?

Log in

No account? Create an account

Previous Entry | Next Entry

Still sick...

...and I've got just two words for that. Code signing.

Seriously. Code signing.

Viruses work because our cells contain machinery which will read, accept, and translate any RNA strands they see into proteins. Any RNA strands they see. Including RNA strands injected into our cells from viruses, or RNA strands transcribed from DNA injected into our cells from viruses.

Which is, from a security standpoint, pretty fracking stupid.

Code signing, I'm telling you. If our genetic material were signed with some sort of unique code that means "yes, this really does come from us, it's safe to translate this RNA and build this protein," and the transcribing and translating machinery would refuse to process RNA that wasn't signed, then viruses could inject their bits into our cells from now 'til Doomsday and it wouldn't mean diddly.

Code signing. Just one more reason why if we were designed by some Grand Creator, he wasn't very good at his job.

Tags:



Comments

roadknight
Dec. 23rd, 2008 08:20 am (UTC)
I see what you're getting at, but the analogy really does break down at a number of places. First, I'm not sure that DNA/RNA can be shown to be Turing Complete. Second, it's a fairly well-known-and-popular fact that the DNA/RNA transcription mechanisms are quite prone to error by omission, substitution or repetition. How can you do any sort of signing of a key of any useful length when tolerance or acceptance of a key that is "mostly correct" is a design requirement of any signing system you develop. How do you do key revocation ?

tacit
Dec. 23rd, 2008 04:09 pm (UTC)
By making the key not code for production of anything. We've already got well-defined codons that mean START and STOP; add a bit of complexity and robustness to the START codons and set it up so that RNA without the proper START sequence doesn't get translated.