No, you misread me. In that case where you know something, I'm saying it's not random. So either you know absolutely nothing about the deck (effectively random) or you know something about the deck (not random).
Sorry, but I don't see how this sentence means anything different from what I said as below:(aside from the claim that I made this distinction is useless)
What I see you are saying is that you either know completely about a deck (not random), or you know it from partially to none (random). That is sure binary, but a rather useless statement.
Ok I guess I need to elaborate.
By your definition, for a shuffled 52-card deck, if I somehow only know where SA is, is this random? I guess not.
How about that I only know SA is among the upper half 26 cards? I guess not.
Or if I only know that SA is next to SK? not random.
How about that I only know SA has a higher than usual probability let's say (25%) next to SK? still not random.
If you agree with my distinction here (which I assume so), then practically any hand-shuffled deck is not random. Even random generators in the strict sense does not generate random numbers. This distinction is thus useless.
(note that in the above discussion "random" means "uniformly random", as there is no ambiguity)
Okay. The stuff where you elaborate matches what I've been saying, though probably I am a little wrong here, and the 25% chance kind of situations are indeed random. The reason I resist saying this is because insofar as this matters for a card game, it is not sufficient to have it randomized to this extent - what is needed is the players having absolutely no idea about cards, their placement relative to each other, etc.
The big thing is, what you say in the quoted section above seems to me to be the opposite of what you say in the unquoted section. Maybe I was misunderstanding. But if the 'let me elaborate' stuff is how you understand, then your understanding of what I'm saying is right.
No, this is the whole point. This doesn't give you different random distributions. It gives you different NOT random distributions.
Now here the terminology changes. When you say "random" without further specifications, depending on context it can mean a few things, including a uniform distribution, as the above discussion. But when you say "random distributions" (especially when I have cited "uniform distribution" as a special kind of random distribution), in any scientific context I have seen, it just mean something can take certain values with certain probabilities. Come on, if you work in Physics you must know about expontential distribution, Gaussian, or Maxwell distributions right? They are all random distributions, but not uniform in any sense.
Well, you have thousands of kinds of distributions, and I in no way mean to imply that uniform is the only kind of random distribution in general. Insofar as a card deck is concerned though, it's the only meaningful one. Of course you can have other distributions, and whether they really are random or not IS in fact the philosophical question - there is a big question about about whether random number generators really are random or not, and you can say that's stupid, whatever, but I don't think it is, and there are lots of philosophers at universities who don't either. Now, I realize I'm weird, and that's fine, but it's not a settled thing and not just a 'come on, that's stupid' thing either.
Again, I don't think anything compensates for insufficient shuffling except for sufficient shuffling.
Well, we can still play word games here. Sure, shuffle more is better. And probably I shouldn't use the word "compensate". But my point (and the question from the open post) is, ok, given insufficient shuffle, do different ways of dealing give different results? And the answer I think is yes, and in most cases dealing one card at a time will deal you hands which are statistically more likely from a uniform random distribution, for the purpose of the game.
Of course they give you different results! And my large point is this - these shuffling methods give you hands which more closely resemble VARIED hands, but people associate randomness with these varied hands, when randomness indeed gives you more clumped hands than people think. The same reason why if I ask a million people to come up with a random number between 1 and 1000, the numbers like 100, 200, etc. won't come up as often as they 'ought' to, because people don't think of these as 'random' numbers, just because they're round.
Well, my problem goes back to the philosophical question - are there really things that aren't deterministic like that? Well, I think so, but they're effectively quite rare when we're talking about a physical process like shuffling, especially on a macroscopic scale.
I use the 'effectively random' label to talk about things which I know are deterministic, but which none of the pertinent actors are actually consciously determining. If that makes sense.
I don't think this has anything to do with the philosophical question-- and it is even not a philosophical question (unless you take the side saying everything is a philosophical question.)
For all practical random things we see in a game, they are actually all deterministic. When you throw a dice, which face ends up on top is completely determined by the initial and boundary conditions. The "randomness" comes about by both the sensitivity to the initial and boundary conditions, and the inability to control those conditions. For a shuffled deck it is similar; the deck once dealt is of course fixed. It is just that we neither control the precise way we shuffle nor remember the starting deck before shuffle so that we have no idea how it will order. The reason why both cases a uniform random distribution emerges is because all the factors that determine the outcome is somewhat insensitive to the spots of the dice or the numbers and drawings on the card so by symmetry we can deduce that if we shuffle long enough or when we throw the dice from high enough the outcome will be uniformly random.
This certainly has nothing to do with whether a Gaussian distribution is random or not random. This also has nothing to do with quantum mechanics and Copenhagen interpretation.
1. Who said anything about a Gaussian distribution?
2. It does have something to do with quantum mechanics and the Copenhagen interpretation, because Copenhagen will tell you that the first thing, which you take for granted, about things being deterministic, is in fact wrong, and you actually CAN get things which are TRULY random. Of course this is impractical, but I never tried to bring this up as an actual practical consideration, only in the philosophical realm.
Again, my larger point is that it's not really a uniform distribution that emerges, but the important thing is that the people playing can't tell the difference between the actual distribution and the truly uniform distribution, that people can't predict the distribution better than by guessing how they would via a uniform distribution of the cards. And what I am saying regarding shuffling here is that stacking the cards in some particular way does not help achieve this goal at all. How you shuffle thereafter, how non-standardized your shuffling pattern is, how long you shuffle, etc. WILL help achieve this goal, but stacking the similar cards does not, because it doesn't actually make your guessing of the thing closer to random, but rather you replace your rough ability to estimate that SK and SA are near to each other with a rough ability to estimate that they're far apart.
