What is Digg?Sponsored by Dragon Age: Origins
Can't get enough Dragon Age: Origins? Play the flash game. view!
DragonAgeJourneys.com - Play the free companion flash game to Dragon Age: Origins.
115 Comments
- parax, on 04/29/2008, -2/+24Simple algorithm to break quantum encryption.
1. Buy gun.
2. Stand next to receiver with gun.
3. Read message.
4. Kill receiver.
5. Code broken! As an added bonus, the sender won't know you broke it! - karmabandit, on 04/29/2008, -0/+16Most likely people will still break it due to human error, or due to people using the same quantum-encrypted key to send multiple unencrypted messages. If you mis-use a perfect system, it will still be flawed.
- martoq, on 04/29/2008, -3/+17Never use the word never when talking about cryptography and codebreaking.
- jeuhrn, on 04/29/2008, -0/+10Oh *****, they hadn't thought of THAT! Years of research, millions of dollars! And here you come, "*****", and put the proverbial stick through the spokes of their wheel. :(
Wouldn't you have to look at the stuff you're trying to virtualize, or in any extent look at the result of your virtualisation, thus breaking it? - dkapuchino, on 04/29/2008, -1/+9Never allow TV shows that show some "genius hacker-cracker-biologist" to shape your conceptions about cryptography.
One time pads have been Mathematically proven to be secure. NEVER will you be able to obtain the plaintext encryted with a one time pad, unless you have the key. - NathanielJ, on 04/29/2008, -0/+8That article was significantly better-written than anything else about quantum cryptography/networks/computers that's been on Digg in the last few months. Dugg for not over-sensationalizing and spouting crap like "quantum computers can perform twice as many tasks for every new qubit they get!"
- trogdor282, on 04/29/2008, -0/+8The whole point is that that's impossible. Not just hard, impossible.
- karmabandit, on 04/29/2008, -0/+7Not only has it been worked out for decades, there are actual working demonstrations of it-- like the one mentioned in the article. And, no, this does not require people to have control over the phase of everything involved.
- Shadowgamers, on 04/29/2008, -5/+12Codebreakers will still break it, even if it breaks the laws of physics V:
- karmabandit, on 04/29/2008, -0/+7It's unbreakable since you know with 100% certainty whether it has been eavesdropped or not. If it has been eavesdropped, don't use that key and send a new one! If it hasn't been eavesdropped, then you can send your message over normal unencrypted lines. (using the key)
- trogdor282, on 04/29/2008, -0/+6That's because people use "never" when they really mean "the mean time required to break this code is impractically long". But with QC, "never" means "the mean time required to break this code is infinity".
- NathanielJ, on 04/29/2008, -0/+5http://en.wikipedia.org/wiki/BB84
You can read about the Key Distribution Scheme yourself. What you described is completely impossible via quantum mechanics. When person A sends the key to person B, they are sending superpositions of states. No one but person A will EVER be able to find out exactly what those superpositions were; reading the data collapses the superposition into "basis" states (which are like 0 and 1). - Reaper2806, on 04/29/2008, -1/+6So if I follow this right it's about the uncertainty principle? That by viewing the code you change it, and therefore the codebreakers are at a loss?
- pinchduck, on 04/29/2008, -0/+5Bingo. Becoming enthralled with technological solutions and ignoring the human element will lead to a lot of confident cryptographers inside a leaky organization.
- neko6, on 04/29/2008, -0/+4Best comment in the thread, and I'm saying it as a graduate student in Cryptography.
RSA is effective. It hasn't been broken in 30 years. Its good enough for every real-life scenario, and its used by banks and armies. But it was still surpassed by human factors numerous times. The Enigma was also broken mostly due to operators' mistakes (using the same daily 3-letter code over a long period). - protogenxl, on 04/29/2008, -1/+5Well I will beat them with good old fashion One Time Pad
- Suricou, on 04/29/2008, -0/+3The abstract mathematics might be unbreakable, but there will certinly be a few implimentation flaws. Authentication comes to mind - not a lot of use in having a perfectly secure connection if you can't tell if the other end is your trusted business partner or Eve, who just cut the cable and jacked her own router in.
- diggerine, on 04/29/2008, -0/+3Or unless the key is not truly random, in which case the ciphertext will be vulnerable to cryptanalysis (see examples of one-time pad code-breaking: http://en.wikipedia.org/wiki/One-time_pad#Exploits ...
And like someone else said, no matter how secure the encryption method is, you can always try to obtain the plaintext or the key by force or guile. - interbeing, on 04/29/2008, -0/+3The first thing I thought when I read this was why don't we have voting machines with quantum encryption here in the US? It seems the Swiss are already using this technology while we can't even get a machine that will accurately print a piece of paper...
- fluxion, on 04/29/2008, -0/+3look up "quantum key distribution" on wikipedia, they have a straight-forward article. BB84 is actually pretty simple protocol. there's another protocol that uses quantum entanglement, which is probably a bit less intuitive.
i agree though, the article was lacking. i think they couldve easilly gotten the jist of the protocol across but instead they threw out stuff like HUP and "quantum wierdness". - mcla0181, on 04/29/2008, -0/+3There is a brnach of Q Crypto that will very likely replace all known cryptography, essentially render a key to be an infinitely replenishable one time pad. This is unbreakable, w/o/ the key.
And AES 128 is still "unbroken" - but there's better, stronger AES versions out there, like AES 256 or 512... - hadees, on 04/29/2008, -0/+3I found it kind of lacking. It doesn't really explain how you are going to be able to send this information from one person to the next. I mean I am sure the theory of Quantum Encryption is sound but if requires a total rework of our infrastructure to use it then no one, except in very special cases, is going to use it.
- ToadLeg, on 04/29/2008, -1/+3This can still be broken by quantum-photonic computing. Not taking the data and running it through a quantum or photonic computer, but running the actual photons through a quantum algorithm that keeps the "uncertainty" (polarization) intact while reading the data encoded in the photons. Basically, AND, OR, XOR etc... logic gates of a normal computer can be designed to work on photons with encoded data (a photonic computer) - and using weird quantum mechanics such as entanglement, the photon can be "copied" and calculations can be run in parallel so that the "guesses" of what the polarization of the photon is that are wrong fail, and the guesses that happen to be right yield something that is like an exact copy of the original photon along with a "measurement of the photon", which is actually what the guess was that yielded a result.
To put it an other way, you simply have to keep the polarization of the photon uncertain while performing a measurement which is also uncertain until it is finished. Note that this is not yet possible, but will be in the future. - jjesusfreak01, on 04/29/2008, -0/+2No one here seems to understand the concept of quantum cryptography (ill give it a go, as it was explained to me by some of the people working on it at Mitre)...
Quantum Cryptography is not a method of creating codes, but it is an integrated system capable of allowing completely unbreakable encrypted communication. It has always been possible since the dawn of written language to create an unbreakable cipher. Thats easy, you simply have a key equal to the length of the message. The question is, how do you get these keys to the people who need to decode your messages? The answer is, you use a system that allows you to send messages and know that so long as you are connected, the line is secure. It does this by sending photons with different spins or polarizations, and if the message is intercepted, both sides will know instantly. You dont have to worry about it being jacked though, as I highly doubt that it would be possible to "jack" these lines, as the equipment needed to do so would be prohibitively expensive...all you really need to do is secure the authentication with a pre shared key (a unique key for each session) and you are set.
There are two uses for this system. You can either
1) Simply use it to send encrypted or unencrypted messages over the quantum connection...encryption doesnt matter if it cant be intercepted
2) Use it to transmit encryption keys in a secure manner. After encryption keys have been transferred, then the medium by which you send the encrpyted messages is completely irrelevent (as you will be using a perfect cipher)...if you want, you can send a copy of every communication to the russians and osama bin laden, and it wont matter, because they wont be able to crack it. - mcla0181, on 04/29/2008, -0/+2peaky is right... the point of QC is onetime pad distribution...
- karmabandit, on 04/29/2008, -0/+2It's not really like that at all. You should read up on it from a better source than digg comments, but suffice it to say that not only will observing it change it, but you will know *who* changed it. They can't so much as look at a small piece of it without you knowing it's messed up. Wikipedia is a good place to start: http://en.wikipedia.org/wiki/Quantum_cryptography
- jeuhrn, on 04/29/2008, -0/+2Sounds highly uncertain!
- digg1520, on 04/29/2008, -1/+3Well, 128 Bit AES is still state of the art, and quantum cryptography will not replace classic cryptography, but I agree about the article.
- StealthTomato, on 04/29/2008, -1/+3I apologize; clearly I have received far less technical education than "u."
- bentrinh, on 04/30/2008, -0/+2TooManySecrets
- webcrumb, on 04/29/2008, -0/+2And pressing the letter L for an entire message.
- ZekeSulastin, on 04/29/2008, -0/+2Is your absolute foaming-at-the-mouth hatred for creationists that bad that you have to froth about them in a COMPLETELY UNRELATED ARTICLE?!
Save your vitriol for the proper context. - NathanielJ, on 04/29/2008, -0/+2***** - Of course computers have to change how they work for them to be able to perform this key distribution scheme, classical computers don't manipulate quantum states, and thus can not perform quantum key distribution. You would need a (drumroll, please) QUANTUM computer.
But that's beside the point anyways - early uses of quantum key distribution of course won't be taking place on quantum computers, but they will be over photon channels and the like. Quantum computers are still quite a ways off, while all the tools to perform quantum key distribution are ready. - Shadowgamers, on 04/29/2008, -1/+3Human stupidity, even physics can't cover for that :V
- diggerine, on 04/29/2008, -0/+2It wasn't "long before", just something like 3 years ahead. And because the British inventors were working for the UK's equivalent of the NSA, their invention was kept secret. Diffie-Hellman-Merkle and Rivest-Shamir-Adleman all developed their methods independently of the secret British breakthrough.
Second, regardless of that, 128-bit RSA is not "current state-of-the-art". Even RSA-200 (663-bit RSA) has been factored already (http://en.wikipedia.org/wiki/RSA_Factoring_Challen ... - JHW539, on 04/29/2008, -0/+2128 bit RSA is "state of the art"? Not really. Hell, just going to 256 bit gets you basically unbreakable and you could crank it up to 512 bits if you want. This is actually quite a problem for law enforcement going after kiddie porn a-holes - if they don't have the password, they can't crack the computer and dictionary attacks take forever.
And "the remaining stumbling block, the distribution of the code key to the right person" was solved quite a while ago. In fact, the name of the most commonly used approach is "public key" cryptography. There is absolutely no need to exchange secret keys due to the magic of factoring obscenely large numbers into obscenely large primes. Did you ever exchange a key with Amazon before passing out your credit card number? Wonder why?
I am not a cryptography expert, but I am married to one and have lost money at poker to ones better than she is and this article sounds like a reporter trying to write well beyond their level of comprehension. - Bluezdood, on 04/29/2008, -0/+2no more secrets...
- ToadLeg, on 04/29/2008, -0/+2Highly "theoretical" - it will almost certainly be possible in the near future.
;-) - Suricou, on 04/29/2008, -0/+2"running the actual photons through a quantum algorithm that keeps the "uncertainty" (polarization) intact while reading the data encoded in the photons"
Though I know little of the intricacies of quantum mechanics, isn't the whole point of quantum encryption that doing this would violate some fundamental law beyond my comprehension? You cannot read the encoded data without ruining that uncertinty. - webcrumb, on 04/29/2008, -0/+2You want to know who people have voted with the assurance of no tampering - with this system you know immediately if tampering has happened, therefore it is useful.
- strictnein, on 04/29/2008, -0/+2WTF is this V: and :V stuff?
- Suricou, on 04/29/2008, -0/+2I have an alternative means:
1. Threaten to label the reciever an unlawful enemy combatent and disappear him to prison without trial.
2. Get reciever to hand over message.
3. Disappear reciever to prison without trial. - geekchic, on 04/29/2008, -1/+3I'll try to ignore the subtle fact that it was Europeans who invented a working version of a "public/private key" system long before the founders of RSA dreamt of it.
- webcrumb, on 04/29/2008, -0/+2"their invention was kept secret"
Just think of what's being worked on now. ;) - Stevo23, on 04/29/2008, -0/+2Yes, I'd say you misread the article.
- Bakebehe, on 04/29/2008, -0/+2Dear god my brain
- grumpyrain, on 04/29/2008, -0/+1Or what they already know but would not dare tell the public.
- wiredDeath, on 04/29/2008, -0/+1A more in depth article:
http://www.aip.org/tip/INPHFA/vol-10/iss-6/p22.htm ... -
Show 51 - 100 of 120 discussions
© Digg Inc. 2009
— Content created and posted by Digg users is