August 27th, 2013, 07:05 AM
Rune's deadbolt application
maybe this is not the right forum but someone may be able to help.
i dont seem to be able to contact the above company re their deadbolt product, anyone got any knwledge why?
it use the vernam otp cipher.
email doesnt work
the site has no download or any other email contact
they are based in singapore
could the website have been stripped for uk users ?
August 28th, 2013, 02:08 PM
From the USA, I see only postal addresses (one USA, one Singapore) and telephone numbers.
But don't worry -- when someone offers "Vernam OTP" encryption, you can be confident that either (a) they don't understand information security, or (b) they are fraudsters.
Vernam is so massively clumsy to use, that it almost never offers a security advantage in practice. Almost everybody who claims "Vernam OTP" is, in fact, selling a stream cipher. There are thousands of n00bs out there who think that stream ciphers are the same as Vernam, because they don't (or can't) understand elementary math. You can find examples of this ignorance in posts by crytpo-wannabes on this very forum!
Now, stream ciphers are not necessarily bad. Several famous ones have been broken, but there are also good strong stream ciphers. But they offer ABSOLUTELY NO SECURITY ADVANTAGE over what anyone can download for free (GNU privacy guard, or gpg).
Note well that you can only use Rune "OTP" to exchange confidential information with a person having a matching Rune mass storage device (that is, one that was purchased as part of the same set).
At a guess*, the Rune product embeds keys in its mass storage devices. If my guess is correct, then using their "OTP" mode:
(a) avoids standard methods of key exchange, which MIGHT someday become practical to break (for example, if the public keys are too short, and/or the quantum computer becomes a reality); in other words, it could be more computationally secure than standard cryptography
(b) should be proof against impersonation attacks such as "man in the middle"
(c) guarantees that if the embedded key is ever compromised, ALL of the "OTP" encrypted data ever exchanged using those devices is also compromised
(d) makes the security of the messages no better than Rune's protection of the keys they create (are you confident that the government of Singapore doesn't have a complete key list?)
(e) makes it possible for an adversary gaining custody of one of the devices (at any time -- for example, before you received it, or after you finish using it) to read ALL of the encrypted messages
(f) makes the encrypted messages indecipherable (to the customer) if the device is lost or destroyed
To sum up, the difficulty of reaching this company has just saved you thousands of dollars you might have wasted -- congratulations!
*I am speculating about the embedded keys, but if they haven't done that, then the mass storage devices they sell are preloaded with "random" data (hopefully, properly generated), which would limit the lifetime total of securely exchanged messages to the size of the drive. Note that in this case, there is nothing that would preclude Rune (or some government) from keeping a library of all of the "random" data sets.
August 28th, 2013, 03:36 PM
yes I appreciate what you say.
I just found out the cost........RIDICULOUS
actually, another application they have called locksmith generates the random pads for the deadbolt app to work.
yes they would have to be paired up to message each other with the same pads.
locksmith also accepts random data from external devices like id quantiques trng.
anyway, way out of my price range, so I saved my money
August 28th, 2013, 06:13 PM
If they sell an "app" for "generating random pads," that is in itself another cause for suspicion (not unlike selling a perpetual motion machine).
If the resulting pads really ARE random, then it is necessary to _securely_ send them from the party who generated them to the other side of the communication link ... but if you have such a secure channel, then you don't need encryption at all! Or to put it another way, the security of your "encrypted" traffic will be no better than it would be if no encryption were used.
If there's a way to get the "random pad" across to the other side WITHOUT sending the entire thing (which of course, if seen by an adversary, completely breaks the security of communication), then the "pad" is ABSOLUTELY NOT RANDOM, and the security is not equivalent to Vernam.