If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(University of Cambridge)   Quantum 'sealed envelope' system enables "perfectly secure" information storage   (cam.ac.uk) divider line 28
    More: Interesting, quantum, special relativity, overwing exit, applied mathematicses, quantum cryptography, Niels Bohr, old quantum theory, Cambridge's Department of Applied Mathematics  
•       •       •

1302 clicks; posted to Geek » on 04 Nov 2013 at 8:09 AM (1 year ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



28 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest
 
2013-11-04 06:18:28 AM  
If information can be seen by one person, it can be seen by another - it will just take a extra dose of trickery.
 
2013-11-04 07:23:33 AM  
No one has EVER claimed to have created an uncrackable code before.
 
2013-11-04 08:05:01 AM  

colinspooky: If information can be seen by one person, it can be seen by another - it will just take a extra dose of trickery.


Or a $5 wrench.
 
2013-11-04 08:07:16 AM  

devildog123: No one has EVER claimed to have created an uncrackable code before.


Actually, there is one kind of uncrackable encryption:  Using a one time pad.

If you follow the very simple rules of OTP use, the secrets are safe forever.  That's why it's still used for critical national security-type applications, for example by  governments to communicate with their agents in foreign countries.
 
2013-11-04 08:09:43 AM  
I didn't see any mention of "man in the middle" attacks where the "man in the middle" can fark up the message just by reading (or attempting to read) the message, thus denying the communications.

Best encryption in the world is useless if your opponent can prevent you from communicating.
 
2013-11-04 08:29:33 AM  

dittybopper: I didn't see any mention of "man in the middle" attacks where the "man in the middle" can fark up the message just by reading (or attempting to read) the message, thus denying the communications.

Best encryption in the world is useless if your opponent can prevent you from communicating.


Agreed, but here at least you know immediately if your network is compromised.
 
2013-11-04 08:30:07 AM  

colinspooky: If information can be seen by one person, it can be seen by another - it will just take a extra dose of trickery.


This.
 
2013-11-04 08:36:38 AM  

dittybopper: Actually, there is one kind of uncrackable encryption:  Using a one time pad.


In the Cryptonomicon, it was suggested that one time pads are only as secure as the randomness of the random number generators they are based on.

Also, last week I recall a submission that utilized observation of quantum states without entanglement (indicating the universe does not change, but that time is merely a byproduct of quantum mechanics) and, thus, not change the spin. I'm not even sure how observation without entanglement is possible, but it seems like that could be a way to read a message without anyone finding out.
 
2013-11-04 08:49:28 AM  
joshua
 
2013-11-04 08:51:41 AM  

Feepit: dittybopper: Actually, there is one kind of uncrackable encryption:  Using a one time pad.

In the Cryptonomicon, it was suggested that one time pads are only as secure as the randomness of the random number generators they are based on.


That's actually true.  Most computer "random number generators" are actually pseudo-random number generators, and because their output can be predicted, they are useless for generating one time pads.

There are some computerized versions that can be used, but then the weak link is the computer:  It has to be completely secured against intrusion, which is nigh-on impossible for mere mortals.  You're better off using a low-tech hardware random number generator:

img.fark.net

It is virtually impossible to "bug" a manual typewriter*, the ribbon and carbon (if used) can be easily destroyed, and the dice can generate non-deterministic random numbers.

That's the real trick:  There can be a statistically significant bias in the numbers, but if there is no algorithm that can determine the next number or group of numbers based upon the previous ones, it's still cryptographically secure.   Soviet one time pads from the 1950's had some interesting statistical anomalies suggesting that they weren't truly random, but were done by typists, but they didn't have enough of a pattern to be able to break them cryptanalytically.


*Manual, meaning no electricity, the entire energy for operation is supplied by the movement of the operator.  Electric typewriters have been bugged.
 
2013-11-04 09:06:53 AM  

Feepit: entanglement is possible


Einstein was walking on the beach (iirc) with someone and mentioned that, by observing things, we're changing them.

Bizzare stuff, I've just spent the last few hours reading about a. 2001/monolith, its sequels, von neumann probes, our location in the Milky Way (our impending collision with the Andromeda galaxy)...

/TOO MANY SECRETS
 
2013-11-04 09:10:57 AM  
Some notes: Subby, the article only mentions transmission of encrypted data, no word on storage. Quantum encryption is hardly new, is this a record distance or something?

Finally, as far as one time pads go, random numbers are painfully slow to produce. Computers can do it slowly(best way is likely an old NTSC capture device tuned to an unused channel and hash the static, but anybody who cares enough about encryption to do a one time pad isn't going to be someone isn't jamming the channel), but it takes time to find something random "enough". Also there is that issue that you still need to deliver just as much data in the one time pad securely (no point in encrypting it) as you will in the final encrypted data. The difference is that you can use the "up-armored humvee stuffed with USB hard drives (again, note the weak link of the computers)" to send the one time pads then send the resulting encrypted data over the internet.
 
2013-11-04 09:31:07 AM  
And I hear that writing your data to the special Unix file /dev/null is very, VERY fast and super secure!!!  No one's ever going to read your data after that!!

:)

/knows Unix
 
2013-11-04 09:38:06 AM  

dittybopper: Actually, there is one kind of uncrackable encryption


It's not uncrackable, the ciphertext just doesn't provide any information.
 
2013-11-04 09:38:45 AM  

Feepit: dittybopper: Actually, there is one kind of uncrackable encryption:  Using a one time pad.

In the Cryptonomicon, it was suggested that one time pads are only as secure as the randomness of the random number generators they are based on.

Also, last week I recall a submission that utilized observation of quantum states without entanglement (indicating the universe does not change, but that time is merely a byproduct of quantum mechanics) and, thus, not change the spin. I'm not even sure how observation without entanglement is possible, but it seems like that could be a way to read a message without anyone finding out.


What if you just read it twice?
 
2013-11-04 09:41:03 AM  
I'll prove it. Send me your credit card info, and I will protect it in my specially-produced, limited-edition, ultra-secure quantum sealed envelope. Completely secure. Trust me.
 
2013-11-04 09:55:18 AM  

sprawl15: dittybopper: Actually, there is one kind of uncrackable encryption

It's not uncrackable, the ciphertext just doesn't provide any information.


Well, you can brute-force it, but you just end up compiling a list of all possible plaintexts the same length, which doesn't help you:  The correct plaintext would be known to you, but you have no way to separate it from all the incorrect ones.  Was it "HELP ME", "SAVE ME", "KILL ED", "STAB JO", "SUCK IT", etc.?

No possible way to tell, so secrecy is preserved.
 
2013-11-04 09:56:23 AM  

dionysusaur: Feepit: dittybopper: Actually, there is one kind of uncrackable encryption:  Using a one time pad.

In the Cryptonomicon, it was suggested that one time pads are only as secure as the randomness of the random number generators they are based on.

Also, last week I recall a submission that utilized observation of quantum states without entanglement (indicating the universe does not change, but that time is merely a byproduct of quantum mechanics) and, thus, not change the spin. I'm not even sure how observation without entanglement is possible, but it seems like that could be a way to read a message without anyone finding out.

What if you just read it twice?


Forcing anyone to read the Craptonomicon more than once is just plain cruel.
 
2013-11-04 09:56:33 AM  
imgs.xkcd.com
 
2013-11-04 10:10:09 AM  

dittybopper: but you just end up compiling a list of all possible plaintexts the same length


Which assumes both that the message is plaintext and has no garbage characters, two assumptions that you gained from sources other than the ciphertext. That's why I made the distinction I made, and why that's kind of an important distinction. Especially when talking about systems like the one in TFA that claims no vulnerability to secondary sources.
 
2013-11-04 10:15:27 AM  

Tyrone Slothrop: [imgs.xkcd.com image 448x274]


imgs.xkcd.com
 
2013-11-04 11:10:19 AM  
I don't believe any science article that cites Einstein or gives primers on black holes, the speed of light and how nothing can go faster than it or evolution.
 
2013-11-04 11:51:58 AM  

mrlewish: I don't believe any science article that cites Einstein


Einstein once said, "I haven't said everything pithy, but in the future, I will have."
 
2013-11-04 12:04:00 PM  

sprawl15: dittybopper: but you just end up compiling a list of all possible plaintexts the same length

Which assumes both that the message is plaintext and has no garbage characters, two assumptions that you gained from sources other than the ciphertext. That's why I made the distinction I made, and why that's kind of an important distinction. Especially when talking about systems like the one in TFA that claims no vulnerability to secondary sources.


If you don't have to worry about bandwidth issues, you can send the balance of the entire pad to, in essence, "pad out" the message.  That way, there isn't anyway for an eavesdropper to know the actual length of the message.  Here is an example:

Say I wanted to send the message "sprawl15 is cool."  It's a nice short message.  So we encode it using the straddling checkerboard:

 |0 1 2 3 4 5 6 7 8 9
 |A   S I N   T O E R
1|B C D F G H J K L M
5|P Q U V W X Y Z . /

S  P  R  A  W  L  /  111 555  /  I  S  C  O  O  L  .
2  51 9  0  54 18 59 111 555  59 3  2  11 7  7  18 58


So we then add that to the pad, converting into 5 figure groups for ease of reading:

25190 54185 91115 55593 21177 1858
88254 83402 02029 48286 92288 75235
----- ----- ----- ----- ----- -----
03344 37587 93134 93779 13355 83715


And we just add the rest of the pad to it, because it's just going to be destroyed anyway, so we end up with something like this (with the page at the beginning):

05231 05231
03344 37587 93134 93779 13355
83715 74209 56361 29287 21099
34997 78488 33268 14349 47824
98740 79443 01369 97010 36292
68086 34509 88345 23771 43639
46726 31704 41347 18797 20708
44588 52627 50903 60352 99426
63376 58119 02778 37355 85235
07420 70322 58626 64161 26497
64244 21232 38842 22697 92963


I've underlined the length of the actual message, but there's no way for a cryptanalyst to know exactly how long that message is, or even if it's just part of a larger message.  He doesn't even know if it's just random crap intended to stifle traffic analysis.

When the intended recipient gets the message and starts decoding it, this is what they'll get:

03344 37587 93134 93779 13355 83715 74209 56361 29287 21099
88254 83402 02029 48286 92288 75235 74209 56361 29287 21099
----- ----- ----- ----- ----- ----- ----- ----- ----- -----
25190 54185 91115 55593 21177 18580 00000 00000 00000 00000
SP RA W L /  1  5   / I SC OO L . A AAAAA AAAAA AAAAA AAAAA


It's pretty damned obvious where the message ends.
 
2013-11-04 12:08:29 PM  

syrynxx: mrlewish: I don't believe any science article that cites Einstein

Einstein once said, "I haven't said everything pithy, but in the future, I will have."


Niels Bohr is like a dose of the clap.  Before he arrives is pleasure, and after a pain in the dong.
 
2013-11-04 12:10:47 PM  
dittybopper:

could you put that in a memo for me and title it "shiat i already know and mentioned in the post you responded to"

thanks
 
2013-11-04 12:24:02 PM  

sprawl15: dittybopper:

could you put that in a memo for me and title it "shiat i already know and mentioned in the post you responded to"

thanks


Yes, I know, but my thing is spreading the information about it.  You know that also ;-).
 
2013-11-04 01:51:38 PM  
Quantum grammar system ensures you'll never get to space.
 
Displayed 28 of 28 comments

View Voting Results: Smartest and Funniest


This thread is closed to new comments.

Continue Farking
Submit a Link »
On Twitter





In Other Media


  1. Links are submitted by members of the Fark community.

  2. When community members submit a link, they also write a custom headline for the story.

  3. Other Farkers comment on the links. This is the number of comments. Click here to read them.

  4. Click here to submit a link.

Report