Re: setenv

[Richard Stallman]

      (4) getenv and setenv both work on process-environment as they do
          now, but before returning, ensure that Emacs' actual environment
          reflects the content of this variable.  This will allow the
          Elisp user to modify Emacs' own actual environment, won't
          require any special knowledge of the system, and won't break any
          existing code that may modify process-environment directly.

Option 3 works completely when users modify process-environment
directly.  Option 4 does not.  So if option 3 is not too terribly
hard, it is better.  I think it is not too terribly hard.

To tell people that modifying process-environment directly is not
reliable isn't totally out of the question, but it is ugly.  If it
is the only way that works, we can do it.  But it is worth a small
amount of work to avoid the ugliness of this.

    I don't know that forking is really the issue.  More, it's the
    behavior of libraries linked into Emacs.  Unless the Emacs developers
    want to keep super-careful watch over any possible points when any of
    these could look at the environment, for all possible versions of
    these libraries, it will almost certainly create a problem.

If we needed to be absolutely certain these problems would never
happen, we would have to be extremely careful.  But this kind of
problem is not terribly serious even when it bites, and it is
very unlikely to bite also.

So I think it is fine to fix them when and as someone notices them.
That won't be much work--I expect less than one per year.