RFS vs. NFS

[Brandon Allbery]

As quoted from <1053 at mcgill-vision.UUCP> by mouse at mcgill-vision.UUCP (der Mouse):
+---------------
| In article <10219 at steinmetz.steinmetz.ge.com>, stpeters at dawn.steinmetz (Dick St.Peters) writes:
| > UNIX file system semantics.  However, if program foo physically
| > resides on a remote machine, typing foo causes foo to run on the
| > remote machine.
| 
| This is `transparent' if, and only if, foo would also run on the remote
| machine after copying it to some local filesystem.  Otherwise, well,
| 
| > Even under RFS, exec'ing a VAX executeable on a Sun doesn't work very
| > well, so it can matter whether the file is local or remote.  RFS thus
| 
| It doesn't matter whether the file is local or remote; it matters
| what's in it!  If you copy the VAX executable to the Sun's local disk,
+---------------

I dunno about you, but if I've got my home directory on some machine "A"
mounted on machine "B" (with a different hardware processor), I do *not*
consider it "transparent" if I can't run a binary in my bin directory on "A"
from "B" without invoking rsh!  (I encounter this using Worknet between
8086 and 80286 machines.)

Perhaps what you describe is "transparent" under some definition, but to Joe
End-User what you describe is a discontinuity:  the file is "on both systems"
but only useable on one of them.  As far as the end user is concerned, the
network isn't transparent, it just reached out and punched him in the nose.
-- 
	      Brandon S. Allbery, moderator of comp.sources.misc
	{well!hoptoad,uunet!marque,cbosgd,sun!mandrill}!ncoast!allbery
Delphi: ALLBERY						     MCI Mail: BALLBERY