Book 3 - Page 263

Say "perhaps" to Flower Power

Book 3 - Page 263
Comic - Book 3 - Page 263
Recent posts... (See full thread)
Chiu ChunLing wrote:
Morgaln wrote:
That's only true for events where every possible outcome has the same chance to occur as any other. If they probabilities differ, some coincidences are less likely than others.
What you're describing here is that some descriptions of an event cover a wider range of possible events, like it is slightly more likely that a lottery number will have a 9 than that it will have a 9 AND a 7, and less likely than that it will have a 9 OR a 7.

When you get to a certain level of specificity that permits a wide range of numbers, and then narrow that down to a more specified description that eliminates most of that range, then the more specific description can be said to be less unlikely to match the actual event. But every actual event is itself perfectly specific, and thus as unlikely as any other possible actual event, even though it may be commonly described in terms that would match a much larger set of possible events.

For instance, in rolling a pair of dice, it is more likely that your total will be 7. But there are six different rolls which produce a seven, compared to only one roll that would produce 12 and one roll that would produce 2 (and two each -> 3 and 11, 3e. -> 4 & 10, etc.). We could see the same principle in play with coin-flipping or whatever. When we have a lot of orthogonal elements added together to make a total, the average/mean/median total will tend to be the most likely because there are more different ways of arriving at that total. But we impose that by the way we distribute and measure (or implicitly weight) the number scale we use to describe each element...even with dice rolls and coin flips, but only counting the final 'total' of which facet ends up facing upwards rather than considering every detail of the event.

The point being, in a lottery you don't win because the total of the digits in your number added together is the same as the total of the digits in the winning number, so you're not more likely to win by picking a number that contains an even distribution of digits (or just 4 and 5). But you can win at craps (or betting on most sports) that way...it's kinda arbitrary.

So 'winning' or 'losing' can definitely be more or less likely, depending on your game. But all the actual events are equally unlikely even if you define more of them as 'winning' events than 'losing' events.

Of course, Luckamancy (and likely Carnymancy) seems to do exactly this in Erfworld. But Predictamancy doesn't.


But we're not talking rolling dice or flipping coins here. We're talking about much more complex events. When I go down the street, it's more likely that a car runs me over than that a rhino appears and gores me; they don't have the same chance of happening because there isn't a simple die roll that decides what will kill me on the street today. There are underlying causalities that make one more likely than the other. Likewise, the chance that a particular beam will fall at a specific point during a fire at a specific time is not the same for every beam. It is affected by where the fire started, how hot it burns, how it spreads, and many other variables that can't be modeled with easy numbers. So it looks like a coincidence when that exact beam falls right before Parson can cast the spell.
Of course in this case, it is a believable coincidence, because that beam could have fallen at just that point. Since we don't know all the variables, it's a realistic coincidence, and might even have been a likely occurrence. But that doesn't mean any beam falling at that exact point in time had the same probability and neither does it mean that a beam falling has the same probability at any point in time.
Chiu ChunLing wrote:
But every actual event is itself perfectly specific, and thus as unlikely as any other possible actual event

This is complete and utter nonsense.
Adept wrote:
Crisco wrote:
That doesn't confirm where he's from. It could, perhaps, serve as evidence that he isn't a native of Erf, but even that is still speculation. He has extra- dimensional channels and he's constantly glued to the dish in most circumstances, so it makes just as much sense for him to have learned vernacular through that as it is for him to be from somewhere else.

Nope. Charlie is also the Wizard of Oz, who is, like Judy, from Earth.

It's feels 99% sure at this point.

Not true. Very near the entirety of book zero alludes (without directly confirming) to him being an Erf native. Additionally, Judy being summoned is portrayed as the first use of a SPW scroll (and by extension, likely the first extra universe character to come to Erf at all, but not necessarily). Evidence currently available points more towards Charlie being native than alien, imo.
greycat wrote:
Lilwik wrote:
Even so, it's still awfully subtle in the sense that it's something that could have happened without any help from Fate.

Fate doesn't violate physics. It just makes highly improbable coincidences happen.


Yet ...

Sorry to anthropomorphize it but Fate has proven to be a cheater altering the numbers to fit his narrative if Charlie were to get enough wins in their competition Fate might just snap and go "You know what the chances are for a wild dragon to pop in your bedroom at night Chuckles ? 100%"
Fate = Rob. No need to be sorry for anthropomorphizing the author.
Morgaln wrote:
But we're not talking rolling dice or flipping coins here. We're talking about much more complex events.
Yes, many orders of magnitude more complex...but you could in principle emulate that much more complexity with enough dice/coin flips.
Morgaln wrote:
When I go down the street, it's more likely that a car runs me over than that a rhino appears and gores me; they don't have the same chance of happening because there isn't a simple die roll that decides what will kill me on the street today.
That's because there are more cars with a lower chain of events that would put them in the road to run you over.
Morgaln wrote:
There are underlying causalities that make one more likely than the other.
There's nothing "underlying" about the facts that make being hit by a car more likely than being hit by a rhino...if you replaced all the cars in the world with rhinos suddenly, then there would be large numbers of people being hit in the street by rhinos for reasons every bit as obvious as the reasons they are currently hit by cars.
Morgaln wrote:
Likewise, the chance that a particular beam will fall at a specific point during a fire at a specific time is not the same for every beam. It is affected by where the fire started, how hot it burns, how it spreads, and many other variables that can't be modeled with easy numbers.
If you flip a heads first in the run, you exclude every formerly possible run that started with tails.

Binary numbers can simulate most events to the limits of our ability to understand them at all. You might not want to call them "easy" numbers, I'm not aware of any 'easier' system that is remotely useful.

The thing is, I'm not saying that some descriptions don't cover a larger range of possible events than other descriptions...what I'm really pointing out is that our mental processes are oriented towards creating descriptions that cover the largest perceptually useful ranges of possible events. We don't speak of each given car (or each given collection of molecules that could form a set with the same total mass as a car), we speak of all cars in the world. And that creates some illusions, like the idea that it is more likely I'll be hit in the street by a particular car rusting in a scrapyard in India than by a rhino that might escape from a local zoo.

Once you get particular enough, all events are widely unlikely. If we take an event chain that has already mostly occurred, we can say that the final events of that chain are more likely...but I trust you see the result of that? It means taking any part of an event that has already occurred for granted.

Which eliminates the idea that any event that has actually happened might be considered unlikely in the least. Though the bracer does work that way to some extent.
Chiu ChunLing wrote:
Morgaln wrote:
But we're not talking rolling dice or flipping coins here. We're talking about much more complex events.
Yes, many orders of magnitude more complex...but you could in principle emulate that much more complexity with enough dice/coin flips.


You can certainly simulate every situation with enough coin flips, but not with equal chance for each outcome. Some will require more specific outcomes than others, because at some point you will need to differentiate between meaningful and meaningless. Simple example:
Just imagine a bowl with three balls, two black and one white. The chance to draw a black ball is different from the chance of drawing a white ball, and you will never be able to make those chances equal without changing the setup. Of course the chance to draw each of the three balls is equal (although not a simple coin flip, because three options), but the difference between the two black balls is meaningless.
If you want something more complex, take a poker deck. Draw two cards. It doesn't matter which cards you draw, the chance to draw that specific combination is of course equal to every other combination (1/52 * 1/51, to be exact). But if we want two events to happen together, the chances suddenly change. Let's say we want two diamond cards; the chance would be 1/4*12/51. That's distinctly different from, say, drawing two queens (1/13*1/17).
Likewise, even if we assume that the chance for every beam to break at a specific point in time is equal, the chance that Parson chooses that exact beam to sit under is lower than that he chooses another beam to sit under (assuming there are more than two beams). Because, again, it's not a binary choice. And even if each choice has the same value, we have two different events (each with several choices) that both have to come up the same result. That's actually pretty similar to drawing two queens from a deck, just more complex.

Chiu ChunLing wrote:
Morgaln wrote:
When I go down the street, it's more likely that a car runs me over than that a rhino appears and gores me; they don't have the same chance of happening because there isn't a simple die roll that decides what will kill me on the street today.
That's because there are more cars with a lower chain of events that would put them in the road to run you over.
Morgaln wrote:
There are underlying causalities that make one more likely than the other.
There's nothing "underlying" about the facts that make being hit by a car more likely than being hit by a rhino...if you replaced all the cars in the world with rhinos suddenly, then there would be large numbers of people being hit in the street by rhinos for reasons every bit as obvious as the reasons they are currently hit by cars.


Both the amount of cars and the amount of rhinos in my area, among other things, are underlying facts that make it more likely to being hit by one and not the other. In my area, the more likely one would be cars, it might be different for other people. If you trade cars for rhinos, you don't change that, you just change the numbers for both events. You don't even switch them, because rhinos behave differently from cars and thus the numbers would be completely different (complex issue, again, not just a coin flip), but that doesn't change that the probabilites for either occurrence won't be the same.


Chiu ChunLing wrote:
Morgaln wrote:
Likewise, the chance that a particular beam will fall at a specific point during a fire at a specific time is not the same for every beam. It is affected by where the fire started, how hot it burns, how it spreads, and many other variables that can't be modeled with easy numbers.
If you flip a heads first in the run, you exclude every formerly possible run that started with tails.

Binary numbers can simulate most events to the limits of our ability to understand them at all. You might not want to call them "easy" numbers, I'm not aware of any 'easier' system that is remotely useful.

The thing is, I'm not saying that some descriptions don't cover a larger range of possible events than other descriptions...what I'm really pointing out is that our mental processes are oriented towards creating descriptions that cover the largest perceptually useful ranges of possible events. We don't speak of each given car (or each given collection of molecules that could form a set with the same total mass as a car), we speak of all cars in the world. And that creates some illusions, like the idea that it is more likely I'll be hit in the street by a particular car rusting in a scrapyard in India than by a rhino that might escape from a local zoo.

Once you get particular enough, all events are widely unlikely. If we take an event chain that has already mostly occurred, we can say that the final events of that chain are more likely...but I trust you see the result of that? It means taking any part of an event that has already occurred for granted.

Which eliminates the idea that any event that has actually happened might be considered unlikely in the least. Though the bracer does work that way to some extent.


You specifically said that every event has the same probability. Here are some of the exact quotes:

Chiu ChunLing wrote:
But every actual event is itself perfectly specific, and thus as unlikely as any other possible actual event


Chiu ChunLing wrote:
Every actual coincidence that occurs is just as unlikely as any other.


That's what I'm refuting here. Of course we are assuming probabilities based on what has already happened. That's what we call a premise. That doesn't make the "final events" of the chain that happened so far more likely, because the various options for "final events" and their respective likelyhoods are exactly what we are discussing. Otherwise, all we can do is go down to a quantum level and back to the Big Bang and check everything that happened up to this point. And even then, we will find that things happened that were more or less likely than others, because sooner or later events interact in some form and have complex outcomes. That kind of discussion only has philosophical worth, though, and is of no use to any meaningful debate.
Um...you seem to be sorta getting what I'm saying but not getting it.
Morgaln wrote:
Just imagine a bowl with three balls, two black and one white. The chance to draw a black ball is different from the chance of drawing a white ball, and you will never be able to make those chances equal without changing the setup. Of course the chance to draw each of the three balls is equal (although not a simple coin flip, because three options), but the difference between the two black balls is meaningless.
Exactly, you actually have the exact same chance to draw any of the balls, but two of them have been designated as being 'equivalent' outcomes...but that's just a matter of what you decided would be the significant attribute. The chances of drawing each of the actual balls themselves is not changed, you've changed the chances of drawing a ball that belongs to a particular set by defining a set that has more balls in it.
Morgaln wrote:
Both the amount of cars and the amount of rhinos in my area, among other things, are underlying facts that make it more likely to being hit by one and not the other.
You're missing any definition of "underlying" which makes sense in this context. If you omit the word "underlying", then your statement makes sense. But the fact that there are more cars than rhinos isn't "underlying" in any sense here, the greater feasibility of using cars rather than rhinos for transportation would be "underlying" in that they are slightly less obvious than the simple fact that there are many more cars than rhinos. I hate to get hung up on what seems such a trivial point, but I really don't know whether it has some significance to you that is not clear to me.

To clarify my last point, some events fall out of possibility because part of their sequence has become contradictory to established reality. They are no longer merely improbable, they are rendered impossible. It is only the events that are still considered possible which can all have the same probability. Whether or not you consider the actual occurrence of an event to make all other events which could have happened instead become contradictory to established reality is not the issue. When we talk of the "probability" of an event that has already occurred, we must go back and 'suspend' the certainty of the event until it no longer flatly excludes the occurrence of the other events to which we are comparing it in probability. But that means we must do it to the same degree for all other events, we can't go back five seconds for one event (getting hit by one car that is on the road heading towards us five seconds ago rather than another that matches that description), and then compare that probability against a chain of events that must have occurred many hours or days ago (rhino escaping from zoo and wandering the streets till it is in a position for a collision with me). We must wind everything back the same amount to have comparable events. If we only wind back 5 seconds, the rhino hitting me is a simple physical impossibility...unless we want to invoke quantum teleportation.

In which case we have to accept the corresponding multiplication of the quantum variations of the cars hitting me...even though at the Newtonian level a car hitting me may become a near certainty in the seconds before the collision, from a quantum perspective the continued existence and behavior of any of the quantum waveforms which compose the car becomes merely probabilistic, and thus it is impossible to even say that it is the same car, or a car at all, which hits me (leaving aside the plausibly more crazy variations on what it means to be the "me" that is hit). The actual event of the car that hits me in that case is, at the quantum level, just as unlikely as the rhino quantum teleporting (or simply materializing spontaneously as a result of disparate quantum teleportations) to hit me because the Newtonian motion of the car is actually an aggregate of uncountable successive quantum teleportations of its uncountable quantum waveforms. There are just a LOT of those results that result in a car moving to collide with me, and very few that cause the car to turn into a rhino, or cease existing entirely, or whatever.

Take a chunk of any radio isotope. It has a half-life, and at the end of that half-life about half the chunk will have decayed into something else. But which half? At the quantum level there are 2^n different halfs that might be left, where n is the number of atoms of the isotope in the initial chunk (keeping in mind that it won't be exactly half anyway, we can say that statistically it will be within some value of half but that's taking it for granted that we only care about the total number of atoms that have decayed or not rather than which particular ones did).

Anyway, this is all fun but getting off-point. The point being that in order to have any chance of predicting (or understanding) an actual event we have to describe a general category of events that includes a large number of the particular events that have a chance of occurring...we are not actually picking out a specific event at the quantum level, that's impossible even to conceptualize because of the limits quantum mechanical existence would impose on any physical means of sustaining the process of conceptualization of events. All events are picked from an essentially bottomless bag of balls, each with some variable surface albedo, and we draw a category to call most of them "black" and a small number "white", but the real odds that we pick any particular ball are unchanged.
Man, what does his love of all these archons mean???????????
Anomynous 167 wrote:

But the long scale is easier, you don't even have to learn the -iards.
A million is a thousand thousands, a billion is a million millions, and a trillion is a billion billions. The only time we get to a new word is when we experience a data-overflow error.


a billion is a thousand millions, a trillion is a thousand billions. Learn yo' shit