Found at: gopher.quux.org:70/Archives/usenet-a-news/FA.poli-sci/81.08.25_ucbvax.2809_fa.poli-sci.txt

Tue Aug 25 21:10:52 1981
TELECOM Digest V1 #1
>From JSol@RUTGERS Mon Aug 24 22:25:21 1981
TELECOM AM Digest 	Tuesday, 24 Aug 1981       Volume 1 : Issue 1
Today's Topics:     Administrivia - Welcome Aboard
	 	   USRNET - Alternative to A. T. & T.
	       Problems with Dimension - One Persons Views
Date: 24 Aug 1981 0118-EDT
From: the Moderator <JSOL AT RUTGERS>
Subject: Administrivia
Welcome to TELECOM. This digest is a spinoff from the HUMAN-NETS
discussion on the telephone network and switching equipment.  Parts of
this digest are in fact submissions to HUMAN-NETS which were never
published, and are presented here to spark the discussion.
The archive for this is in the usual place, DUFFEY;_DATA_ TELCOM at
MIT-AI, and we will shortly be adding to the archive the discussions
that have taken place in HUMAN-NETS relating to telecommunications.
I will be moderating this list from Rutgers, as I do with POLI-SCI,
but you can still send mail to TELECOM@MIT-AI, or TELECOM@RUTGERS.
If you want to communicate with the maintainers then you should
Date: 12 Aug 1981 1257-PDT
From: Daul at OFFICE  
Subject: USRNET
I was wondering what kind of information is sitting in the
readership's minds regarding the proposed Shell's inter-company
Usrnet.  This is the alternative to the high cost of AT&T's services.
Any information is welcomed.
		-Bill  (DAUL@OFFICE)
Date: 21 Aug 1981 1109-PDT
From: Jwagner at OFFICE  
Subject: One writer's problems with Dimension
This editorial recently appeared in the San Jose Mercury, the local
newspaper for San Jose CA and vicinity.  It discusses one writer's
problems with a new in-house telephone system.  I'm sending it along
without comment with the hope that it will spark a flame or two.
          -- Jim Wagner/jwagner@office
GIVE HIM A RING -- IF YOU DARE  (c) 1981 San Jose Mercury
by John Askins, editorial writer
   Remember how it started out?
   The machines were supposed to be the servants and we were supposed
to be the masters?
   Right now I figure I'm working for machines about half the time.
The computer I type on decides that we're going to take a little
break, and we take a little break.  Never mind if I want to or not.
Never mind if it's DEADLINE or not.
   When the remaining humans here want to tell me something quickly,
they don't come running over to my desk.  They flash a message across
my screen.  Sure I know it's them, but it feels like it's the computer
getting all excited over something.  It's unnevering.
   I'm not complaining, it's just that slowly and surely I'm losing my
sense of, well, human dignity.
   Just this week we've been getting used to a new telephone system
called Dimension which is supposed to make our jobs easier.  They
always say that, don't they?  My telephone now does a lot of tricks.
"Dimension is a system that listens to what you tell it and then talks
back you you," said the Pacific Telephone representative sweetly.
Just what I wanted, a phone that talks back to me.
   It has a lot of abilities, I'll give them that.  All things that
humans used to do, using English.  It can forward all my calls to
another number, or it can forward them only while I'm busy on another
call.  It can let me know I have a call waiting, and how important the
caller is.  It can put me in line for a WATS line and then call me
back when the line is free.
   In other words, it's another goddam computer.  You talk to it in
simple number codes and it responds.
   It not only does what you tell it, but it has a limited vocabulary
of its own, a sort of dial-tone Morse code that you're supposed to
learn to interpret into English.  Three short beeps means "Action
accepted, proceed."  Siren intercept, which sounds like a cop car
responding to a riot call, lets you know you made a mistake.  Three
shorts and a long, followed by a dial tone, indicates that you didn't
depress your switchhook properly.
   Yes, it's come to this.  If I depress my "switchhook" -- the button
under the receiver that I used to cal "the button" -- too quickly or
slowly, the telephone is not happy.  It is a sensitive "instrument,"
as the call it now, and I've got to learn new habits to avoid
offending it.
   Pardon me, telephone sir.  I don't want to be a bother or anything,
but I'd like to make a call.  If it wouldn't be too much trouble.
   At the training session, the nice woman (yes, they're still using
people for some jobs at Bell) got up and explained to us that the days
of just casually picking up the phone and using it, without paying too
much attention to what you're doing, are OVER.  For one thing, you
have to make sure to hold the receiver real close to your ear, so you
won't miss anything the telephone says.
   This training session was marvelous.  Can you imagine spending an
hour and a half to learn how to use the telephone?  They had a color
movie, a lecture, a practice session, everything but a quiz.  And
believe me, with Dimension you NEED it.  A few days later they
installed the system.  Monday morning the phone rang in its new,
authoritative way.  Ring, ring.  Ring, ring.  I stared at it in
horror, a jumble of codes flashing through my head.  Finally I got up
the nerve to lift the receiver expecting to hear a warbling siren
   There was nobody there.  It had just been testing me.
   Later I tried to hang up by pressing the, let's see, switchhook,
and got three longs and a short.  "Scotch and soda!" I said quickly.
"Scotch and SODA, goddammit!"  Didn't do a bit of good.  I finally
replaced the receiver of my instrument and waited five minutes or so
to get over being mad.  Sure enough, there was a dial tone again.  At
least it doesn't hold a grudge.
   Well, so, big deal.  New technology comes along and tells us we
have to change our way of doing things.  This is nothing new in the
workplace.  It's just that --
   The typewriter, OK, I remember how reporters complained when
electric typewriters replaced manuals, and how we complained again
when video display terminals replaced typewriters.  But at least these
machines have a little class, a little allure.  But a stupid
telephone?  Hell, I remember when their dials took forever to spin
back around.  I remember when all they were were devices for letting
people talk to each other without being in the same room.
   Those were the good old days.  Last week.
Gumby@MIT-AI 08/04/81 19:24:28
After all that phone discussion, can anyone dissect the digits
in an international direct-dialing sequence? I just called my
parents direct (in Australia) and it was 14 digits including
7 of the australian number and a # at the end. What was the
# (pound sign) for?
JSOL@Rutgers 08/24/81 00:49:00
You can look in your phone book for the explanation of international
calling. If not there, the local phone store distributes free "calling
guides" for international calls. Perhaps someone has a synopsis handy
(Lauren?) and can provide a brief explanation, if you tell us the
phone number you called then perhaps we can break it down.
Telephone numbers in the U. S. have an Area Code (3 digits), and phone
number (7 digits).  Sometimes you have to dial a "1" or something to
get outside your "local calling area", but the switching system need
only count the digits and then when you get all of them start
completing them. International calls aren't so lucky. The number of
digits can vary quite a bit, so the switching equipment times out if
you don't dial any more digits and then tries to process the call with
the digits it has. This timeout takes a few seconds. Touch Tone(tm)
users with the # key can hit that at the end of the sequence to force
it to complete the call, saving time.
Date: 28 Jul 1981 1358-PDT
From: Daul at OFFICE  
Subject: Phone Company Deregulation
What are the details (general or specific) on the phone company
de-regulation? What are the rumors about MacDonald's wanting to
buy in?
Date: 28 May 1981 17:52:28-PDT
From: ihnss!hobs at Berkeley
Cc: ihnss!hobs@Berkeley, ihnss!wlh@Berkeley
Subject: ESS
I want to correct some erroneous statements that I made about the
ESS machines.  (I bet Lauren beats me to it.)
#1/#1A is a large local switch (100K lines or so), #2/#2B is a medium
local switch (30K lines), and #3 is a small local switch (12K lines?).
#4 is a TOLL switch, it handles no lines but up to 107K trunks.  Both
#4 and #5 use a digital network to perform the actual switching, the
others use an analog network. #5 will handle up to 100K lines by
adding capacity on a modular basis.  The first #5 is currently
scheduled to go into service late this year.
                                May you be spared egregious errors,
[This discussion of ESS telephone switching picks up from HUMAN-NETS
Digest V3 #108. Shortly a transcript of the entire discussion should
be made available in the archive at MIT-AI. -JSOL]
Date:  28 May 1981 14:54 edt
From:  JSLove at MIT-Multics (J. Spencer Love)
Subject:  #5 ESS
Sender:  JSLove.PDO at MIT-Multics
cc:  JSLove.PDO at MIT-Multics
A recent message to Human-Nets contained a slight inaccuracy on this
subject: The #2 ESS is used in rural applications, and the #4 ESS is
used in toll switching only (not in rural offices as stated).  The
characteristics of ESS machines that I can remember easily are:
#1: A sort of glorified crossbar machine that uses banks of reed
relays (called ferreed switches) of some sort arranged in the same way
that the old crossbar switches were used.  It also uses some fancy
magnetic technology switch that are polled in a manner reminiscent of
core memory to detect on/off hookness and pulse dialing.  It is
controlled by a multiply redundant (2.5x) 40-bit computer running out
of a vast quantity of magnetic(?) read only memory.  In short, it
appears not to have used ANY off-the-shelf computer technology at all.
#2: A smaller machine that uses time-domain multiplexing to switch
conversations around withing itself, using an 8KHz sampling rate to
provide the 3.6KHz bandwidth that TPC generally delivers.  With a
capacity of about 2000 subscribers (as opposed to 40,000 for #1) it is
generally intended for rural or other small-capacity application.
Much of the hardware of the #2 has also appeared in the automatic
intercept machines: the ones that deliver you individualized recorded
sorrow when you reach a nonexistent number.  I think the computer is a
more relatively conventional 16-bit architecture.
#4: This machine handles 4-wire interconnections.  That is, completely
separate pathways for the two directions of a conversation.  The #2 is
internally 4-wire, but externally 2-wire, and the #1, when used in
regular central offices at least, is 2-wire.  There were some crossbar
machines that were built 4-wire, so there may have been some #1's
built that way for similar applications.  2-wire technology is fine
for voice grade, short haul interconnections, and is commonly used in
intra-city communications, but problems with echoing caused by
impedance mismatches cause 4-wire technology to be used in most
long-distance network equipment.  Unfortunately, echo-suppressors are
still needed in many cases because the ends of the conversations are
on 2-wire equipment with sloppy impedance matching.  The #4 is also an
all-digital switch.  All kinds of signal processing problems plague
designers of analog telephone switching equipment, and their less than
perfect handling of these problems is the reason that many long
distance connections are so bad.  By first digitizing the signals,
parts of the long distance network preserve the signals from any
further distortion until they reemerge into the analog world at the
other end of the conversation.  When the #4 ESS was designed, this was
too expensive for central office use, but looked very good for the
long distance network.
There may be minor inaccuracies in this presentation since it has been
years since I payed serious attention to these issues.  I don't know
what the #3 ESS is (if any), and I can't remember any other convincing
applications for 4-wire technology in the scale appropriate to a telco
central office.
Based on the above, I would guess that the #5 ESS is an all digital
(and hence internally "4-wire") switch with a modular design and
sufficient "distributed intelligence" that it can fit equally well
into the small (#2), midrange (#1) or gigantic (#4) offices.  When
connected to actual subscriber loops (wires leading to telephones), it
probably uses CODEC's (like MODEM: modulator/demodulator, except
encoder/decoder), which have become cheap enough that having one for
every phone line is now feasible.  You can buy them as single
integrated circuits, these days.
This has the additional implication that four wire circuits to
subscribers will become quite cheap in such exchanges (although it may
be years before the rates reflect this).  This has great implications
for data communications, since the CODEC for a given line might be
replaced with a pair of short-hail MODEMSs, at a great saving in
bandwidth and an eventual reduction in data-communication rates.
End of TELECOM Digest
 gopher://quux.org/ conversion by John Goerzen <jgoerzen@complete.org>
 of http://communication.ucsd.edu/A-News/
This Usenet Oldnews Archive
article may be copied and distributed freely, provided:
1. There is no money collected for the text(s) of the articles.
2. The following notice remains appended to each copy:
The Usenet Oldnews Archive: Compilation Copyright (C) 1981, 1996 
 Bruce Jones, Henry Spencer, David Wiseman.