As has already been said earlier in this thread, this makes no sense. The individual or company providing cryptographic services would not have the client’s key anyway. I am quite sure whatever the bill might have said it did not say what you say it said.
Not necessarily. It is entirely possible that NSA was shuffling the S-box values around so as not to leave any backdoors that might be created by IBM.
Not for 128-bit keys.
But since DES uses 56-bit keys exclusively, 128-bit keys are irrelevant.
from http://www.wikipedia.org/wiki/DES:
The original research announcement by Eli Biham and Adi Shamir.
More information about DES
Information about the EFF’s custom-built, $250,000 DES cracker, which completed the DES Challenge II in under 3 days.
More info on DES from IBM
Most real-world implementations of DES use triple-DES, which is an encrypt-decrypt-encrypt scheme using three keys and an effective key length of 112 bits. There are several variations on how to implement triple-DES/3DES/DES3, but I believe the encrypt/decrypt/encrypt with three different keys is the FIPS standard.
Sure, but if you’re going to do that, why not just use IDEA or AES? they’re both more secure and much faster.
Of course, DES isn’t the only way to go.
Yeah, RC4 is a symmetric cipher, so you can’t do certain things with it. (It can’t be used as the basis for a PGP implementation, for example.) But it does provide good security as long as you keep the key secret. What’s more, the algorithm is simple enough people have memorized it. It can be shrunk into a few lines of <insert favorite non-toy language here> and used as a sig, thereby exporting it as a rider on Usenet and messageboard posts. (As has been mentioned, this is no longer a legal issue. But it was when these sigs were created.)
The site I linked to is an activist page that advocates even casual coders create their own RC4 implementations in whatever language they choose. (The site mentions a 16-line QBasic implementation.) Their viewpoint is that strong encryption is under attack by the governments of the world and the only way to reverse this is to make at least one algorithm so widespread it can’t be outlawed effectively.
DES is a US government standard, which meant it was the required algorithm in many government uses. In many cases, using IDEA or Blowfish or any other algorithm was not an option because the product had to conform to certain specs. There was a long period of time in which triple-DES was also an approved standard but AES didn’t even exist yet.
Of course, all of this is OT as far as the OP is concerned…
BTW, the mathematical basis of the PGP algorithm is trivial to describe to any mathematician in about 30 seconds and was known to Euler 300 years ago. It is slow, however, and I believe that PGP uses it only to encrypt the key for a different algorithm. But if you really cared you could use it for the whole message and anyone could write the program. My real belief is that the reason the US govt restricted the export was that they hoped they would thereby discourage Americans from using it.
Yes, PGP generates a random key and uses that key to encrypt the message and it then encrypts that key with the key or keys of the recipients and sends it all along. This has several advantages. One is that encryption is much faster and also that you can send the message to several recipients as you only encode the (short) key several times and not the entire message.
Yes, that’s IDEA. PGP uses the public key to encrypt a randomly generated session key for IDEA, and uses that key to encrypt your file.
The proposed law that starfish mentioned in the RIP bill which stands for Regulation of Investigatory Powers. If I remember correctly what it boils down to is if the police have obtained your encyrpted e-mail (for example) via a legal wire-tap, then you are required to hand over the key used to encrypt it. Failure to do so can result in a fine and/or prison time.
I believe you can choose IDEA or other algorithm in your setup.
Wouldn’t that be the key to decrypt it? I know nothing about this law but common sense tells me the police would need a judicial warrant? In any case, now it begins to make more sense. Once you are being investigated and the police have a warrant then you are obligated to cooperate.
I believe we had a thread about this discussing whether that could be protected as self-incrimination. We went into all sorts of analogies and I do not remember if we reached any conclusion. Someone might want to do a search.
Dunno. See, I haven’t bothered to upgrade PGPi for awhile, so maybe that feature is available in the newer versions.
It depends on whether the cipher is a single-key one or a public key one.
Oh come off it. The police want the key to decrypt the message. In one case it happens to be the same as the key used to encrypt and in the other one it is different. So what? The police want to decrypt. That’s what they want and that’s what they want the key for. They want the decrypting key and they are not much concerned with whether the flavor is chocolate or strawberry.
Regarding PGP I am using V5.5 and it allows CAST, triple DES and IDEA. It will decode any of them and you can choose which one to use by default when you encrypt. I am using CAST and I imagine that’s the factory default. You can see this in PGP Preferences / advanced.
Of course, if anyone can ever get Quantum Crytography to work, then the whole question of government interference goes away, as it is completely unbreakable.
We have discussed this in past threads and i still believe there is much hype and exageration and confusion. until someone can better explain it to me I believe so-called “quantum cryptography” is not cryptography at all but a means of building a secure transmission channel which is a completely different thing.