Re: Patents on RSA will expire soon....

> The algorithm that factored RSA129 takes about exp(sqrt((log n)(log log
> n))) steps.
> Indeed 10^17 instructions is just about how much work was required to
> factor RSA129--.
> That formula gives about 10^29 for a 1024 bit number. If computers double
> in speed every 18 months then they will be only 32 times as fast when the
> patents expire.

If that rate of speedup held long term, and no significantly better
factoring algorithms showed up, it would be 60 years until
a 1024 bit number were as easy to factor as RSA129 was this year.

-fnerd
- - - - - - - - - - - - - - -
and i dreamed i was flying
high up above my eyes could clearly see
the statue of liberty
sailing away to sea        --Paul Simon
-----BEGIN PGP SIGNATURE-----
Version: 2.3a

aKxB8nktcBAeQHabQP/d7yhWgpGZBIoIqII8cY9nG55HYHgvt3niQCVAgUBLMs3K
ui6XaCZmKH68fOWYYySKAzPkXyfYKnOlzsIjp2tPEot1Q5A3/n54PBKrUDN9tHVz
3Ch466q9EKUuDulTU6OLsilzmRvQJn0EJhzd4pht6hSnC1R3seYNhUYhoJViCcCG
sRjLQs4iVVM=
=9wqs
-----END PGP SIGNATURE-----

FutureNerd Steve Witham says:
> If that rate of speedup held long term, and no significantly better
> factoring algorithms showed up, it would be 60 years until
> a 1024 bit number were as easy to factor as RSA129 was this year.

That cuts it rather close for some applications. Consider that a
constant factor of a few thousand is easy if a really good new
factoring algorithm shows up. If you are concerned that no one be able
to read your messages for the next twenty years, you have trouble.

Perry