NOS is evil

["Brandon S. Allbery" <bsa@linux.kf8nh.ampr.org>]

In your message of Thu, 08 Dec 1994 19:29:22 CST, you write:
+---------------
| jmorriso@bogomips.ee.ubc.ca said:
| > The mind boggles. I get a laugh when I see people using whatever hack
| > NOS has for SMTP, when they have the far more powerful Berkely
| > sendmail or Smail sitting under their noses. <Sigh> I guess old
| > habits die hard. Time to do some re-education...
| 
| On the other hand there are many valid reasons to use a user process
| for amateur radio network interface, all have been covered here before.
+------------->8

I intended to send to the group the first time, but forgot to change the 
address...

The *biggest* plus of [JT]NOS for most users is that they can move 
step-wise, learning as they go.  I haven't played with TNOS, but I've done 
everything I could to keep JNOS/Linux compatible with the DOS 1.09 
distribution (and still want to move it to 1.08df; before someone wonders 
about the versioning, Doug Crompton released 1.08df (1.08d fixed), then 
Johan rereleased it as 1.09.  I don't plan to go to 1.10, but Doug's current 
1.08dff shouldn't be too much different from what I already have).

A DOS JNOS user can move their existing system, virtually unmodified, to 
JNOS/Linux.  From there, he can establish a link between JNOS and Linux's 
native networking, then begin moving services from JNOS to Linux.  
Ultimately, all that will be left in JNOS is the AX.25 protocol; the final 
step, of course, being the move to kernel AX.25.

While Linux's (or, for that matter, BSD's) native networking is 
superficially similar to NOS, in fact configuring it --- and all the 
services --- is an order of magnitude more complicated.  JNOS (and, 
presumably, TNOS) give the NOS user a chance to make the move in stages, 
learning how to configure Linux services and switching them over *one at a 
time* instead of having to learn and configure *everything* before moving.

(I haven't checked recently, but at one point JNOS's SLIP was 25-30% faster 
than native Linux SLIP, both tested on the same serial port and modem to the 
same destination.  I assume this has been fixed...)

++Brandon
-- 
Brandon S. Allbery KF8NH	 [44.70.248.67]		bsa@kf8nh.wariat.org
Linux development:  iBCS2, JNOS, MH					 ~\U
Controlling application developers is like herding cats. --Oracle DBA Manual