Interesting blog post concerning the telecommunications act of 1997 and the modern day. It seems Senator McCain? is advertising what he got done. http://talkingpointsmemo.com/archives/217239.php Reed Hundt's response is interesting too.

-- TomGlaisyer - 16 Sep 2008

Reviewing where we left off last time: Eben was discussing the relationship between the concept of The Internet as a physical object, and the “Internet Society” meaning “everybody connected to everybody else without any intermediary.” That connection is the new thing. There isn’t an object “the internet.”

In 1965, what mattered was INTER net, the ability of networks to network without regard to content. That is no longer a thing worth naming. I talk about “The Net” because it’s no longer important to capture “the network of networks”.

Still, it’s relevant that the Net is made of physical things that can’t be copied without marginal costs. In particular it’s made of two things: pipes and switches. Defined abstractly, a PIPE is an entity which moves a signal from point A to point B without change and without selection. It has two ends; it can be unidirectional (“broadcast”) or bi-directional. A SWITCH selects packets to be put into pipes -- connects pipes by deciding who receives what signals/when/how/at what price.

The characteristic of 20th century communications technology was a reduction in the cost of piping, and no reduction in the cost of switching. The reduction in the cost of ductile metal products gave us, e.g. miles and miles of wire. In 1920, the piping was already reasonably inexpensive. The Edison/Bell set—phone wire, electric wire—had made piping very cheap. Switching was still in human form, e.g. a skilled operator. Around WWII, “cyborg” switching was made very expensive (NLRB / communist accusations) so capitalism began replacing switches with _ _ exchanges (… “click click”)

So here we have to understand how switches work. Circuit Switching: when 2 points in the net want to be connected to one another, you can find pipes matching the 2 points, and you FREEZE circuits in place connecting the points, and you leave that circuit in place as long as the contact lasts; and then that circuit is taken apart. This is the traditional structure of the telephone network. ATT in 1946 proved to itself (w/ help of Claude Shannon) that this was the only way this could happen. What we knew about information theory (created by Shannon) wrongly proved that this was the only way. Because that’s how switching worked, certain characteristics followed. A 20th century telephone bill had 2 elements. One was RENTING TERMINAL EQUIPMENT. This was necessary because the network [note-taker spaces out] So there would have to be a monopoly in manufacture and distribution of the terminal equipment.

But that was less than what MOST of your phone bill was for: SWITCH RENTAL.

Two theories evolved about switch rental in … In the US, you rented for a fixed rate, regardless of the length of the contact. The purpose was __ for short calls because the assumption was that people were going to make short calls. The European approach (related not necessarily economics but to the politics of listening) was to rent you the switch by the minute – by the impulse or tick. You paid switch rental by the unit. This was one way to make switches. It was the dominant way of making switches before the late 1960s.

But there’s another way to makes a switch: you can say that any form of communication can be decomposed into small digital bursts (think of a burst = “one line of text.”) And think of the whole world broken up into a series of units, called “packets”, of length of one line of text. Each packet is given to a switch, with instructions, “get this to its destination.” Each packet has a little header on it: “here’s where I’m from, here’s where I’m going, get me there.” The rules that translate headers into getting packets places are what we call “internet protocol”. In a packet switch network, anything broken up into a large number of packets and sent to its destination along many different paths may fail in various ways.

1. Some packets fail to arrive—some route has failed.
2. more likely, some packet arrives twice, because some mid-path entity “sends another one just in case.”
3. packets arrive out of order. Of course.

Student question: What percentage of _, because Professor Wu doesn’t know. Eben’s answer: 1) There are a handful of stepper-bar exchanges left in rural America. 2) . 3) The beauty of the __ without having to declare anything at either end. Since 1962, a lot of the copper wires in the ground or overhead have ceased to be good. So what Verizon does is it doubles up or triples up calls: by using one circuit switch loop, and … So the reason Tim Wu doesn’t know is because Verizon doesn’t know. … But the overwhelming majority of calls made over telephones in the United States are packet switched; and every cell phone call is packet switched.

So in this world, what we have essentially is pipes & switches moving packets; and everything can be broken up into packet networks. We could in fact say we have one ring to rule them all: at the bottom of everything is IP – how to get things from A to B according to the addresses.

But it doesn’t tell you how to reassemble packets. We need a TRANSMISSION CONTROL PROTOCOL (TCP). It doesn’t say what this stuff means. TCP doesn’t care. It only tells you how to regulate transmission over IP.

That combination, TCP/IP, is what we now mean when we talk about The Net.

It doesn’t have to be this way: the are other rules for transporting packets—the electric system in this building, etc.

But the rules for assembling data out of packets are the rules for how the Net functions.

The regulatory structure of telecommunications in the United States knows nothing about what I told you. It knows about phones (long and short distance), radios, and televisions.

The regulatory structures are stupid. And the owners of these devices humiliate the regulators. It’s not because the regulatory structure was made in 1927 – because it was make in 1996. Because the way it was made, was the way that infrastructure was built in the 19th century. (See “The invisible barbecue”).

…

And the regulation works beautifully. Because it is so tedious, so full of junk, you will glaze over. It is built to make you want to stop learning it. It is a great example of modern American regulation. It is made by people in temporary control of government to be so complicated that you need to hire people to be in temporary control of government to make it work.

If you fall asleep, I won’t be offended, but you need to hear it. Nobody read the whole statute, and almost nobody has read the whole CRS report. I’ve taught this statute for a dozen years since it was published. Nobody has read it yet. Even Peter Strauss. You can understand how Rupert Murdoch got away with it. The companies whose job it was to explain it to you, were the companies who benefited from it. That’s why I wrote the invisible barbecue.

Rick Schwartz Question: Because Congress wants companies that have the capital to put in the networks, how do you reconcile the need to build up the capital, with the fact that they need some kind of return on their investment? Eben: Map first, war second.

So all I want to figure it is what was happening. So first we transport our minds back to 1996. “Telecommunications Reform” has been on the mind for two decades. It’s the second biggest reform on the mind of congress. The biggest has been on the mind for four decades—the repeal of the Glass-Steagall Act (financial industry deregulation.)

We’re going to get both of those done because a master of triangulation is in control of the White House. And because he’s lost the health care reform, and with it he lost the congress.

So now there’s a question: who will be president in 1997. Remember, Mr. Clinton (Mr. top of his class) had a near death experience when he got voted out as Arkansas Governor. This is his fear again. He and Dick Morris have a plan: to bury the other side with money. He and Dick Morris (a theorist of media in politics) have concluded that this is the Last Television Election. They believe that television is becoming fragmented. In 1965, there were just three networks. You just needed to buy three networks. Today, there are hundreds of networks; therefore we need vast amounts of money to buy the networks. (Phillip Morris actually said that.)

In order to __ Clinton had three instruments. 1) the death of welfare 2) a new-and-better deal for television 3) the repeal of Glass-Steagall, and the unleashing of the financial industries, to operate in a free market. (The heart-child of Bill Rubin—who is now trying to say, as the advisor of Citi Bank, he … Characteristically, he did a brilliant job of making it happen. There were not children starving in the streets. The repeal of Glass-Steagall was the Jarndyce v Jarndyce. But under Clinton it HAPPENED.

The 1996 Act is a product of the people who made it, and their formidable skills; and a more loathsome piece of regulation would be hard to find.

One of the weaknesses of our telecom law is that it views the world in these baskets (phone, cable TV …). To understand it, we have to grow it from below.

Excerpts from the Telecommunications Act of 1996 CRS Summary

• (Sec. 202) Directs the FCC to modify its multiple ownership rules to eliminate its limitation on the number of radio stations which may be owned or controlled nationally. Limits the number of radio stations an entity may own, operate, or control in a local market, with an exception when the FCC determines that such ownership, operation, or control will increase the number of radio broadcast stations in operation. Directs the FCC to: (1) eliminate its limitation on the number of TV stations which may be owned or controlled nationally; (2) increase to 35 percent the national audience reach limitations for TV stations; and (3) conduct a rulemaking proceeding to determine whether its rules restricting ownership of more than one TV station in a local market should be retained, modified, or eliminated.

• (Sec. 702) Makes it the duty of every telecommunications carrier to protect the confidentiality of proprietary information of other carriers, equipment manufacturers, and customers. Permits a carrier that receives proprietary information from another carrier or a customer for purposes of providing any telecommunications service to use such information only for such purpose. Directs a carrier to disclose customer proprietary network information upon the customer's request. Permits a carrier to use, disclose, or permit access to aggregate customer information for other purposes.

<--/commentPlugin-->