Evaluating future technology

[ciphergoth]

How do you tell a reasonable guess about what future technology might bring (eg a manned mission to Mars) from unreasonable guesses (eg teleportation)?

I'm inclined to think that you have to get down to the technical nitty-gritty. If you don't know the field, it might be reasonable to think that in the future we'll prove that our ciphers are unbreakable. Actually, for everyday useful ciphers, a proof that they are secure with no unproven assumptions is much harder than you might think if you've not studied CS. There's no reason I'd expect you to know that if you're not a computer scientist, but if your vision of the near future included provably unbreakable ciphers, I'd want to explain why that doesn't look very likely at the moment.

What do you think?

Comments

[josington.livejournal.com]

I'm having to explain why provably unbreakable ciphers are unlikely, including a discussion of the Diffie-Hellman assumption, for my final-year project at the moment.

[ciphergoth.livejournal.com]

Communications of the ACM edition 09/2009 is all about the state of the P vs NP problem, which might be relevant. Of course even if we could solve P vs NP, that still only applies to worst-case problems, and we need average-case for cryptography.

The Wikipedia entry on "Results about difficulty of proof" is also interesting.