Prev: Error in Lapack la_dgesvd.f90 (Was: who maintains lapack95?)
Next: Text or pictures? (was: non portable - a problem with direct access)
From: Nick Maclaren on 7 Aug 2010 09:47
In article <i3furd$r3k$1(a)news.eternal-september.org>,
e p chandler <epc8(a)juno.com> wrote:
>A lisp interpreter. [smile] I don't know about using that language for
>numerical calculations. When I first read about it, it was slow, partly due
>to its highly recursive nature.
No, sorry. Interpretive nature. Algol 60, Egtran, BCPL and others
proved that recursion often IMPROVES efficiency, and rarely reduces
it by a significant amount.
>LISP is about as far from Fortran as you can get.
Try Prolog :-) There are others even further away, but I can't
imagine anyone sane writing an ODE solver in them.
From: jfh on 8 Aug 2010 19:30
On Aug 8, 1:47 am, n...(a)gosset.csi.cam.ac.uk (Nick Maclaren) wrote:
> In article <i3furd$r3...(a)news.eternal-september.org>,
> e p chandler <e...(a)juno.com> wrote:
> >LISP is about as far from Fortran as you can get.
> Try Prolog :-) There are others even further away, but I can't
> imagine anyone sane writing an ODE solver in them.
A friend at Nick's university said to me "Real programmers do their
numerical analysis in LaTeX and their word-processing in Fortran."
-- John Harper