Re: Emacs Lisp's future

[David Kastrup]

Taylan Ulrich Bayirli/Kammer <address@hidden> writes:

> David Kastrup <address@hidden> writes:
>
>> I think you are misunderstanding the problem.  The problem is not that
>> technical problems occured after large changes.  The problem is that
>> GUILE developers cannot be bothered with the fallout in affected
>> projects.
>
> I don't mean to insult but I have the impression that, what the Guile
> developers could not be bothered with might have been the LilyPond
> maintainer rather than LilyPond itself.

Shrug.  Naturally, after Andy Wingo prohibited me from bothering the
Guile developer list any more, the number of contacts went down.  I sent
several major developers who had _after_ the interdictum offered to help
the respective information about the state of the GUILE 2.0 port and how
to check out and work with LilyPond.  None of them ever got back to me.

At any rate, it has been mentioned previously in this discussion that
the Emacs developer list is not always considered a happy-go-lucky
environment.  So if it is the habit of GUILE developers to take out
revenge on a project and its users for enmity with single developers,
that's also relevant for making crucial GNU software depend on GUILE.

I think this interpretation of events is not making for a much better
outlook.  Particularly because Emacs development is more often than
other GNU projects governed by unpopular political decisions.  A habit
of retaliation and "see-where-this-will-get-you" in order to pressure
for a change in project lead is not likely going to work out well.

I don't think that you are better off selling this situation as a
personal vendetta, and it is not like the GUILE 2 problem was not
already there when I started to get involved with LilyPond.

>> GUILE development will cater better for Elisp than for Scheme?  That
>> does sound peculiar.  Is this pitch being made anywhere else apart
>> from the Emacs developer list?
>
> Guile has authority over Guile-Scheme, but not over Elisp; it has to
> and will support Elisp as defined by Emacs as much as possible.
> That's pretty obvious I'd say.

"This is not possible" will be defined under the constraints of GUILE
remaining Scheme according to GUILE's vision of interpretating the
Scheme standard and its further evolution.

> If the Emacs/Guile merge became fully complete in several years and
> all remnants of non-Guile Emacs disappeared, then it could be vaguely
> imaginable that Guile took control over Elisp semantics, but that
> "ain't gonna happen" under such Emacs community pressure. :-) You can
> bet on Guile taking Elisp only to where Elispers want it to go.

The question now is not where GUILE will take Elisp but rather whether
its interpretation of Elisp can get close enough to make a switch
feasible in the first place.

-- 
David Kastrup