🚀 go-pugleaf

boucher@csl.sri.com (Peter K. Boucher) writes:

>cbbrowne@csi.uottawa.ca (Christopher Browne) writes:

*** Stuff deleted ***

>So, 512-bit RSA keys aren't looking so great anymore.  1024-bit
>keys seem barely adequate in this light.  2K- to 3K-bit RSA keys
>seem prudent to protect any data that you are seriously 
>concerned might be attacked by a big-budget ($25G) enemy.

Despite what Peter said above, $25G programs are not easy to start
or stop, witness the AIDS research effort.  That effort is still
ramping up and, despite the great political pressure, already has
more money appropriated than can be wisely spent.  That situation
is much like the defense buildup in the early 80's.  More dollars
than sense.  A group with $25G to spend would have the same problem,
years and years before a solution would be available.  With that
level of funding, mathematicians would have to be trained and
algorithms developed.  There would also need to be a period for
ramping down of the effort and the money to pay for it.  The
result of the effort would likely eliminate RSA (or DES or whatever)
as an encruption technique. In all of its years of existance, it is
unlikely that NSA has spent more than $50G or $60G on this problem,
though that is only a guess on my part.  They have got lots of other
irons in the fire, and so do the French, and so does the medical
community.
Also, regarding the French Intelligence Service, my understanding from
newspapers is that if they come across something in the pursuit of their
mission that would be useful to their industrial service, they will
share the information with them.  Industrial intelligence is not a
primary mission.