On Sun, 27 Apr 2008, SL Baur wrote:
On 4/27/08, Tim Connors <tconnors(a)astro.swin.edu.au> wrote:
> On Sun, 27 Apr 2008, Mike Kupfer wrote:
> It's more that I'm being overly complicated and want some way of forcing
> the xemacs process on one machine (machine 1) to use one filesystem,
> except for when it is invoked by the gnuclient process on another machine
> (machine 2; I don't want to just run the xemacs process on machine 2,
> because machine 2 is maintained by central ITS who are useless
> morons^W^Wrun several year old unmaintained versions of linux). Since a
> wrapper around gnuclient pointing back to machine 1 is my $EDITOR of
> choice on machine 2, I can't just put a symlink in for ~/thesis on machine
> 1, because sometimes I really do want to edit other files on machine 2.
>
> Probably more of an ill thought-out idea than something worth persuing.
> Never mind me :)
No, it actually bears some thought now. I face similar Linux braindamage
Perhaps if gnuclient could accept a flag to tell it to tell emacs to
operate on a file on the remote's filesystem, with the requisite
bidirectional communications this would require (yay! Yet another
protocol!), and the lack of an ability to use ange-ftp and tramp syntax
(since the user is explicitly asking for a file on the remote's local
filesystem, I see no harm in barring them from using tramp/ange-ftp
syntax).
issues at work. We could always assume in the past that a Linux user
was
free to change/fix his environment and that isn't true any more.
It's always been the case :) I had a 3 meg quota on $HOME, and 3Meg in
/tmp in undergrad CS in 2000. Almost the third freaking millenium, and
still only 3Meg!
--
Tim Connors
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta