Path: utzoo!mnetor!uunet!husc6!cmcl2!brl-adm!umd5!ames!elroy!mahendo!jplgodo!wlbr!scgvaxd!trwrb!aero!venera.isi.edu!cracraft
From: cracraft@venera.isi.edu (Stuart Cracraft)
Newsgroups: comp.protocols.appletalk
Subject: Thoughts on AppleTalk
Message-ID: <4360@venera.isi.edu>
Date: 17 Dec 87 17:22:28 GMT
Organization: USC-Information Sciences Institute
Lines: 31

As far as we can tell, the AppleTalk network is one of the more difficult
and frustrating networks to maintain. So that others may benefit from
our experience, we are here summarizing our major problems with the net:

  1) a lack of good, solid connectors means that every week or two,
     some portion of the network 'breaks' and we end up with several
     nets.
  2) as far as we can tell, print software version 4.0 does not like
     to run on Mac II's, which mostly likes version 5.0. This means that
     in our mixed network, different versions of the print software
     end up having to co-exist and subsequently a new print seems to
     require a lengthy printer down-load half the time.
  3) adding new hosts is bulky, time-consuming and difficult. Cutting
     a cable in half and then attaching new three-prong end-plugs to
     each of the new ends is a lengthy, stripping-and-soldering process.
  4) we are not aware of any good network diagnostic software that we
     ourselves can get ahold of, as much as we would like to have some.
     rumors abound of a network diagnostic tool that verifies all 
     possible 'pairs' of inter-Mac communication, but we haven't found
     a supplier yet.
  5) the serial nature of the network is somewhat cumbersome. two wires
     usually need to be run to each Mac unless it is the end of the net
     (rare), so the cable-duplication and general cable-confusion is
     high.

Remarkably, the only thing we presently use our network for is to ensure
that each Mac or Mac II can print to one of two LaserWriter's on the net.
It should be a simple, stable configuration, but we have found this not
to be the case.

	Stuart