telecom@ucbvax.ARPA (10/18/84)
From: Jon Solomon (the Moderator) <Telecom-Request@MIT-MC>
TELECOM Digest Thursday, 18 Oct 1984 Volume 4 : Issue 103
Today's Topics:
Re: TELECOM Digest V4 #102
Jove: joint networking proposal by IBM and British Telecom
Name and address service in Alabama
Net 1000 - AT&T's Answer To AUTODIN II?
----------------------------------------------------------------------
Date: Tue, 16 Oct 84 20:16:18 EDT
From: Robert Jesse <rnj@BRL-TGR.ARPA>
Subject: Re: TELECOM Digest V4 #102
"Please hang up and try again. If you need help, hang up and then
dial your operator."
A recorded message that interrupted a news conference
between astronauts in the space shuttle and reporters
on the ground.
-- U.S. News & World Report, 22 October 1984 "Current Quotes"
------------------------------
Date: Tue 16 Oct 84 22:32:06-CDT
From: Werner Uhrig <CMP.WERNER@UTEXAS-20.ARPA>
Subject: Jove: joint networking proposal by IBM and British Telecom
[ excerpts from an editorial in The Economist, Oct 13-19, 84, page 13
]
NO, BY JOVE
THE PROPOSED IBM AND BRITISH TELECOM JOINT VENTURE FOR A DATA
TRANSMISSION NETWORK IS ANTI-COMPETETIVE
... [such a request] lies in the British government's in-tray, in the
form of an arcane request for a telecommunications license. Britain's
trade secretary has to decide on a proposal by BT and IBM for a joint
venture, aptly if unwisely called Jove, to run a value-added network
(van) in Britain.
...
As British trade secretary, Mr Norman Tebbit should none the less
refuse to let BT and IBM do what they want: their proposal runs too
big a risk of interfering with the competitive free-for-all that is
Britain's, and Europe's, only real hope of getting back into the
technology game.
LET THE MARKET SET THE STANDARDS The hundred or so computer and
communications companies that have filed hostile comments about the
proposed venture make two main objections to it. The first, which
they wrongly rate more important, is that the van will run on a
communications standard called Systems Network Architecture [SNA]
which is owned by IBM. The opponents say that the use of IBM's
proprietary standard will give IBM a big advantage - not just in the
market for British vans, but also in the markets for the computers,
other machines, and other networks (such as local ones) which will
eventually hook up to the van. Jove's detractors want the British
government to back an alternative called Open Standards
Interconnections [OSI], which is being developed in international
committees and will be owned by nobody.
These arguments of Jove's opponents against SNA are hard to credit.
An industrial standard, wether proprietary or not, is best established
not by a committee but when enough customers buy it to make it a
standard. That process of standardisation by consumer choice should
not be interfered with. If SNA does become a network standard, IBM's
advantage, if any, will be slight. It will be stuck with the
standards as much as anybody; and even some of IBM's critics admit
that it will not require superhuman effort to build good links between
SNA and OSI ones.
The charge of anti-competitiveness, the second main objection to Jove,
is much more serious. Believers in competition need to be thoroughly
suspicious of a proposal by the world's (and Britain's) biggest
computer company and Britain's near-monopoly telecommunications
company to start holding hands. It is technically true, though only
just, that IBM and BT are in different businesses. But any IBM
executive who proposed such a deal with one of America's regional (and
still mostly monopoly) telephone operating companies would be laughed
out of IBM's boardroom, just before being laughed out of court.
The real problem is that the British government's too-cautious partial
deregulation of BT has left the company with powers that are
inherently anti-competitive in the context of the deal it wants to
make with IBM. It is not clear wether anybody else who wants to offer
a service like Jove would be allowed to provide it. Even if
competition in this kind of van market were open to all comers, the
overawing team of BT and IBM, backed up by BT's control of the
physical network, would all too likely be an effective deterrant.
IT IS THE SECONDARY COMPETITOR WHO MATTERS Worst of all, the deal
threatens to discourage secondary competitors - those who want to
provide not a van of their own, but equipment or services to ride on
the back of the IBM-BT van. These are exactly the people on whom a
revival of British electronics depends. Competing with BT should be
made as easy as possible for them. In America, anybody with a product
or service to latch on to the public network has the right to full
technical details about how the network operates. In Britain,
unhappily, even a privatised BT will be able to keep much of that
information secret.
The simplest course would be to say NO to the two companies. Until BT
proposed something grander, IBM was preparing a limited van of its own
- which could still offer BT some salutary competition. But an
outright rejection is not essential. It might be more helpful to let
the van run, but on conditions that forced BT and IBM to make
knowledge about the inner workings of the network available to
competitors. However it is done, the British government needs to make
sure that Jove has some company on Olympus.
[ either way, it seems to me as if IBM is ready to gobble up another
big market. while Europeans still huddle in their traditional, and
mostly government controlled committees, IBM is ready to do something.
It's amazing when you think about it, to see IBM leading the pack.
But European governments are simply too unwilling to give up control
over the communications industry, otherwise the market forces would
have achieved long ago what governments have successfully obstructed,
the merging of the European companies into fewer but more competitive
giants of international scale. The products would not necessarily be
better, but at least there would be products coming out at a faster
rate, possible even competing more successfully in the world markets.
Why are these developments important to us outside Britain? Well, for
one thing, we'll have to interface and live with whatever de-facto
interface the British communications scene presents to us, and
furthermore whatever alternative technologies are defeated by IBM
there, will be less likely to even be available as an option for us
here. Besides, when not depressing, it's often amusing, even
educational, to watch developments in Europe. - Werner ]
------------------------------
Date: 17-Oct-1984 1243
From: covert%castor.DEC@decwrl.ARPA (John Covert)
Subject: Name and address service in Alabama
South Central Bell has just announced that in conjunction with a
change in the pricing for D.A. service within Alabama (now 40 cents
per call with NO allowance, no exemption from hospitals and hotels,
and 25 cents from coin phones unless charged to a calling card, in
which case 40 cents applies) names, addresses, and ZIP codes can be
provided if a number is given.
Alabama used to refuse to give you the address, even if you needed it
to verify that you had been given the number for the right subscriber.
The bill insert claims that the service is only provided on Alabama
customers to callers within Alabama, but it seems to work from outside
the state as well (it should, since that costs 50 cents, except from
Canada).
There is a form with the bill insert so that name and address will not
be provided to callers who provide a telephone number.
------------------------------
Date: Wed, 17 Oct 84 12:37:06 EDT
From: dca-pgs <dca-pgs@DDN1.ARPA>
Subject: Net 1000 - AT&T's Answer To AUTODIN II?
I was reading in the 17 Oct issue of MIS Week that Net 1000 is in
heavy seas ("A Hole In Net 1000", Robert Feldman, p. 1). Apparently
Ford Motor Co., who was to have been a majotr customer, dropped Net
1000 last week in favor of Tymnet/Tymshare. Reason: Insufficient
nodes/access points into the network. That was a big part of why
AUTODIN II was cancelled; of course , the DoD's reqt for survivability
made that part of the argument even stronger. EDS Inc. appears to
have been successful with a strategy of highly centralized nodes, (e.
g., VIABLE), but VIABLE is less a distributed network than a number of
quasi-independent host servers which are interconnected by thin VG
pipes for non-real-time Q/R and nighttime file dumps. I would guess
that remote VIABLE terminals are heavily multidropped.
I guess NET 1000 and AUTODIN II could be termed how-to-do
packet-switching unprofitably, by overcentralizing, while EDS/VIABLE
could be called how-to-do "everything but packet-swirching"
profitably.
Best,
-Pat Sullivan
------------------------------
End of TELECOM Digest
*********************