@DJMcMayhem I browse /tg/ (traditional games) and /ck/ (food) occasionally and aside from the typical reddit/4chan name-calling shit they're not that bad. This was far worse than what I was expecting though
This would get me called an idiot on both 4chan and reddit but I actually don't see a whole lot of difference between the culture of the two. Reddit is just more tame usually because the off topic stuff is spread out across a lot of related subreddits while 4chan only has one board for a thing
I think that the random string is generated in several "chunks". That is, the output of the mersenne twister algo each time it is run includes a chunk of characters (output) and a "generator" that acts for the seed of the next chunk
Zwei: Generally, a pseudorandom number generator just generates output that "appears random". PRNG's do tend to generate numbers based on a stored state, and some function. Not sure what you mean by "secretly pure"...
But let's say I have a PRNG. I could use that to encrypt a message--simply generate numbers using the PRNG, and xor it to a message. This may or may not be a hard to crack encryption scheme, depending on the PRNG. Roughly if it is, this is a "cryptographically secure PRNG" or CSPRNG. If not, it's not.
anyways, I think the idea is to go from last character to first character
find a seed that generates the last character you want, then find a seed that generates that seed/character combination, and repeat until you get to the front of the string
I'm going to say that as long as it is of reasonable size, not too low because of an inappropriate challenge, or too large because they weren't even trying, and not obviously the result of a bitfuck to Turtlèd translator someone made, I will award 50 rep
I recently revisited the following classic challenge: Showcase your language one vote at a time. The gist of the challenge is as follows:
An answer is a set of program snippets, each with a unique length.
The maximum length and maximum number of snippets are given by the answer's current vote ...
@ConorO'Brien ok the dynamic STL items stuff is on leading branch so you can get the new changes using git fetch origin && git rebase origin/develop -X theirs
then you'll be able to use them the way I described earlier
reticular + Turtlèd
"trick"oll"eat";
Try reticular online! Try Turtlèd online!
reticular explanation:
"trick"oll"eat";
"trick" push the string "trick" to the stack
o output it
l push length of stack (0)
l push length of stack (1)
...
@LuisMendo I was looking over the "Functions arising by coin flipping" paper, and realized that, in the case of finite events, matrices are really useful in describing the relationship between the coefficients of the target polynomial and the coefficients of the coin-flipping polynomial.
@miles have you seen that googol challenge? I've been thinking about doing it in J, but I don't know if there's a way to avoid using an explicit verb to use for.. Thoughts?
here is how it works. You have string var, #Foo#, and string pointer. if the pointer points to the very last char of string var, the char var, initially *, set with @ Will be written to the current cell. Otherwise, a space will be writetn
I am not very good at golfing by myself, but once in a while, I dome across something small that I notice that could help. These are typically tricks I have found while looking at tips. Some users would not think of these, as some are very strange. I would like to comment, but I do not have 50 re...
@LuisMendo For example, let's say that you wanted to create the function x - x^2 + x^4, then the relevant coin-flipping polynomial is 0*x^0*(1-x)^4 + 1*x^1*(1-x)^3 + 2*x^2*(1-x)^2 + 1*x^3*(1-x)^1 + 1*x^4*(1-x)^0. This is given by the following matrix multiplication:
@PhiNotPi Does that work in general? Take example 4.4 in the paper, i.e. target polynomial 1−8x+20x^2−13x^3. That one requires a coin-flipping polynomial of degree 46
But this can't work for all polynomials? Some polynomials don't have a coin-flipping polynomial (i.e. are not realizable as the probability of a finite coin-flipping event)
And, in many cases it is clear that it will never become valid (because there's enough negative numbers that they never die out). But in other cases (I believe in exactly the cases in which a solution exists), eventually a valid state will be reached.
@PhiNotPi That's interesting, in connection with the open problem (well, I think it's still open) mentioned on page 21, right before the ackonolwegments
Here's the steps of my "algorithm:" (1) take the polynomial 1-8x+20x^2-13x^3 and represent it as [1 -8 20 -13]. (2) multiply it by the corresponding 4x4 pascal-triangle-matrix. (3) Take the result [1 -5 7 0] and apply pascalization until it is valid (which may never happen).