One point that feels mysterious when you are new to these is the
storing. Is this the same for all different mail agents in Emacs?
The Emacs Lisp values should all be stored by Emacs the same way.
Emacs is a virtual lisp machine. Everything that is in an Emacs Lisp
expression is interpreted by the virtual machine. Programs that are
not part of the virtual machine are often operating system dependent
and their executables are architecture dependent.
Thus, the variable
rmail-delete-after-output
has a value that is kept by the Emacs virtual machine.
Regarding your second question,
Another mysterious point is the settings for different mail
agents. The setting above seems to suggest that this if for RMAIL
only?
Any variable can be used anywhere in Emacs. They are global. But the
naming convention suggests they are for rmail only.
It certainly makes sense that rmail-pop-password be advertized to
other email clients by changing the name. We think of `rmail' as
standing only for `read mail' but it could also stand for `retrieve
mail' ... That alternative expansion suggests that we use the
`retrieve mail' variables and functions in various programs.
In any event, nowadays, I think your point should be well taken. The
variable should be called mail-retrieve-remote-password (or something
like that) rather than rmail-remote-password as is currently
suggested. Moreover, all the movemail code should be moved to a new
Emacs Lisp library called mail-retrieve-remotely.el (or something like
that).
For the past 20 or 30 years, rmail has been the default mail reader
for Emacs. Since its beginning, Emacs has gained many email readers.
For example MHE was originally written in 1983 for Gosling Emacs and
ported to GNU Emacs shortly after. (From this perpective, as a mail
reader, gnus is recent.)
My thought is that the current names and lack of a separate library
are a consequence of Emacs' history. (I am not sure of this; I have
not paid attention to rmail.) RMAIL variables are like laws and
social customs that may once have returned value greater than losses
resulting from their suppression of cooperation but do not do so now.
However, Emacs variables are global and voluntary so they do less
damage to cooperation as they go obsolete than laws, since laws are
involuntary.
--
Robert J. Chassell
[EMAIL PROTECTED] GnuPG Key ID: 004B4AC8
http://www.rattlesnake.com http://www.teak.cc
_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel
