Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!ALLIANT.MCS.ANL.GOV!catlin
From: catlin@ALLIANT.MCS.ANL.GOV (Bob Catlin)
Newsgroups: gnu.emacs.bug
Subject: (none)
Message-ID: <9001082351.AA01082@alliant.mcs.anl.gov>
Date: 8 Jan 90 23:51:58 GMT
Sender: daemon@tut.cis.ohio-state.edu
Distribution: gnu
Organization: GNUs Not Usenet
Lines: 53


Documentation of emacs-18 (GNU Emacs Manual, Sixth Edition, Version 18; p.20)
states:

  "To kill Emacs, type C-x C-c (save-buffers-kill-emacs)...Unless a numeric 
   argument is used, this command first offers to save any modified buffers."

I guess the key here is "modified" buffers.  It is my experience, with 
emacs-18.55 on our Solbourne (Sun-4 clone) as well in emacs-18.50 on the 
Alliant here at Argonne National Labs, that if one creates new buffers, none 
of those are saved unless explicitly written by "C-x C-w".  For example:

	csh%  emacs <RET>

	*scratch*> <C-x b>

	minibuffer> Switch to buffer: (default *scratch*) a.a
	-----

        a.a>  This is buffer a.a <C-x b>

	minibuffer> Switch to buffer: (default *scratch*) b.b
	-----

        b.b>  This is buffer b.b <C-x b>

	minibuffer> Switch to buffer: (default b.b) c.c
	----
        
	c.c>  This is buffer c.c <C-x C-w>

	minibuffer> write to file ~/whereverIam/c.c
	----


This editing session will save buffer c.c in file c.c and discard buffers a.a
and b.b without prompting the user for reconfirmation of the kill and/or 
whether or not the user would like to save the contents of the other buffers
currently opened.  This seems to me to be an inconsistency.  Granted there are
ways around this problem ( invoking emacs seperately with a file name for each
new file desired; using C-x C-w on each buffer before C-x C-c), but it would
be better, in my opinion (which is only an opinion), that consistency between
the operations for creating files and modifying old ones is preferable.

Thanks for your attention to this matter.  Please let me know if there is 
anything I've omitted, overlooked, etc., or if you just think I'm full
of GNU-goo.  I am a fan of yours and hope that little suggestions like this
are helpful in producing the best software available anywhere. Thanks.

						Bob Catlin
						Alphatech, Inc.
						(617) 273-3388 x236
						catlin@alliant.mcs.anl.gov