Can the FCC Handle Phone Service over the Internet? [telecom]

Mr. Lazarus appears to understand VoIP about as well as a good farrier understands electronic fuel injection. Perhaps he's shilling for the Bells, who want to maintain confusion in order to get out of their regulatory obligations.

Mr. Moderator is going easy:

Well, that didn't work very well, though it took a while before they figured out twisted pair.

Crossbars and panel only existed in certain areas. Both were "Bell" inventions, to some extent NIH reactions to their non-ownership of Strowger's patents. Panel and Strowger couldn't directly interconnect (no common signaling protocol). Strowgers remained in service in service here and there until around the turn of this century, after the last crossbar and panel were long gone.

But in any case, the technological transition from manual to Strowger to crossbar to analog-SPC to digital did not require a change in the fundamental rules. Technology changes.

And that's where he goes totally off the rails. He confuses IP with the Internet. He assumes that the telephone network is going to become an Internet application. This simply isn't true, at least not with the Internet we have. The majority of "VoIP" is not really Internet at all. It is voice *using* IP, on separate, managed networks such as PacketCable and MPLS SIP trunks. The better term for this non-Internet term is VuIP; the term "VoIP" should be reserved for actual Internet voice, which runs *over* ("the top" of) the Internet.

In VuIP, the voice service provider controls the layer below IP; in VoIP, they don't, and that's a huge difference. VuIP works much better, since the Internet is *designed* to discard random packets, which is incompatible with good voice. So telephone companies are gradually replacing old TDM gear wit newer VuIP gear. It is transparent to the user. It can still feed a copper analog loop, via a media gateway (inside your cable modem, in a street pedestal, or in a central office). So it needs new rules as much as cars with fuel injectors NEED absolutely new traffic rules because they don't have carburetors, and the rules of the road were written by people who assumed that cars had carburetors. (Sort of "like a fish needs a bicycle"). But I don't think the rules of the road were written by farriers.

About 6% of US "phone" lines are actual over-the-top VoIP, and that number may have peaked. VuIP is growing, since it is merely an internal technical change. Mr. Lazarus gets that totally wrong and pretends that VoIP over the Internet is taking over.

The Internet was created as the content of common carriage, so it is not regulated as the PSTN is. The telephone companies are claiming that by using the Internet's fuel injector, IP, they are somehow no longer telephone companies. Beware. It is merely a ruse to pick your pocket.

Actually Panel and Crossbar could communicate with SxS exchanges. It was a horribly complex thing but the Bell System did it because they had such a sunk cost in the SxS.

The determining factor of who got Panel was the total volume of telephones. For example I know here in Providence, RI we had Panel that was upgraded by #1 ESS.

And I know Pawtucket, RI was served by a #5 Crossbar until upgrade to a Nortel DMS-100.

***** Moderator's Note *****

IIRC, Crossbar to Panel used Revertive Pulse senders in the Crossbar office, and Panel to Crossbar was MF. SxS Panel was always via a tandem.

Bill Horne Moderator

What regulatory obligations do the 'baby bells' still have? In our area, Verizon announced only residential POTS service was regulated. All business services and all premium residential services are not regulated.

POTS is becomming obsolete. Lots of users have dumped their traditional landline complete in favor of only using their cell phone. Many others get their telecom services from their cable TV provider, having nothing at all to do with the baby bell company. The modern baby bell services, such as DSL and FIOS are unregulated.

So, while there are some 'regulated services' still in place, it seems to me that the majority, perhaps vast majority, of service provided is unregulated.

By the time exchanges grew enough, Strowger's patents were long expired.

In the 1910-1920s, most of Bell's customers were in cities. At that time, the combination of capital and operating costs made manual service more cost-efficient in cities. Adding switchboard positions and girls to handle high loads was relatively cheap as compared to adding dial equipment--the dial equipment was there 24/7, while the girls were only needed during the peak periods.

With traffic rapidly growing in the 1920s and labor becomming scarce, Bell recognized the need for automation. But step-by-step was inadequate for heavy city service due to its 10x10 matrix. Big city service had more exchanges and more trunks than could fit. The panel system allowed a much wider selection; which included the flexibility of translation to a non-decimal scheme.

In the meantime, smaller areas were automated with step-by-step, which eventually reached about 49% of Bell customers.

"Eventually (by 1958) more than 3,830,000 lines of panel equipment had been produced and placed in service. The last new office was installed in New York City (HYacinth-9) in 1950. Although the panel system was designed to function for 30 years, many systems served well over longer periods." (Bell Switching History)

A "Panel Sender Tandem" was established that could outpulse dial codes. The first was placed in service in NYC in 1931.

A major interface goal of panel was an automated interface to manual exchanges so that callers would not need to know whether their called party was served by a manual or dial exchange. For dial users calling a manual exchange, the desired number lit up on a screen in front of the operator (later by sound). For manual users calling a dial exchange, the operator punched in the number.

Note that some panel exchanges accepted eight digit local dialing. A common use was the party line suffix, which was dialed as the 8th digit. In a few cases, a manual exchange had more than 10,000 numbers (up to 10,500), and a prefix digit was required, eg HOllis 5-10245.

While panel existed only in and near larger cities, crossbar was in many places. The early No. 1 crossbar was a large city system. However, the postwar No. 5 crossbar was intended for growing suburbs.

Due to the high cost of common control, which needs to be spread out over many lines, step-by-step remained supreme in the smallest offices.

"By the end of 1976 close to 27 million lines in the Bell System were served by No. 5 crossbar. Extensions to existing systems were still being installed, and a peak of 28.5 million customer lines was reached in October 1978." (Bell Switching History).

Step by Step remained very popular in smaller offices due to the high expense of common control of crossbar and ESS. It also had the advantage of being rugged and requiring less skill to maintain than crossbar or panel. As mentioned, it wasn't until the 1970s that electronic components became cost-competitive with electro-mechanical components.

"No. 2 Crossbar System: As a result of the successful introduction of crossbar switching to these large city applications, three less successful developments were started, applying some of the techniques to smaller offices. In

1938, development was started on a modified version of No. 1 crossbar, known as the No. 2 crossbar system. This system was developed to serve offices with a high percentage of intraoffice calling, such as would occur in a single-office city, or in a city with but a few offices. In those days the high community of interest and rate structure in suburban offices also tended toward high intraoffice calling. Development of this system, using many frames developed for No. 1 crossbar, proceeded to the point where a laboratory model was constructed by 1940. The system was designed so that on intraoffice calls, only one sender and one marker--the originating marker-was used. The call was routed directly from a district link frame to a line link frame on which the called line was located. One group of markers served all types of calls. Price studies made as the design proceeded indicated that, while the system was lower in first cost than No. 1 crossbar for this type of traffic, it did not compete with a senderized version of the step-by-step then being considered; therefore, the No. 2 crossbar system was abandoned."

"Early Crossbar COOs and PBXs" In 1935, development was also started on a crossbar community dial office (CDO). It was known as the No. 380 crossbar community dial office and had a capacity of 400 lines. One installation of 80 lines was placed in service in Jonesville, New York in

1940. The system was not standardized since its design did not include sufficient safeguards against double connections that might occur due to crossed-wire troubles. This office gave a good account of itself but was removed from service in 1946 when an addition was necessary. A similarly designed PBX with only a single marker, known as the 745A, was placed in service in the Sun Oil Company, Philadelphia, and as a result of the poor performance it too was removed from service and never standardized (see Chapter 13, section 2.1). Another plan for CDOs was known as the "43" system. This identification came from a dash number associated with a planning case. The system was frequently used as a basis for comparison in price studies because it made very efficient use of crosspoints. It never competed favorably with step-by-step CDO equipment, so development was not undertaken."

"No. 2 ESS" Although successful, the No. 2 ESS was not economical below

dial offices, even though the No. 2 ESS offered much more capability. The No. 2 ESS processor, to meet its development schedules, had to forego monolithic integrated circuit logic and the PBT memory. Furthermore, the No.2 ESS network could expand beyond 10,000 lines to 20,000, but the call-handling capacity of the No. 2 ESS processor limited the practical size in almost all offices to

10,000 lines or fewer."

"The step-by-step system provided automatic service for more Bell System lines than any other switching system when it reached its peak of 24,440,000 lines in 1973. (Later the same year the No.5 crossbar system exceeded this number.)"

It wasn't until the mid 1970s that they got the cost of small ESS units low enough so as to be cost-competitive with small step-by-step offices.

The key obligations are wholesale, not retail. The whole point of the PSTN is that it is a catenet, functionally one network under multiple owners, not an internet, which is voluntary agreements among network operators. In the PSTN, intercnnection is mandatory: in other words, "The call must go through", even if one of the carriers doesn't like the price they're getting to route it.

To prevent abuse, the price for interconnection is regulated. The originating carrier may charge its own customer whatever retail rate it wants, but the originating carrier MUST hand it off to the terminating carrier, who may only charge what the regulators approve. Note that in the internet model, that is not the case, and if the parties can't reach agreement, packets may be routed indirectly or not sent at all (though in practice some connectivity is usually reached; absent QoS, it is not like the PSTN where quality matters).

What the Bells want is for interconnection to be treated like the Internet. Then they could refuse to pay competitors to deliver their calls, and demand higher rates from competitors who send them calls, or refuse them entirely. Size gives an advantage in that situation.

Ok, but are the "Baby Bells" _today_ still of a big enough size to have that much market power? It certainly seems that today many callers are not using Baby Bell service, rather, cable TV provided, direct trunking (for large businesses), or wireless. So, a Baby Bell wouldn't be involved at all.

Further, who owns today's long-haul circuits?

In addition, if the Baby Bells tried unscrupulous wholesale charging practices on interconnections, then their competitors could retaliate in the same way. The competitors today are big enough to have a solid foothold in the marketplace, even if not dominant. For instance, Comcast today is a powerful company with deep pockets.

One cloudy area is today's "at&t", which is both a long haul carrier and local company.

Lastly, wasn't the whole idea of Divestiture to create a non-regulated marketplace? In the real world, marketplaces are not perfect--some companies grow stronger than others, some take advantage of special circumstances to charge very high rates when possible. So, a convenience store will charge more for groceries than a supermarket. That is the real world.

We can't have a regulated marketplace when it suits us, and a competitive one when it suits us.