On Tue, Apr 10, 2001 at 02:23:03PM +0200, Akim Demaille wrote:
: So? Let's give a try to Python? Move to Perl 5.5? Other?
I suddenly got the urge to do some ranting :-}
The situation today is that perl has the upper hand and Guile is more or
less just another embeddable scheme interpreter. When/if the real Emacs+
Guile project emerges, knock on wood, Guile will suddenly be more widespread
than Emacs (some vi-users use Guile too ;), basically causing Guile to
become one of the most widespread script interpreters around. More wide-
spread than perl is maybe wishful thinking, but it's not totally unlikely.
In this situation, people would if they knew what's best for them ;)
rather want to be programming Guile than Perl, basically because Emacs
will be *the* dynamic development platform for Guile-based environments.
Anyone who's done some dynamic programming on running environments know
they don't want to go back to the slow turn-around time associated with
traditional software development...
[If this seems like an emacs-rant, you should know that I am primarily a
vi-user, and a power-user at that. Morten can verify this.]
As for rewriting Automake in Guile, how realistic is it that this will
happen any time soon, let's say the next couple of years? There is a
strong point in having the same foundation for both Automake and Autoconf
- having installed one, the requirements for installing and using the
other should be fulfilled. I therefore feel that it's basically Perl
and Guile (if Automake is heading that way) that are the options.
Python is also a strong contender (I hear people say it's a nice language,
but come on; programming without braces?! ;) but I'd discourage going
there if Automake is headed another way.
Lars J
