome than having six outcomes and also less random than the fair coin.
Another way to think of it: if you wanted to generate randomness, would you prefer to use a fair die or the one where you're almost certain to roll a six?
More probability theory: pure randomness is unbiased, which is what makes it self-correcting. This eventually leads to fun things like the bell curve.
That vid is a fantastic example of something we call random that isn't random.
If you recreated all initial conditions precisely, you would get the same outcome. Maybe there's a clearer definition of random. An event that doesn't necessarily have the same outcome when all initial conditions are identical and the experiment is repeated.
And I still insist a fair dice and a fair coin are equally as random... ie, 100% random. Both are equally predictable over a large enough sample.
You're talking about hard vs. soft determinism. I'm talking about statistical randomness.Quote:
Originally Posted by OngBonga
Nope.Quote:
Originally Posted by OngBonga
Far a coin with 2 outcomes, we can show that the "most random" case is when the probability of flipping heads is the same as the probability of flipping tails, and that any imbalance makes the coin flip less random, to the limit where one side flips 100% of the time and the other flips 0% and we've lost all randomness.
Shannon Entropy for this system is
- ( p_0 * ln_2( p_0 ) + p_1 * ln_2( p_1 ) )
where p_0 is the probability of flipping heads, p_1 is the probability of flipping tails.
ln_2(x) is the logarithm function in base 2.
Obviously, p_0 + p_1 = 100%
so p_1 = 1 - p_0.
LET
p_0 = p and p_1 = 1 - p
Now we have a function of a single variable.
Rewrite the Shannon entropy equation with this substitution
-p * ln_2(p) - (1 - p) * ln_2(1 - p)
A graph of this function is concave down, with maximum at p = 0.5. The Entropy approaches 0 as p goes to either 0 or 1.
The "problem" is that we're using a rigorously defined mathematical definition of random, and but you're not.Quote:
Originally Posted by OngBonga
I'm not sure I follow your 2nd point.
We're not separating those ideas, as both are relevant.
A "fair" 10 sided die is more random than a "fair" 6 sided die. There are more outcomes, equally distrubuted, so there's more randomness.
An RNG that produces a random number from 1 - 10 has more possible outcomes than an RNG that produces a random number form 1 - 6. E.g. it's more random.
I had to stipulate both facts to draw that conclusion. 1) that both dice are fair and 2) that one die has more possible outcomes.
It takes both pieces of information to apply entropy (as applies to information theory) in order to produce a value that is a measure of the randomness, the Shannon Entropy.
This is just a mathematical tool to help us fill the gaps in when we don't know all the variables.Quote:
statistical randomness.
Yep. If randomness is a variable concept, then it has to be a deviation from expected probability (well,the inverse). A fair dice has zero deviation (maximum randomness) from expected probability, over an infinite sample. Same with the coin. Therefore, equal (maximum) randomness.Quote:
Nope.
If you're going to say that the dice is more random than the coin because six not two, while also insisting that a loaded dice is less random than the fair dice, then you're using two different definitions of random, one to describe the number of possible outcomes, and the other as a measure of the difference in probability between different outcomes.
I'm defining random rigorously. An event that happened for no reason.Quote:
The "problem" is that we're using a rigorously defined mathematical definition of random, and but you're not.
It's worth noting that QM is about both. Hard determinism coupled with statistical randomness.Quote:
Originally Posted by Poopadoop
Hard determinism in that if we know all the knowable information about a particle's state, we can exactly calculate the probabilities that we will later observe it in another state.
It can't be soft determinism - because the predictions made when we assume that what "can be known" is a True fact about the system and not some reflection of our ignorance about the system's "hidden" properties - those prediction produce extremely precise predictions. If there was soft determinism, then we'd see outcomes that DO NOT exist in our prediction or that would take place at a probabilistic rate that is NOT what we've calculated as our EV.
It can't be non-random, because the outcomes vary.
That is not mathematically rigorous.Quote:
Originally Posted by OngBonga
"For no reason" seems like a very hard thing to prove, under the best circumstances.
One of my favorite demos involves a piece of tech that I must describe first.Quote:
Originally Posted by OngBonga
Each student is required to have a little remote control device called an iClicker. That remote control has 6 buttons on it. Power and A,B,C,D,E. It's simply a device so we can ask a multiple choice question to 100+ students in the room and track their participation and attendance for a tiny fraction of their final grade.
OK
In this demo, I put out a bin of pennies and tell each student to take one as they enter the class.
The demo goes like this:
Each student flips the coin 4 times and enters the appropriate A,B,C,D,E corresponding to 0, 1, 2, 3, or 4 heads flipped.
The cool part is that the graph of student responses comes in real time and is shown on a projector. It always produces that binomial distribution. We could never tabulate all those initial conditions to reproduce the experiment exactly, but we don't need to and that's the point. We could never predict which students would flip exactly 1 head or exactly 2 heads, but we can definitely predict the % of students who will flip that number of heads with remarkable precision (taking error bars into account).
*nearly always. Sample size too small.Quote:
It always produces that binomial distribution
What you describe there is something that is seemingly random. Once again, if you know all the relevant variables, however many trillions that might be, and were capable of processing them,then you could predict the outcome without the need to rely on probability. Statistical analysis is a mathematical tool, very closely reflecting the behaviour of the universe. Because it's so close, we're led to believe randomness is a real life thing. It's not. It's an emergent property when we lack information.
Whatever you say.Quote:
Originally Posted by OngBonga
You're still talking about determinism vs. statistical randomness. If you want to argue those are different things, then sure they are. If you want to argue determinism means there is no such thing as statistical randomness then no, it doesn't.
Fuck me, learn how to post images.
You can flip that Galton board 'till the end of time and it's always going to produce that binomial distribution, within the error bars of that specific geometry and number of balls.
Same for the coin flipping. It's not like my one class is the only ones to perform this experiment. It's been performed thousands if not millions of times for over a hundred years across the world. That's not a small sample size, though my classes' specific contributions to the overall population is small.
eh, yeah I forgot what soft determinism was. It's actually about free will. I thought the argument was that since QM can include randomness this somehow gives us free will.Quote:
Originally Posted by MadMojoMonkey
IAC, it's irrelevant since Ong is just repeating over and over stuff about how everything is predetermined, which is orthogonal to my explanations of statistical randomness.Quote:
Soft Determinism is the theory that human behaviour and actions are wholly determined by causal events, but human free will does exist when defined as the capacity to act according to one's nature (which is shaped by external factors such as heredity, society and upbringing)
The balls are bouncing off each other, so their behaviour is governed by the dynamics in question.Quote:
You can flip that Galton board 'till the end of time and it's always going to produce that binomial distribution, within the error bars of that specific geometry and number of balls.
Sample size too small.Quote:
thousands if not millions of times
What's the probability that every student flips heads with every flip? Is it precisely zero?
Statistically, a "large sample" is n > 30. So assuming he has at least 30 students in the class each year, it's practically guaranteed to work.Quote:
Originally Posted by OngBonga
Indeed.Quote:
practically
.0625^n, where n is the number of students.Quote:
Originally Posted by OngBonga
For 30 students, it's ~ 1 x 10^-37
(that's 0.00000000000000000000000000000000001%)
I will if you make that image your avatar.Quote:
Originally Posted by OngBonga
Like I was saying, sample size too small. Thanks for doing the maths for me, I couldn't be bothered.Quote:
(that's 0.00000000000000000000000000000000001%)
Is "I'm a twat" not cutting it?Quote:
I will if you make that image your avatar.
The whole point is that random variables can be the result of deterministic physics. Whether or not the minutia of motions is random or not is another question entirely. I agree that the system is rigorously deterministic on scales above the QM realm.Quote:
Originally Posted by OngBonga
That doesn't erase the binomial distribution of outcomes that results from the statistical analysis of all possible results and their expected frequencies.
No, of course not. It's just very small. Statistics isn't about exact numbers, it's about probabilities and relationships. It's about ranges of outcomes with rigorously defined error bars on those possible outcomes.Quote:
Originally Posted by OngBonga
If that could never happen, that would be proof that the distribution is not a binomial distribution, after all.
In fact, this isn't about any 1 class, really... it's about the population of all classes and all sets of 4 coin flips.
***
It's fair to argue that randomness isn't mathematically provable as such. We can only say that something behaves like a perfect mathematically random variable or it does not. We need infinite information to make the error bars go away entirely, and that's not possible.
The best we can do is compare how a perfectly random variable would behave to how our system under investigation behaves and note that there is no statistically provable difference between them.
BWAAAHAHAHAHA!Quote:
Originally Posted by OngBonga
Nice
Not quite infinite information, and in some cases we might one day be able to measure and process all relevant information sufficiently to make accurate predictions... that's where I see QM going.Quote:
We need infinite information to make the error bars go away entirely, and that's not possible.
Error bars are the standard error, which is stdev/sqrt(n). So as long as n < infinite and stdev > 0 then se will always be se > 0.
So yes, infinite.
You're completely missing a very important fact about QM.Quote:
Originally Posted by OngBonga
Presuming mathematically perfect random variables are "good" models of many phenomena just works. This is a core assumption of the Standard Model of Particle Physics, which provides more precision than any other predictive model, ever. Does that mean it's "True?" No. Not remotely.
What it does mean, though, is that whatever is ultimately "True" has to manifest as this perfect randomness within rigid boundaries.
Nothing new can make the data go away. The data says there are many properties of particles which are accurately modeled as mathematically perfect random variables. In many, varied experiments, this has been shown. For decades. Despite many intelligent people like yourself feeling like it can't be the Truth.
The doubting physicists have been won over by the data. I am one of them. Though my gut still wants to agree with you, my mind simply must accept what it knows to have personally witnessed.
We can get into the many interpretations of QM and maybe find arguments like the "pilot wave theory" which compel a less probabilistic view of the underlying mechanism. This is a bastion of many physicists whom still trust their gut on the matter, even though the data shows something irrefutable. The thing is that the "pilot wave" is the imaginary* portion of the solution to the position wave function. We can't measure imaginary values, only real ones. The position is a real value... it's pilot wave is imaginary. It's an untestable hypothesis. It can neither be proven, nor disproved.
EDIT:
* imaginary as in i^2 = -1
This is certainly not the case. Many neuroscientists and physicists have put a hard no on this one.Quote:
Originally Posted by Poopadoop
The problem is exactly in the random variables, and the precision of predictions the data shows.
If "you" can affect the outcomes in any way, that's not a random event anymore. Not in the way the Standard Model says it's random. There would be measurable deviations from the statistical prediction. We'd be able to look at the particle physics in brains and see definite missed predictions from our model. The model stipulates perfectly random variables. Any deviation from that perfect randomness, by imposition of free will, e.g., would be measurable. It's not.... to VERY large N.
My bad... I guessed at the distinction from context and guessed wrong.Quote:
Originally Posted by Poopadoop
Which is not to say that consciousness is not a very useful way to describe human behavior.
The models we use can vary on different scales.
The properties of the constituents can be vastly different... unrelated, even... to the properties of a complex body.
As a poor analogy.
Both Sodium and Chlorine are poisonous chemicals and you shouldn't eat them.
Sodium Chloride is table salt and is an essential food that you need to survive.
All things are poison, and nothing is without poison, the dosage alone makes it so a thing is not a poison.
The solution to pollution is dilution, yes.
I heard a lot of soldiers in WW1 died from eating chlorine.
Another one of those "but guns" cases
https://www.youtube.com/watch?v=J0o_eVeMVr0
Killer Kampground Kathy lol.
Y'all seem to not see the difference between the immorality of an outright ban vs. the immorality of unregulated access.
It can be perfectly legal for Kathy to own a gun and carry it on her person while simultaneously being perfectly illegal for her to draw the gun in any manner that escalates a situation from "not potentially fatal for anyone involved" to "potentially fatal for someone involved."
Looks like Kathy needs her permit to open carry suspended while she attends a training course in general gun safety and she completes the exam associated with said course.
And my argument is as follows:Quote:
Originally Posted by MadMojoMonkey
Whatever reason she may have top have had a gun, she, at that point, draws her gun on people for a in every case frivolous at best reason. She has the power to easily end their lives over a permit dispute. She should not have that power in my opinion. If she chooses to do so, it should be very hard or require particular skills, so not anyone who can pull a trigger can do it.
People like her spoil it for the rest of the people who would actually carry a gun in a responsible manner. She has a gun, we have no idea whether it's licensed or not, she is not using it responsibly in any way, shape or form. She drew it upon people over a fucking camping ground permit dispute. Are you fucking kidding me
I've seen people draw a gun and shoot another because they were punched. The general populace should not have the power of an executioner over whatever generally trivial matter they may deem of utmost importance. Your personal judgement should never lead to someone else's life being lost over nonsense.
Oh, and about legal guns (and carry permits, this is the reality right now)
https://www.youtube.com/watch?v=TCSpokckHC0
DO YOU WANT COMMUNISM? What part of 'inalienable right' don't you understand? What about innocent until proven guilty?! How will she unsuccessfully defend herself against tyranny?Quote:
Originally Posted by MadMojoMonkey
Hey you still haven't told me if you're in favor of less strict gun laws and why/why not.
"in my opinion"Quote:
She should not have that power in my opinion.
How refreshing.
I agree this women is wholly irresponsible. If these people were carrying a weapon also, it very quickly turns into a very serious situation. You've got an argument for self defence when someone pulls a gun out on you for something so trivial. You have to question their sanity.
She needs more than training. She needs to lose her right to own a gun.
What you're missing here is that if it wasn't so easy for her to own a gun, her inclination to escalate from non-lethal to lethal wouldn't even be relevant. Such escalation would not even be an option for someone of her ilk.Quote:
Originally Posted by MadMojoMonkey
He suggested private citizens should be allowed to own nukes. So he has answered the first half of that question at least.Quote:
Originally Posted by oskar
WHY DO YOU HATE FREEDOM?Quote:
Originally Posted by Poopadoop
https://i.imgflip.com/22cq2y.jpg
I'm pretty sure that was nanners.Quote:
Originally Posted by Poopadoop