Re: Encyrption back door?
The maths say that you can break it with enough power and time.
A meaningless statement, in practice.
First, of course, "it" hasn't been defined. RSA? ADH? ECC? Some other key agreement protocol? AES? Some other symmetric cipher? Or is this just hand-waving?
Second, once you assume unbounded resources, the question is no longer interesting. If you have a decision procedure for determining what the correct plaintext is, you can just try every possible key, or even every possible plaintext, "with enough power and time".
Third, it's quite easy to scale cryptographic algorithms up to the point where there aren't enough resources in the visible universe to brute-force them using a conventional computer. It's quite easy to do that for symmetric algorithms and hashing even with general quantum computing. It's a bit harder to do that for asymmetric crypto (key agreement and signatures), but we have candidates with strong evidence for being secure under GQC.
It's vanishingly unlikely that any correctly-implemented, well-studied, modern cryptography was broken in this case. Any of the mooted alternatives -- bad implementation, false implementation (the "it was a trap" theory), insider compromise -- are all much, much more probable.
Years ago, Bruce Schneier famously claimed that cryptography was good enough, and that "if you think your problem is cryptography, you don't understand cryptography and you don't understand your problem". Since then there have been successful attacks on widely-deployed cryptographic algorithms (MD5, SHA1, RC4) and protocols (all SSL/TLS versions prior to TLSv1.2, pretty much anything using CBC and not making a special effort to mitigate padding oracles, etc.). And we have the perennial worry that maybe someone will get feasible large-scale GQC working and so we need post-quantum asymmetric cryptography. But Schneier's basic point was right: implementations and people are the big threats to communication and data security, not the underlying cryptography.