[Baypiggies] Eric Raymond likes Python

Damon McCormick damonmc at gmail.com
Mon Sep 14 10:36:58 CEST 2009

I only know of a few impressive industrial uses of OCaml.  However, those
are in exactly the sort of places where OCaml really shines, e.g. theorem
proving, program verification, and program transformation.
OCaml was used to do the C code generation for FFTW, an outstanding piece of
engineering which is pretty much the only cross-platform, high-performance
Fast Fourier Transform library that matters (it's essentially as fast as the
fastest hand-tuned libraries on all platforms).  AirBus used Ocaml for a
program that verified the fly-by-wire software used in the A340 and A380--it
proved the absence of runtime errors like memory violations and overflows.
 Microsoft used OCaml to implement a utility that developers used to detect
rule violations in Windows kernel-mode drivers.   IBM used OCaml to convert
between DB2 and Oracle in the IBM Migration Toolkit.

You could write this stuff in Python, but the idea of doing it in OCaml is
that with some cleverness, you can leverage the state-of-the-art program
transformation and theorem-proving facilities that exist in the language
interpreter to do the heavy lifting, rather than implementing those
algorithms yourself.

But for the 100 - epsilon percent of code that is not a theorem proving /
program manipulation problem, I would (and do) choose Python, Java, or C++
over OCaml.


On Sun, Sep 13, 2009 at 3:45 PM, Stephen McInerney
<spmcinerney at hotmail.com>wrote:

> > > A fairer comparison might be comparing OCaml to Psycho.
> > >
> > > Also, can Python do things that OCaml can't?
> >
> > LOL. Have you ever heard of "Turing-completeness", or "Church's Thesis"?
> Edward,
> Yeah I know all about the theoretical stuff;
> I was looking for a commercial answer, aka "does anyone actually seriously
> use OCAML and for what?".
> As far as I know noone much uses OCAML much, so a historical rant about
> Python 1.x from
> an OCAML enthusiast doesn't carry much weight.
> The only criticisms he made about OO back in Python 1.x, I was slightly
> curious whether there was
> more or less to OCAML than meets the eye.
> (By analogy, I looked into XSLT 18mths ago and that was overstyled
> nonsense, as in,
> I could write clean functionally-equivalent Python with >10x fewer lines
> and much more clarity.)
> Regards,
> Stephen
> ------------------------------
> Bing brings you health info from trusted sources. Try it now!<http://www.bing.com/search?q=pet+allergy&form=MHEINA&publ=WLHMTAG&crea=TXT_MHEINA_Health_Health_PetAllergy_1x1>
> _______________________________________________
> Baypiggies mailing list
> Baypiggies at python.org
> To change your subscription options or unsubscribe:
> http://mail.python.org/mailman/listinfo/baypiggies
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/baypiggies/attachments/20090914/39966c33/attachment.htm>

More information about the Baypiggies mailing list