In article <firstname.lastname@example.org>,
> Robert Bonomi wrote:
>> If the "Bell System/AT&T/Western Electric" had remained a monolithic
>> entity, The rate of change in the "Internet" would likely have been
>> much slower. There probably would not have been the telecom boom/bust
>> of circa 5 years ago,
> What is your basis to claim "the rate of change in the Internet
> would likely have been much slower"?
A knowledge of telecom history. :)
Bell system/AT&T/WEco was _very_ conservative in outlook. They didn't
deploy anything until it was thoroughly tested and determined to be
They were driven mostly by "what is best for the telephone company",
and only incidentally by 'what is best for the telephone customer".
Most "improvements" in customer capabilities came because the monolith
waqs -pushed- into it. And were implemented on a "gouge the end-user
for all it is worth" basis. Look at the deployment of 'touch tone'
dialing for one example. The primary advantage of touch-tone was to
the telco. it reduced the time that 'expensive' switching resources
were tied up for the handling of a given call; this allowed the telco
to reduce the _number_ of such expensive resources required to handle
a given volume of calls; thus reducing the telco's _cost_ for handling
that volume of calls. Yet the telco made the user _pay_extra_ to use
this 'save the telco money' feature. As a result, some FORTY YEARS LATER,
the telco _still_ has to support 'pulse' dialing. And the expenses of
*two* signal schemes for dialing.
Imagine what would have happened if the telco had offered a _discount_
(even a small one, say $0.50/month) for lines that were touch-tone
dialing *only*, from day 1. I suspect that 'pulse' dial would have
dropped out of the tariffs 20 years ago; because nobody was using it.
The Bell/AT&T/WEco record on 'data communications' is similarly
'lagging edge'. I've _never_ heard of a "Bell Standard" for speeds
above 1200 baud over voice-grade (aka dial-up) circuits. And that
specification was 'unusable' except for direct-connect hardware; If
you were stuck with 'acoustic coupler' interface requirements, The
best telco spec was "Bell 103", 450 baud or below, (an 'enhanced'
specification, added later, pushed the upper-limit to 600 baud --
wow!) Racal-Vadic, on the other hand, had an acoustic coupler capable
specification for 1200 baud that worked. And with far less of a
'noise' problem than "Bell 212". Don't forget U.S.Robotics, or Hayes,
who had 9600-baud over dial-up units. Or Trailblazer, that had one
capable of 19,200 baud.
Yet, the Bells only offered 'lease-line' service for 2400 and above.
either 4-wire lease-line for 202, 206, 208, 209, etc type modems, or
DDS. And, DDS was priced at "an arm and half a leg".
> It seems that many critics of the former Bell System "freeze it" at
> the time of divesture. That is, they presume the Bell System's
> physical plant and operating policies would never change and remain in
> 1983 technology. That premise is absurb.
> Throughout its history the Bell System was improving its plant. The
> system of 1983 was radically different than the system of 1973, and
> clearly the system of 1993 and 2003 would be radically different
> than 1983.
> I know that data communications improved greatly just during the late
> 1970s, for example. Digital lines replaced analog lines for faster
> speed and higher reliability. Private line costs were coming down.
> It is also clear operating policies and service plans would have
> changed, too. (They were always envolving in the past). How or what
> is tougher to say -- it depends on the external environment.
> Don't forget the Bell System was heavilly controlled by (1)
> regulation and (2) the consent decree. We know that deregulation
> became popular later on. It's possible the Bell System may have
> been allow to adjust its rates so the the profitable corridors (the
> "cream") may have gotten discounts so Bell could compete fairly
> against newcomers. It's possible the Bell System may have escaped
> the consent decree -- just as IBM was able to do -- and go into new
> markets previously closed to it.
> Who knows, perhaps LANS and WANS would've been bult FASTER had
> the Bell System been allowed to be involved and use it strengths.
Actual history argues against that 'unfounded speculation'.
Starting with their 'concept' of how high-speed/high-volume data
networking should be done (ATM). Which, because of the miniscule size
of the 'payload', has an _incredible_ layer of overhead; with all
sorts of 'per connection' overhead carried in _every_ packet --
primarily to ensure that every packet could have 'cost accounting'
done at _every_intermediate_ point.
And it disregards the existance of the various 'public packet-data
networks' (Tymenet, Telenet, Bitnet, Autonet, etc.) that were
_already_ in existence.