Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[bunnyThor]

Check it out. When you hit random strip, you never get a strip past 800.

At least I didn't, and I clicked it about 60 times. 

[C Patrick Carolan]

I just got 816 on my fourth or fifth click.

[MC]

the highest I was able to get up to was 817. So I'm able to get past 800 but not very far past 800. I remember there being a problem like this before where the random would only go up to strip 500. I think Jeph had to go fix it last time [can't remember] I think for some reason the random button might stop registering new strips or something. Computers are weird, we make em, we can't understand em.

[Jepser]

I also noticed that some strips turn up suspiciously often when you random for some time. 

[Rocketman]

I also noticed that some strips turn up suspiciously often when you random for some time. 

That's the nature of randomness. Avoiding strips because they've been selected already imposes order.

[JorickHorn]

Wow. 50 trials, no 800s. I got both #1 and #2, but nothing higher than 789.

[Hobolicious]

Hmm did it 75 times and got 2 : 813 and 809 got a lot of 600's though

[Is it cold in here?]

Is the random button powered by ohnorobot? They only go up to the 800s.

[Floydian]

Is the random button powered by ohnorobot?

It seems to be. I don't really see why though, it's not exactly hard to write a small script to show a random comic.

[selkie_upsilon]

woo, this can be analysed

assuming button is truly random and will get strips 1-1134 equally

probability of getting "old" strips= 800/1134 = 0.71
probability of getting "new" strips = (1134-800)/1134 = 0.29

probability of getting "old" strips 50 times in a row = (0.71)^50 = 0.000000036 (silly small number)
this was achieved by JorickHorn, so, random button is probably broken...

how broken? maybe it only gets up to say, 816 strips (as shown by C Patrick Carolan)

so now
probability of getting "old" strips= 800/816 = 0.98
probability of getting strips from 801 to 816 = (816-800)/816 = 0.02
probability of getting any strips beyond 816 = 0

probability of getting "old" strips 50 times in a row = (0.98)^50 = 0.36 (more than one third of the time)
probability of getting at least 1 strip between 801 to 816 = (1-0.36) = 0.64 (more than half the time)
probability of getting 1 to 3 strips between 801 to 816 after clicking 75 times in a row = 75*(0.98)^74*(0.02) + 2775*(0.98)^73*(0.02)^2 + 67525*(0.98)^72*(0.02)^3 = 0.72 (nearly three quarters of the time)
as achieved by Hobolicious =)

this is fun.

_{I'm bored. Don't hit me.}

[snalin]

God damn you, that is the only part of math that I do not understand. Yet.

And it needs to be fixed, because brokern things are bad things.

[puck]

You have way too much time on your hand Selkie. But being a techie I appreciate the math involved.

[C Patrick Carolan]

HEAD A SPLODE.

[KamikaziCal]

I failed stats in freshman year

I still blame beer.

[NuclearDog]

http://upload.nucleardog.com/49/qcanalysis2500.gif

Made 2500 requests to ohnorobot.com and recorded the resulting comic ID, graphed them (X - request #, Y - comic ID).

/ Very bored.

ND

[Misel]

http://xkcd.com/221/

*SCNR 

[KamikaziCal]

In high school computer science class, everytime the teacher asked us a question, we would answer 4...
It made that strip much funnier for me.
And this was my fun fact of the day.

[smellslikemagic]

That graph....it makes me strangely happy.  I don't know what conclusions we can draw, but I think I am okay with the fact that it was made. 

[Doug S. Machina]

I'm not sure what that graph is telling me, but it looks random, or at least there's no visible patterns. I thought the selections would be psuedo-random.

[mrjoegangles]

Its obviously a conspiracy headed up by our own computers.   There must be some crucial piece of information in one of these newer strips that we will need to defeat the machines in the future. Maybe its the IP address to SkyNet or something.

Someone call Arnold.  We got work to do.

[Norrit]

Alright, a few geeky things really quickly. Computers at their root cannot be completely random, it is one of their faults. They can be so close to random that they seem it to the casual observer but that would be due to very complicated algorithms.

"A "random number generator" based solely on deterministic computation cannot be regarded as a "true" random number generator, since its output is inherently predictable. John von Neumann famously said "Anyone who uses arithmetic methods to produce random numbers is in a state of sin." How to distinguish a "true" random number from the output of a pseudo-random number generator is a very difficult problem. However, carefully chosen pseudo-random number generators can be used instead of true random numbers in many applications. Rigorous statistical analysis of the output is often needed to have confidence in the algorithm."
http://en.wikipedia.org/wiki/Random_number_generator#.22True.22_random_numbers_vs._pseudo-random_numbers

Now on to the graph, I love that graph. It seems to me in my very basic understanding of Chaos or Complexity, to hold true to some of the signs of chaos. Periodic and seemingly random, maybe we should get the guys from XKCD to analyze.

[snalin]

http://upload.nucleardog.com/49/qcanalysis2500.gif

Made 2500 requests to ohnorobot.com and recorded the resulting comic ID, graphed them (X - request #, Y - comic ID).

/ Very bored.

ND

Now, make a graph showing what strips you got most of, with intervals [0,50> [50,100> and so on.

[css.pirate]

Periodic and seemingly random, maybe we should get the guys from XKCD to analyze.

You might have a hard time getting "the guys" (near as I can tell, it's just ONE, Randall Munroe) to analyze it if you refer to it as "XKCD": the preferred form is "xkcd"--all lowercase.

[IKantSpell]

Dilbert.  Yes, Dilbert.

[Norrit]

Periodic and seemingly random, maybe we should get the guys from XKCD to analyze.

You might have a hard time getting "the guys" (near as I can tell, it's just ONE, Randall Munroe) to analyze it if you refer to it as "XKCD": the preferred form is "xkcd"--all lowercase.

Well css.pirate, I was actually referring to the community and the forums. They have very intelligent people there discussing complex math all the time. I know that Randall Munroe is just one man and very busy most of the time, thank you for updating me on that.

[GHIGHATRON]

GHIGHATRON HAD RANDOM STRIP BUTTON ONCE.

GHIGHATRON HAD REMOVED BECAUSE ORGANO-CHILDREN PRESS BUTTON AND GHIGHATRON GET NAKED/ARRESTED.

[lolaAria]

Depending on how the random number generator was created, it's possible that the max limit has to be adjusted. Statistically speaking, the generator would have to eventually pull a comic between 818 and 12xx (I forget what it currently is, 1225?). The fact that it doesn't leads me to believe that someone needs to alter the CONSTANT MAX line of the generator code.

[hapaxLegomina]

I kind of like the idea of not getting recent strips in a random link. 99% of the people using the link, at least in a serial like this, have already read most if not all of the strips, especially the most recent ones. Always returning older comics means that "random" comics are those less likely to be fresh in memory.