On Fri, 26 Feb 2016 09:20:58 +0100
FRIGN <dev_AT_frign.de> wrote:
> On Fri, 26 Feb 2016 09:11:20 +0100
> Mattias Andrée <maandree_AT_kth.se> wrote:
>
> Hey Mattias,
>
> > Well, that is pain in the ass.
>
> I know Matlab is a pain in the ass, but it's going to be
> academia mostly who would be "eligible" to use factor(1)
> for something.
> What I could live with is having a naive implementation,
> so John Doe can just type "factor 1831" quickly and see
> the result.
> If you want to crack large primes, use something more
> tailored for the job.
>
> We had the same mindset "implementing" sort -m, which
> merges already sorted files without storing them.
> We just sort them anyway, which kind of defeats the
> purpose of the flag, but such a simple solution is valid
> for 99% of the cases and we still have the chance to do
> it "right" in a simple way.
>
> > I'm actually using factor. And it is in base systems, so
> > I think it should be included, but I will be simplifying
> > it.
>
> What are you using it for? To test the primes in your
> favourite RSA-algorithm?
Mostly random things, but regularly when I correct maths
tests.
> I will personally not sign any patch that includes gmp,
> tommath or pthreads or any other abominations of mankind.
> I don't think either that anybody here would agree on
> merging a single tool literally nobody uses if it
> includes these insane dependencies.
> I know there is "need" for a suckless bignum library,
> also in the interest of implementing dc(1) or bc(1) in
> the future (which are programs people actually use),
> however, this has not been achieved yet.
>
> Cheers
>
> FRIGN
>
Received on Fri Feb 26 2016 - 09:37:12 CET