Circa 1976, Long Distance Network Mapping Project

Copyright © 2011 Andrew K. All Rights Reserved.

I had the opportunity to do some university level work in the 1970s related to the cost of operation and ROI feasibility for long-line costs over the old Bell Operated Public Switched Telephone Network (PSTN). The mid to late 1970s were for many the culmination of a golden age of the post-war construction and refinement of a reasonably reliable network before a series of big-bangs in the late 1970s and early 1980s (Bell System divesture, implementation of fiber-optics transmission, rampant inflation, computer technology and related improvements, among others) brought about a profound change in the network.

I am adding some files to the Telecom Digest archive. The files, which are in .pdf format, reflect the nature of "default" routing (or what we might call today the backbone) of long distance traffic in the PSTN as it existed around 1976. My source documentation is the ATT Long Lines Routing Guides that I had access to and kept copies of while I was doing my university work. I rely on materials that were published in late 1975, supplemented with other documents from 1976 and 1977. Much of the network design characteristics and paradigms depicted here were abandoned in the early 1980s when a more direct switching and call routing paradigm was adopted, resulting in part from the big-bang items described above, which gradually replaced the older, more limited copper & coaxial wire and microwave-based networks at the time.

At the time (circa 1975), calls would default to a route based on what was termed a "homing" priority. There were five levels of ranked switches, or a hierarchy of switches based on where calls needed to go. These five levels, or classes, were known as Regional, Sectional, Primary, Toll and End Office. There were 10 Regional switches in the US, and 2 in Canada, with several hundred sectional and primary switches. A call from an End Office would be sent to the Toll Center, and if the destination number was not available and outside the end offices served from the Toll Center, the call would be routed to the next higher center, and switched from that point if feasible, and if not, the call would then be sent so on to the higher ranked centers. For example, under the strict homing hierarchy-topology, a call from Walla Walla, Washington to Moscow, Idaho - a distance of less than 100 miles - would route (absent any other options) from Walla Walla, to Yakima, to Seattle, to Sacramento, to Denver, to Salt Lake, to Boise, to Coeur d'Alene and finally to Moscow. This routing would be the last resort, absent local instructions to route a call differently based on local resources. Likely the call would have been routed from Walla Walla, to Yakima, to Spokane, to Coeur d'Alene and then to Moscow - a much shorter distance and cheaper cost. There were many, many of these local alternative routes based on local requirements. In the event of an alternative network being unavailable, or non-existent, a call would then be switched based on the default homing route. The network was constantly being upgraded and developed to create routes that would represent the cheapest and most expeditious route. Changes occurred almost monthly.

The depictions herein, represents the "backbone" default homing priority for traffic. In studying the layout of the network, it does give the reader some perspective of the population distribution of the time, and planning based on local factors and conditions. Some of the routing decisions do not necessarily make sense in a modern perspective, as population patterns as well as the supporting physical network have changed.

The 12 Regional centers are reflected by cities with a purple color code. I did not color code the various levels as a call is routed based on the terming convention discussed above. I developed this exhibit to allow the reader the ability to perceive visually how calls would have been routed. Thus, from each "purple" center, all lines are red and lead to "red" centers, which might have been a sectional, primary or toll center. From a "red" center, blue lines lead to the next down "blue" center, and then yellow lines onward to a "yellow" center (if required).

Local "tributaries" (also known as offices, exchanges, and prefixes) would represent the last link to the primary office that provides dial-tone to the customer. Regardless of the level of the switching center, lines to tributaries are always in green leading to the tributary. While most centers have tributaries, some do not, e.g. Paso Robles, California.

The same color lines do not cross. A forking or 4-way breaking of a same color line should be construed not as a junction or switch, but a joining of traffic from a number of subordinate nearby locations to a higher ranked switch. If for some reason the same color line does have to cross (and of this writing there are few examples), before the intersection the solid line becomes dotted as it crosses another line and then returns to a solid line. The lines should also not be construed to represent the likely or actual physical route of the actual transmission apparatus.

I work on this more as a hobby. I do not work in the telecommunications industry, nor do I have any business interest therein. My work mainly has to do with people and their disputes and trying to resolve their disputes. I retreat into a quiet moment now and then and work on this project in solitude and as a form of therapy. Additionally, this is a work in progress and I will update it and make the updates available.

The Telecom Digest operates under publication rules that provide that all materials in the archive can be accessed without the purchase of software. I am completing my primary work using Excel and then transferring the work into a .pdf format for the reader. Regretfully, as of this writing, the transfer to .pdf enables some limited breakdown of the clarity of the data, in particular lines moving in a northwest - southeast and southwest to northeast direction.

To receive a more detailed appreciation, it is best to look at the work in Excel. I would welcome users who would be interested in helping me through comment and feedback. In particular, I would welcome engineers who "were there" who might offer some insight to amplify this work. I would be happy to send copies of the work directly to you off-line, for your comments.

The first .pdf files are available now:

AZ.pdfRoutes in Arizona
CA_and_NV.pdfRoutes in California and Nevada
WA_OR_ID.pdfRoutes in Washington or Idaho

-Andrew K