Linux crypto killer apllication

[Robert J. Hansen]

David Picón Álvarez wrote:
> Well, I'm pretty sure if GnuPG had the limit you suggest (2048) it
> would be legally unusable for some purposes, due to legal guidelines,
> "best practices", and all that tosh.

Can you name some?  I'd love to know them.

> I don't know of any of these policies that require keys longer than
> 4096, but I wouldn't discard the possibility, certainly not in the
> future.

If 2kbit RSA/DSA/ElG ever becomes attackable either via cryptanalysis,
brute force or developments in large number theory, the solution will be
to move to entirely new algorithm families, not to just tack on another
few bits to the end.