| switch class | functional designation | no. in Bell System | no. in independents | total |
|---|---|---|---|---|
| 1 | regional center | 10 | 0 | 10 |
| 2 | sectional center | 52 | 0 | 52 |
| 3 | primary center | 148 | 20 | 168 |
| 4 | toll center | 508 | 425 | 933 |
| 5 | end office | 9803 | 9000 | 18,803 |
Originally there were 164 LATAs. Over time additional pseudoLATAs were assigned to independent company serving areas (originally LATAs were only assigned to RBOCs). Other pseudoLATAs were assigned to foreign countries in the North American Numbering Plan. Presently, the total of these additional pseudoLATAs for the non-RBOC companies in the U.S. is 35. If you have any additional questions please contact Phil Cheilik of our staff directly.
I believe that this is true for Australia as of early this year (Geoff Huston posited as much in Memphis) and both PacBell and AADS are making similar claims. I've heard rumors that the curve is near crossover within MCI and Sprint as well. Bill Manning
A senior person at one carrier told me that their leases for rights-of-way are actually measured in 64 Kb/s channels carried.
| USA | Germany | France | Sweden | |
| TV hours per day | 4 | 3 | 3 | 2 |
| Newspapers, /1000 | 226 | 156 | ||
| mail, pc/yr | 670 | 230 | 418 |
For the same reason, we shouldn't try to subject Internet telephony to all the rules that apply to conventional circuit-switched voice carriers. Imposing traditional divisions, like voice vs. data or interstate vs. intrastate traffic, on Internet-based services is wrong-headed and futile. Internet telephony may well become, in time, a competitive alternative to traditional circuit-switched voice telephony, especially in areas like international calls and calls over private corporate networks. We want to encourage that kind of competition, not limit it. I hope the FCC bars any state from limiting the growth of Internet telephony. We want states to regulate less, not more.Carriers engineered and deployed their switches based on the characteristics of voice traffic. Internet users, however, typically engage in far longer calls than voice users. Several local phone companies and Bellcore found in traffic studies that the average voice call lasted between 2 and 5 minutes, while the average Internet call lasted between 17 and 21 minutes. The average end user circuit in a central office was in use 5 to 7 minutes in the busiest hour, but the average circuit connected to an Internet service provider was in use between 31 and 47 minutes in the busiest hour, and some are in use virtually nonstop.
The existing networks weren't built for this sort of use. And the same switches that are being overwhelmed by Internet usage also provide voice connections to other users. According to Bellcore's models, if only 4% of the lines into a central office are in constant use by Internet service providers, users, including non-Internet users, will face a 60-fold increase in the number of calls that don't go through.
AT&T share fell from nearly 90% in 1984 to about 54% as of the third quarter 1996. MCI's share of long distance revenues increased from about 5% to 18% during that period; Sprint's share increased from 3% to 9%; and WorldCom (formerly LDDS/WorldCom) obtained a market share of almost 5%. The market shares of smaller carriers grew even faster, increasing from 4% to 15% during that same period of time. press release:
Average annual expenditures on telephone service increased from $325 per household in 1980 to $623 in 1992. About 2% of all consumer expenditures are devoted to telephone service. This percentage has remained virtually unchanged over the past 20 years, despite major changes in the telephone industry and in telephone usage. Residential toll calling grew by about 10% a year between 1985 and 1989 a period when toll rates declined dramatically. The average American household now spends more on long distance service than on basic local service...
Intrastate toll minutes increased from 8% of all minutes in 1980 to 12% in 1992. During that same period, interstate calling minutes increased from 8% of the total to 15%.
As shown in Table 17, the average telephone line is used primarily for local calling and is used somewhat less than an hour per day. This level has remained relatively constant for a long period of time despite increases in long distance calling and the introduction of facsimile machines and other devices that affect usage. In 1992, services provided by long distance carriers generated more than $59 billion in revenues. The total toll market, including the short haul toll traffic handled entirely by local telephone companies, was $73 billion in 1992, with AT&T accounting for about half of the total. The number of [international] calls has increased more than 600% since 1980. In 1992, Americans spent about $10.3 billion dollars on international calls. Five markets -- Canada, Mexico, the United Kingdom, Germany, and Japan -- currently account for about half of the international calls billed in the United States.
Engineering and Operations in the Bell System (p. 125)
| type of office | CCS/hr | Intraoffice (%) |
| Urban | 3.1 | 31 |
| Suburban | 2.7 | 54 |
| Rural | 2.1 | 66 |
Deutsche Telekom: 50% of local calls are shorter than 90 seconds
CCS: hundred call seconds/hour (fully used: 36 CCS) BHCC: busy hour call capacity
roughly 10 calls (internal/external) per station per hour
GMD: 5 external lines per 100 stations; for 3 minute calls, 20 calls/hour/line; thus, 1 external call/hour/station.