| Author |
Message |
Knyte
2010 SLF Tag Champ*
Title: Curator Of The VGM
Joined: Nov 01 2006
Location: Here I am.
Posts: 6749
|
Found this on io9, and found it interesting:
Generating truly random numbers is notoriously difficult. But now, using a quantum system, researchers have managed to create 42 genuinely random numbers. Their discovery is a major breakthrough for cryptography, and could one day enable truly private communication online.
Violating Bell's inequalities
In today's issue of Nature, a group of scientists describe how they generated those 42 numbers by measuring atoms at various locations. Now here's where we get quantum. The researchers created these numbers via, as Valerio Scarani puts it in Nature, "the violation of Bell inequalities." What that means is that the researchers relied on the fact that the movement of entangled quantum particles isn't dependent on what you might call a "list of instructions" within each particle. You can't predict where the hell a particle is going to go based on the properties of the particle itself - or even the properties of its locality. So there would be no way to guess what a particle's state is going to be at any given moment, based on where or how that particle started. It's just random.
Verifying that the numbers are random
OK, but how do you prove your random number generator is truly creating random numbers? That's even harder than getting those 42 random numbers. According to a release about the Nature study:
Randomness is difficult to prove because it is not readily distinguishable from noise and other uncontrollable factors. Antonio Acín and colleagues show that the non-local correlations of entangled quantum particles can be used to certify the presence of genuine randomness. It is thereby possible to design a cryptographically secure random number generator that does not require any assumption about the internal working of the device. They then perform a proof-of-concept experiment in a system of two entangled atoms in which 42 new random numbers are generated.
In this way, Acín and colleagues hope they're on the way to creating a "black box" that spits out provably random numbers, even if you can't look into the box itself.
What's in it for you?
Sounds great, but how will random numbers help Bob send a private message online that nobody can read except for his partner in crime Alice? Think of it this way. Bob sends his ultra-private emails to Alice by encrypting them, turning his dark secrets into a code so nobody at Alice's email provider can grab them and start reading. But how does Alice read them? Well, Alice needs a key to unlock the code.
The fact is, your code, no matter how tricky, is really only as strong as your key. If Alice's key is "iwuvbob," it's going to be easy for Eve the spy to decrypt Bob's emails. But if Alice's key is randomly generated by this black box using quantum entanglement . . . well, that key is going to be pretty damn hard for Eve to guess. Especially if it's 128 characters long, or longer.
So that, my privacy-loving friends, is why you need random numbers to send private information online. Today you got just one step closer to foolproof privacy.
|
|
|
   |
|
username
Title: owner of a lonely heart
Joined: Jul 06 2007
Location: phoenix, az usa
Posts: 16136
|
huh?
|

| Klimbatize wrote: |
| I'll eat a turkey sandwich while blowing my load |
|
|
     |
|
Knyte
2010 SLF Tag Champ*
Title: Curator Of The VGM
Joined: Nov 01 2006
Location: Here I am.
Posts: 6749
|
Up until now, there has never been a true Random Number Generator. If you tell a computer to pick a random number, it has to have some kind of reference to do it. The most common way, is that the computer will use the number of the CPU's clock cycle at the time in which you ask for it. (Down to the milliseconds.) So if ask for a random number between 1 and 100, the computer will look to see what time it is. So if the time is "12:31:313516", it will pick the number "16".
If you know what the number's seed is, then you can crack what the random number will be. If there is no seed, then it is truly random, and much better to use for security applications and such.
|
|
|
   |
|
glycerine92
Joined: Dec 06 2008
Location: Pearland, TX
Posts: 88
|
I'm not sure how to feel about this.
|
 ^^^All genius^^^ |
|
   |
|
username
Title: owner of a lonely heart
Joined: Jul 06 2007
Location: phoenix, az usa
Posts: 16136
|
ok, the abridged explanation made more sense.
but still, how can you tell it truly is a random number, and not just a collection of different algorithms all coded into one program?
i guess thats what the OP mentions
|

| Klimbatize wrote: |
| I'll eat a turkey sandwich while blowing my load |
|
|
     |
|
Slayer1
Title: ,,!,, for you know who
Joined: Sep 23 2008
Posts: 4274
|
And here my Statistics teacher always thought the computer was random...
|
|
|
  |
|
GPFontaine
Joined: Dec 06 2007
Location: Connecticut
Posts: 11244
|
| username wrote: |
| but still, how can you tell it truly is a random number, and not just a collection of different algorithms all coded into one program? |
I am not sure what the actual name is, but there is a theory that every event on earth can be predicted if you had a scientist who was smart enough to understand every variable. This "Super Scientist" theory describes how every single event since the Big Bang can be shown as relative to past and concurrent events. Anyone who believes in this type of theory would say that no number is random even if your source is based on randomness in nature because there is no such thing as random, only what we don't understand yet.
If however you believe that things have the ability to act and react without predictability, then you can assign numerical values to actions and call it randomness. That is pretty much what is going on here.
So, I suppose you could say it is as random as your faith allows it to be.
Personally I think it is as random as we can get, and something like this won't be cracked for security reasons within our lifetimes. That makes it good/random enough.
|
|
|
   |
|
username
Title: owner of a lonely heart
Joined: Jul 06 2007
Location: phoenix, az usa
Posts: 16136
|
| GPFontaine wrote: |
| username wrote: |
| but still, how can you tell it truly is a random number, and not just a collection of different algorithms all coded into one program? |
I am not sure what the actual name is, but there is a theory that every event on earth can be predicted if you had a scientist who was smart enough to understand every variable. This "Super Scientist" theory describes how every single event since the Big Bang can be shown as relative to past and concurrent events. Anyone who believes in this type of theory would say that no number is random even if your source is based on randomness in nature because there is no such thing as random, only what we don't understand yet.
If however you believe that things have the ability to act and react without predictability, then you can assign numerical values to actions and call it randomness. That is pretty much what is going on here.
So, I suppose you could say it is as random as your faith allows it to be.
Personally I think it is as random as we can get, and something like this won't be cracked for security reasons within our lifetimes. That makes it good/random enough. |
that makes sense too. especially to those who believe in predestination.
|

| Klimbatize wrote: |
| I'll eat a turkey sandwich while blowing my load |
|
|
     |
|
FNJ
2010 SLF Tag Champ
Joined: Jun 07 2006
Posts: 12294
|
if there are only 42 completely random numbers, wouldn't that actually make it easier for people to crack shit, because tehy'd jsut have 42 numbers to try and plug into something?
|
|
|
  |
|
UsaSatsui
Title: The White Rabbit
Joined: May 25 2008
Location: Hiding
Posts: 7565
|
How can you pull a truly random number from an infinite pool of numbers, rational and irrational, real and imaginary?
|
|
|
  |
|
Valdronius
Moderator
Title: SydLexia COO
Joined: Aug 22 2005
Location: The Great White North
Posts: 4465
|
| GPFontaine wrote: |
| So, I suppose you could say it is as random as your faith allows it to be. |
Scientists have no faith.
|

| Klimbatize wrote: |
| A Hispanic dude living in Arizona knows a lot of Latinas? That's fucking odd. |
|
|
   |
|
Doddsino
Joined: Oct 01 2009
Posts: 5316
|
Joe Biden: "This is a big fuckin' deal!"
|
|
|
  |
|
FranzKefka
Title: The Onion Knight
Joined: Apr 14 2010
Location: Wales, UK
Posts: 24
|
I was always under the impression that quantum physics was like the Emperor's New Clothes (or perhaps I can't accept that I'm too stupid to understand it).
|
|
|
   |
|
Probable Muppet
Joined: Aug 05 2008
Location: CA
Posts: 867
|
| GPFontaine wrote: |
| username wrote: |
| but still, how can you tell it truly is a random number, and not just a collection of different algorithms all coded into one program? |
I am not sure what the actual name is, but there is a theory that every event on earth can be predicted if you had a scientist who was smart enough to understand every variable. This "Super Scientist" theory describes how every single event since the Big Bang can be shown as relative to past and concurrent events. Anyone who believes in this type of theory would say that no number is random even if your source is based on randomness in nature because there is no such thing as random, only what we don't understand yet.
If however you believe that things have the ability to act and react without predictability, then you can assign numerical values to actions and call it randomness. That is pretty much what is going on here.
So, I suppose you could say it is as random as your faith allows it to be.
Personally I think it is as random as we can get, and something like this won't be cracked for security reasons within our lifetimes. That makes it good/random enough. |
I think you might be talking about Universality
|
|
|
  |
|
Blackout
Title: Captain Oblivious
Joined: Sep 01 2007
Location: That Rainy State
Posts: 10376
|
|
     |
|
UsaSatsui
Title: The White Rabbit
Joined: May 25 2008
Location: Hiding
Posts: 7565
|
| Blackout wrote: |
| Just roll D20. |
That actually won't give you a truly random result. Imperfections in the die, as well as the material they cut out of it to make the numbers, make it impossible. And that assumes a non-biased roller.
|
|
|
  |
|
Knyte
2010 SLF Tag Champ*
Title: Curator Of The VGM
Joined: Nov 01 2006
Location: Here I am.
Posts: 6749
|
| UsaSatsui wrote: |
| Blackout wrote: |
| Just roll D20. |
That actually won't give you a truly random result. Imperfections in the die, as well as the material they cut out of it to make the numbers, make it impossible. And that assumes a non-biased roller. |
Since, different faces have different numbers carved into them, each face has a different weight to it.
I personally believe that a d20 has the best odds of rolling a "20". Because the "20" face has the most material cut out of it, so it is the lightest face, while on the opposite side, the "1" has the least ammount of materal cut out of it, making it the heaviest face.
|
|
|
   |
|
GPFontaine
Joined: Dec 06 2007
Location: Connecticut
Posts: 11244
|
| Knyte wrote: |
| UsaSatsui wrote: |
| Blackout wrote: |
| Just roll D20. |
That actually won't give you a truly random result. Imperfections in the die, as well as the material they cut out of it to make the numbers, make it impossible. And that assumes a non-biased roller. |
Since, different faces have different numbers carved into them, each face has a different weight to it.
I personally believe that a d20 has the best odds of rolling a "20". Because the "20" face has the most material cut out of it, so it is the lightest face, while on the opposite side, the "1" has the least ammount of materal cut out of it, making it the heaviest face. |
If a computer were rolling a D20 in a perfect environment I would agree as long as the starting position of the die began with a new number face up each time it was rolled.
If a person is rolling there are too many variables for weight of the carved face to matter. The starting position of the roll, the angle of the toss, the friction of the hand, the friction of the landing surfaces, the air temperature, the wind speed, and gravity along with initial force. Lastly the die would most likely have sweat, finger prints, or other human transfer moved to it altering the weight slightly as it was rolled.
|
|
|
   |
|
Rycona
Moderator
Title: The Maestro
Joined: Nov 01 2005
Location: Away from Emerald Weapon
Posts: 2815
|
I don't think this method is truly random. It may appear random now, but what if one day we understand the forces that may truly be at work with quantum mechanics. If there are such forces, they are likely able to be made into an equation, therefore making this seed not truly random even without actually knowing that hypothetical equation. It's just random from our current perception and understanding.
|
 RIP Hacker. |
|
   |
|
|
|