Re: Telephone Exchange Usage in Low-Volume States
John McHarry
> In many places in the U.S. the demand for telephone exchanges is very
> high for a variety reasons. This has result in area code splits and
> overlays. NJ started off with one area code and now has nine.
> But some states still only have one area code. I understand some
> states are not growing very fast in population, indeed, some rural
> towns are losing population. This includes: Alaska, Idaho, Montanna,
> North Dakota, South Dakota, and Wyoming. (Not counting some other
> single-code states).
> Given the rural/low growth aspect of places in some of these states, I
> was wondering if telephone service may still have some old fashioned
> features to it. For example, would such areas have:
> 1) Traditional party line service, since it's not worth the cost to
> upgrade lines out to people's farms?
In almost all places single party lines have been cheaper for the
telco for decades. Where they have been forced to continue to offer
party line service, they have used bridges at the CO, not in the
field. The forcing comes when they go to the PUC and tell them party
lines cost more, and the PUC replies go to all single party lines at
the party line rate. I think they have mostly managed to buy off the
PUCs.
> 2) Five digit dialing in some areas not well populated or served by
> community dial offices?
Five digit dialing was a feature of old panel switches which could be
set up to throw out the first two digits if the local exchange was
being dialed, and use them otherwise. My old hometown had one of those
for a long time.
> I believe everything is ESS nowadays, but that pays for itself by
> eliminating the need for technicians to visit remote unattended
> switches. Probably some community dial offices have been converted to
> concentrators or feeders to a larger CO elsewhere.
There are certainly CDOs still around in the boondocks. Many of those
old brick windowless boxes have a pad out back with a fiberglass hut
mounted on it. Likely that contains a DMS-10 or DCO. Those are small
ESS systems that can serve several thousand lines.
hanc...@bbs.cpcn.com
> ... I believe private line service is pretty
> much available in all but the most remote areas now.
Could you elaborate on the situation in "the most remote areas"? Has
it been cost effective to replace an isolated long loop shared party
line with more modern carrier equipment?
I know that in developed areas party lines are obsolete, in some cases
grandfathered in, in some cases not available at all. But I was
wondering about very isolated rural areas.
>> 2) Five digit dialing in some areas not well populated or served by
>> community dial offices?
> be dialed with all seven numbers with the first few digits "absorbed"
> for local callers. When ESS came into being that all ended.
As long as the dialing is unique, there is no reason that an ESS
couldn't absorb digits just like an SxS could.
In many places the demand for exchanges is so high that the only way
to create unique dialing is require TEN digits. But in the states I
mentioned perhaps there is enough 'space' in the exchange assignments
that five digits could still be unique for a town.
In any event, I was also wondering what kind of exchange demand there
was in the states I mentioned with limited population growth and rural
decline.
Robert Bonomi wrote:
>> 2) Five digit dialing in some areas not well populated or served by
>> community dial offices?
> *VERY* rare. Gotta have full numbers, to handle direct-dial inbound
> calls from outside that exchange. Recognizing 'short cut' dialing
> within the exchange raises all sorts of complexities, having to do
> with 'variable length' numbers, and detecting 'end of dialing'.
By 1965 (probably earlier), almost all '5 digit' exchanges had full 10
digit numbers to handle inward calls. During the 1950s and early
1960s the Bell System was changing exchanges where necessary to make
them unique within an area code.
None the less, the 5-digit dialing capability remained for people
within those small towns. Letterheads carried phone numbers like
this: (505) 34 5-4111. Outsiders would dial 505-345-4111, but those
within the town would need only dial 5-4111. This was easily
accomodated by SxS.
One reason this was possible was because rural areas often had very
small local calling areas, so local dial choices were quite limited.
Anything else required the toll prefix which forwarded the call to a
toll center (or later a SxS add-on memory register).
I should note I had 7 digit dialing to the area code nearby (and they
did to me). This was possible because our exchanges were unique to
BOTH area codes. In other words, if I were 555, there'd be no 555 in
the other area code, so there was no confusion. Obviously when
exchanges became scarce this was abandoned. First I had to dial all
ten digits to cross the area code, now I have dial 10 digits to call
next door. This was common in area code border sections. (To this
day such calls remain local even though they cross the LATA).
As mentioned, it was easy for SxS to handle this, and certainly could
for ESS if conditions and policy permitted it.
Neal McLain
> In many places in the U.S. the demand for telephone exchanges is very
> high for a variety reasons. This has result in area code splits and
> overlays. NJ started off with one area code and now has nine.
> But some states still only have one area code. I understand some
> states are not growing very fast in population, indeed, some rural
> towns are losing population. This includes: Alaska, Idaho,
> Montanna, North Dakota, South Dakota, and Wyoming. (Not counting
> some other single-code states).....
I assume that by "exchange," Lisa is referring to an NXX code.
> Given the rural/low growth aspect of places in some of these states,
> I was wondering if telephone service may still have some old
> fashioned features to it. For example, would such areas have:
> 2) Five digit dialing in some areas not well populated or served by
> community dial offices?
No.
The misconception here is that, after the introduction of DDD, telcos
retained 4- or 5-digit local dialing as a convenience to users. Telcos
retained 4- and 5-digit numbering only as a temporary stopgap measure
during the transition from SxS to crossbar or ESS.
In order to accommodate inbound DDD, it was essential that every
number have a 7-digit format. But SxS switches couldn't accommodate
7-digit dialing, so telcos faked 7-digit numbers by prepending dummy
digits. Local calls continued to be dialable with only four or five
digits; however, if a local caller actually dialed all seven digits,
the prepended digits were absorbed by "absorbing selectors" --
i.e. ignored.
This situation led to numerous conflicts between local numbers
(dialable as 4 or 5 digits) and non-toll calls to nearby communities
(dialable as 7 digits). To avoid such conflicts, telcos had to devise
special dialing plans. AT&T documents dating from 1975 describe such
dialing plans in detail ["Typical Trunking Diagrams for Step-by-Step
Offices." "Notes on Distance Dialing." AT&T Engineering and Network
Services Department, Systems Planning Section, 1975. Appendix A,
Section 4].
The legacy of these old dialing plans can still be seen today in the
numbering assignments of many outlying communities surrounding
mid-sized cities, even though the old SxS switches have long since
been replaced with ESS. In these communities, most/all of the NXX
codes assigned to the outlying communities start with a numeral that
was not used in the central city in the SxS days. Two examples that
come to mind:
-- In Ann Arbor Michigan, local numbers were a combination of
four- and five-digit numbers, all served from the same
central office:
2-XXXX Huron office
3-XXXX Huron office
4NX-XXXX Outlying communities
5-XXXX Huron office
6XXX Huron office
7XXX Huron office
8XXX Huron office
9XXX Huron office
The 4NX codes are still in use today, even though one
of them (South Lyon 437) is now in a different area
code (248).
-- In Madison Wisconsin, all local numbers were five digit,
served from four central offices:
2-XXXX Pflaum office
3-XXXX Sylvan office
4-XXXX Kedzie office
5-XXXX Main office
6-XXXX Main office
7-XXXX Main office
8NX-XXXX Outlying communities
9-XXXX Kedzie office
The 8NX codes are still in use today.
A third example -- Centerville, Iowa -- was cited by Mark Roberts in a
posting here on TD a couple years ago. As I noted at the time, even
though Centerville's old SxS switch retained 5-digit dialing, new NNX
codes, even within Centerville itself, would require 7-digit dialing.
http://tinyurl.com/8axyn
Neal McLain
bv...@aol.com
> In order to accommodate inbound DDD, it was essential that every
> number have a 7-digit format. But SxS switches couldn't accommodate
> 7-digit dialing, so telcos faked 7-digit numbers by prepending dummy
> digits. Local calls continued to be dialable with only four or
> digits, the prepended digits were absorbed by "absorbing selectors" --
> i.e. ignored.
in school.
Carbondale, IL, (Jackson County) 1971 General Telephone
618-453 - So. Il. Univ., Carbondale. IL
618-457 - Carbondale, IL
618-549 - Carbondale, IL
618-867 - De Soto, IL
618-684 - Murphysboro, IL
618-687 - Murphysboro, IL
From/to any Carbondale NXX (1, 2, or 3): 5-digits allowed, 7-digits
supported (618-453 required a ?9? to dial out from the university, but
5-digits allowed within the university PBX/Centrex/whatever)
From Carbondale NXX (1, 2, 3) to De Soto (4): 7-digits required
From Carbondale NXX (1, 2, 3) to Murphysboro, (5, 6): 7-digits required.
From Murphysboro or De Soto to Carbondale, 7-digits required.
(I believe that locally, only 5-digits were required in Murphysboro
and only 4-digits in De Soto.)
Outside of these 3 exchanges, but within the 618 NPA: 1+7-digits
required Outside the 618 NPA: 1+NPA+7-digits required
Carbondale had DDD in the 60's. It did use '150' instead of just '1'
as a toll alert and the operator would come on the line and ask 'Your
Number Please?'
John Levine
> it been cost effective to replace an isolated long loop shared party
> line with more modern carrier equipment?
Yes. My relatives' telco in Vermont has some really long loops out to
remote farms, and they're all private lines. The maintenance is a lot
easier, as is the toll billing. They have a Paradyne DSL system that
works on long loops and they told me about one farmer who wanted DSL
so they took him off the SLC which didn't support DSL and gave him an
18k ft home run so the 60 hz hum on voice calls was deafening but the
DSL works fine.
> In many places the demand for exchanges is so high that the only way
> to create unique dialing is require TEN digits. But in the states I
> mentioned perhaps there is enough 'space' in the exchange assignments
> that five digits could still be unique for a town.
Sure, but for policy reasons dialing is now all 7D or 10D or 1+10D.
There is exactly one prefix in my town, and we tell each other our
phone numbers with four digits, but the dialing is 7D nonetheless.
It's 7D within the area code, which may be local, intralata toll or
interlata toll, 1+10 to other area codes.
R's,
John
Neal McLain
> In order to accommodate inbound DDD, it was essential that every
> number have a 7-digit format. But SxS switches couldn't
> accommodate 7-digit dialing, so telcos faked 7-digit numbers by
> prepending dummy digits. Local calls continued to be dialable
> with only four or five digits; however, if a local caller
> actually dialed all seven digits, the prepended digits were
> absorbed by "absorbing selectors" -- i.e. ignored.
Whereupon bv...@aol.com responded:
> I don't understand. Below is the local dialing plan we had when
> I was in school.
> Carbondale, IL, (Jackson County) 1971 General Telephone
> 618-453 - So. Il. Univ., Carbondale. IL
> 618-457 - Carbondale, IL
> 618-549 - Carbondale, IL
> 618-867 - De Soto, IL
> 618-684 - Murphysboro, IL
> 618-687 - Murphysboro, IL
> From/to any Carbondale NXX (1, 2, or 3): 5-digits allowed,
> 7-digits supported (618-453 required a ?9? to dial out from the
> university, but 5-digits allowed within the university
> PBX/Centrex/whatever)
> From Carbondale NXX (1, 2, 3) to De Soto (4): 7-digits required
> From Carbondale NXX (1, 2, 3) to Murphysboro, (5, 6): 7-digits
> required.
> From Murphysboro or De Soto to Carbondale, 7-digits required.
> (I believe that locally, only 5-digits were required in
> Murphysboro and only 4-digits in De Soto.)
My guess: the GTE Carbondale office used type "AR" (absorb repeatedly)
selectors on the 4th and 5th levels. Local (within Carbondale) calls
were dialable with only five digits, but if a local caller dialed all
seven digits, the prepended digits (4 and/or 5) were absorbed and
ignored. Nearby communities (7-digit dialable) were segregated on
separate levels (6th and 8th).
Which means that you could have dialed any combination of 4s and 5s
before dialing anything else, with no effect on the end result. For
example, you could have dialed 444554444555544443-XXXX to reach SIU.
With this in mind, the Carbondale dialing plan would have been:
----- ---------- ------------ ----------------------
LEVEL LOCAL INBOUND DDD RESULT
----- ---------- ------------ ----------------------
1 11X Vertical service codes
2 Blank level ?
3 3-XXXX 618-453-XXXX SIU., Carbondale. IL
4 absorbed
5 absorbed
6 68X-XXXX 618-68X-XXXX Murphysboro, IL
7 7-XXXX 618-457-XXXX Carbondale, IL
8 867-XXXX 618-867-XXXX De Soto, IL
9 9-XXXX 618-549-XXXX Carbondale, IL
0 0 0 Operator
----- ---------- ------------ ----------------------
If my guess is correct, the 2nd level would have been blank (perhaps an
error tone). Did you ever dial it to see what happened?
bv...@aol.com continued:
> Outside of these 3 exchanges, but within the 618 NPA:
> 1+7-digits required Outside the 618 NPA: 1+NPA+7-digits
> required
I suspect it's now:
Within the 618 NPA: 1+618+7-digits
Outside the 618 NPA: 1+NPA+7-digits
Unless the ICC has adopted the New York/California plan in 618, in which
case it's now:
Within the 618 NPA: 7-digits
Outside the 618 NPA: 1+NPA+7-digits
In this same thread, hanc...@bbs.cpcn.com noted:
> As long as the dialing is unique, there is no reason that an ESS
> couldn't absorb digits just like an SxS could.
True. But it would have been enormously complicated, if not impossible,
to add more NNX prefixes as the community grew. A case in point: the
Carbondale office described above. (Note that I use NNX here, rather
than NXX, because most of the crossbar/ESS conversions occurred long
before 01/01/95.)
I realize that Illinois isn't one of the "low-volume states" that you
mentioned in your posting. But the problems associated with five-digit
dialing are universal, and occur in all states.
In 1971, there was only one unused level in Carbondale (2nd), so only
one new NNX would have been possible: NN2. The obvious choices would
have been 442, 452, 542, or 552 (to maintain the look and feel of the
existing NNXs); however, it appears that all four were already in use
NPA 618. So GTE would have had to find something on the 6th or 8th
levels (as it happens, 618-862 is still available today).
In any case, NN2 would have been the last non-conflicting NNX available
in Carbondale.
Now imagine yourself trying to play this game in every small community
across southern Illinois (or any other area code). In every case, you'd
have to pick a new NNX that:
- Wasn't already in use elsewhere in the area code.
- Didn't conflict with the local dialing plan.
- Didn't conflict with the dialing plan in any nearby community.
And you probably wouldn't even have Lotus 1-2-3 to help you do it!
Footnote: new prefixes in Carbondale today include 319, 351, 503, and
529. SIU's centrex has added 536.
Neal McLain