Holger Schauer <schauer(a)coling.uni-freiburg.de> writes:
Well, if you just consider the Elisp currently at the Lisp archive
or the ones shipping with XEmacs, anything might be fine. But what
about the guys who have written their own private `foo'-package ?
Force a recompile and have them fix their surprisingly-now-buggy
code which has been legal elisp for years ?
All changes lead to such results. Why are you so surprised at this?
Hrvoje> Why do you call non-sense anything that is not strictly
ANSI
Hrvoje> conformant?
If you could tell me where the sense is of using some language (or
even worse: an implementation) not standarized when standarized
versions exist ... ?
When using a subset of a standardized language, you can still write
conformant programs. The reason to use the subset instead of the
unabridged experience is, of course, efficiency.
Haha. I was just thinking of the re-occuring threads in comp.emacs.*
but certainly won't want that for myself. But if the engine would be
able to support language X, and there is somebody who wants to
implement and use X, hey, who am I stopping him from doing so ?
*I* will stop him, personally if it comes to that. :-)
Hrvoje> I did, but you weren't listening.
Nope, you were not. Do you want to replace Emacs Lisp with (a
substrate of a) Common Lisp (_implementation_) or do you want to two
concurrent lisp engines ? AFAIU, you want the former, but I'm by no
means sure.
So you were listening after all. :-) Yes, I would like to see the
former.
--
Hrvoje Niksic <hniksic(a)srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
I'm a Lisp variable -- bind me!