At 09:58 AM 4/19/2002 +0200, Xavier Leroy wrote:
>I don't doubt that, but my experience (as a long-time Unix user who
>installed quite a bit of garbage of the kind you mention) is that
>installing Unix software on Windows machines, even with Cygwin, is
>often surprisingly hard.

True, though Cygwin has improved astonishingly over the last five
years.  One problem with the "new" Cygwin is that it comes in many,
many pieces, so when someone says "Cygwin" you cannot now be sure
what exactly they have installed.

However, back to the subject.  For proselytizing purposes,
you want to avoid bad surprises for new users.  In this case,
I think that means that you should remove any Unix-only demos
from the introductory examples, and put them someplace with a
clear label that says "Unix only".  Unless the benefits from 
impressing the Unix users outweigh the negatives of confusing
the Windows users, such demos should go.

The other reason to avoid Unix-only demos is that, if you wish
to make ocaml more likely to be accepted by new users, then it
doesn't hurt to be "cross platform", which means that support for
Windows is about as good as support for Unix (which ought to
now include MacOS X).  Not only should this be true, it should
also be easy to see this truth from the outside.  This is purely
for marketing purposes, understand, but I am much more likely
to want to write something new in OCaml if I think I'll be able
to run it on any box I am likely to use (meaning, Windows, Linux,
and MacOS).

David

-------------------
To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr
Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/
Beginner's list: http://groups.yahoo.com/group/ocaml_beginners