Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!mips!spool.mu.edu!munnari.oz.au!manuel!cmf851
From: cmf851@anu.oz.au (Albert Langer)
Newsgroups: comp.society.development
Subject: Re: Computers and Telephones
Message-ID: <1991Jun2.052408.21005@newshost.anu.edu.au>
Date: 2 Jun 91 05:24:08 GMT
References: <1991May28.204751.11309@news.larc.nasa.gov> <3506@laura.UUCP> <1991Jun1.124059.8093@hq.demos.su>
Sender: news@newshost.anu.edu.au
Organization: Computer Services Centre, Australian National University, Canberra, Australia.
Lines: 53

In article <1991Jun1.124059.8093@hq.demos.su> avg@hq.demos.su 
(Vadim Antonov) writes:

>Our humble experience of running the biggest public computer network
>in Soviet Union shows that there is a lot of possibilites to use
>old technology to carry digital data. For example we're running stable
>9600 bps links over copper wires leading via our phone exchange which
>was installed in 1929 (Telebit's PEP modems, of course). V.22bis works
>reasonably well over noisy lines but V.32 sucks.
>
>I've got an impression that reasonable UUCP-based network could be
>established everywhere in the world. Our current average delivery delays is
>about four-five hours; we're working on reducing it by replacing
>backbone links with 9600 bps IP links over IP lines.

Thus poor phone networks are not a major issue.

>Anyway the main problem we got is the lack of qualified and initiative
>people to establish regional nodes and keep them operating. 

But skilled labor for sysadmin or sysop work IS a major issue.

These are the points I tried to make before, but nobody bit.

If Vadim's experience and my theorizing about this are correct,
shouldn't we focus on how to reduce the labor requirements for
store and forward email network management?

It is already commonplace for packet network nodes to be unattended
pieces of equipment like any other telephone exchange, with remote
supervision through the network itself and personnel on site only
when required to replace a circuit board etc. Not long ago it used
to be normal for even ordinary telephone exchanges to have permanent
on site technicians. Corporate LANs are being designed for remote
network management. Surely it must be feasible to design store and
forward email networks for remote network management as well. The
building blocks are already available, even including Management
Information Bases designed to allow remote monitoring of such
paramaters as available disk space and page buffers and mail
routing as well as the usual LAN paramaters for packet throughput
and packet routing etc.

Why on earth should a skilled sysadmin be required at every node?

Once this problem is resolved, massive expansion both in developed
countries and developing countries becomes quite straight forward.
The equipment itself is already very cheap and all costs except
network management are declining rapidly.


--
Opinions disclaimed (Authoritative answer from opinion server)
Header reply address wrong. Use cmf851@csc2.anu.edu.au