Transcript: Community Broadband Bits Episode 216

This is episode 216 of the Community Broadband Bits Podcast. Fred Goldstein, a principal at Interisle Consulting, returns for part II of "What is the Internet?" with Christopher Mitchell. Listen to this episode here.

(Listen to part I of the discussion in episode 213.)


Fred Goldstein: The Internet is not a thing, the Internet is a phenomenon.

Lisa Gonzalez: This is episode 216 of the Community Broadband Bits Podcast from the Institute for Local Self-Reliance. Welcome, I'm Lisa Gonzalez. Just 3 weeks ago we introduced you to Fred Goldstein, the principal of Interisle Consulting. He and Chris had an in-depth discussion about past FCC decisions that have influenced where we are now as we consider the question, “What is the Internet?” Well, Fred is back again with more on history, and how our perspective of the Internet, both upper- and lowercase I, has, and will influence innovation. Check out Fred’s firm at to learn more about their work, and about Fred’s extensive experience in telecommunications. Now, here are Fred and Chris.

Christopher Mitchell: Welcome to another edition of the Community Broadband Bits Podcast. I’m Chris Mitchell, and today I’m returning with Fred Goldstein, a principal of the Interisle Consulting Group. Welcome to the show.

Fred Goldstein: Thank you Chris, glad to be here.

Christopher Mitchell: We’re having you back to continue our discussion. Previously we got into a longer history of how we got to the Internet. I think, in some ways, the telecommunications policies that allowed the Internet to develop. And when we finished, you had just told us what the Internet is. Now, I wonder if you wouldn’t mind just briefly reminding people what the Internet is?

Fred Goldstein: Sure. I define an Internet as a voluntary agreement between network operators to exchange traffic for their mutual benefit. Voluntary means it’s like a real business. Not a monopoly, not regulated, they agree on the prices, they haggle on prices, they interconnect how they want, when they want for prices they can agree to. They exchange traffic for their mutual benefit, but it’s not the underlying network. The Internet is the computer mediated interconnection between networks. The networks are a higher layer than telecommunications, which is a raw exchange of bits.

Christopher Mitchell: Now, I wonder if conceptually, would it make sense to think of this as, in a metaphor, traffic is not the road. Traffic is the things on the road. The roads were telecommunications, the Internet would be traffic as we think of our daily commutes, and things like that.

Fred Goldstein: That’s a very good analogy. The idea being that, right, the Internet is the cars riding on a road, and the road is the telecom layer, and the vertically integrated model we have now is that Interstate 90 has become the Chevrolet highway, and only Chevrolets are able to use it.

Christopher Mitchell: Right, and so when it comes to the Internet I think, and this is something that I’ve been railing about, because it drives me nuts that the AP style manual has stopped making it capital, but when we talk about the Internet, is there really one major Internet at this point, or am I thinking of it incorrectly?

Fred Goldstein: Some people don’t know whether Internet is uppercase or lowercase. The answer is, there is an Internet uppercase called “the Internet, capital I.” There is a prototype. There is one public Internet that we call “the Internet, capital I.” This is like, there are a lot of country clubs where rich folk around the country hangout, but in Brookline Massachusetts, there is The Country Club. The Country Club was founded in the 19th century, and it caught on. Other rich people around the country said, “We’ll open our own country clubs.” They used the term country club generically. It’s the same thing with Internet. There is one original Internet, but there can be other internets. Well, at least, there can if the telecommunications is made available. One of the problems with the FCC’s understanding, is that they think all progress halts when the Internet happened, the Internet is a, essentially, finished model, and all you need to do is turn up the speed of the access pipes, and magic just continues to get better. That’s ridiculous. The Internet was always an evolving technology, and it’s based on early 1970s ideas. TCP IP was specified first. TCP in 1974. IP version 4. The only version that works, 1978. We have people talking about evolving the phone network from old TDM, to new IP, IP is older than TDM. They’re going backwards. IP was a lab hack that demonstrated certain principles. It’s not an ideal protocol.

Christopher Mitchell: This is something that I always think about from my days I used to program, and one of the things I learned early on was that a lot of times when you think I’m a “This will just get me from A to B, and then I’ll do something better.” You never come back, you’re always stuck with that barely working bridge, and that’s IP in many ways. Although, it’s gotten us quite far.

Fred Goldstein: It’s gotten us far, but it’s gotten us far with such brute force. It’s a monoculture that people are taught. If you actually objectively look at it, you realize, it’s a 1970s lab hack that grew out of control, and that’s why we don’t have adequate security. It was not designed for security, it was designed to be a closed network, that security by who is allowed to attach. Not the public.

Christopher Mitchell: Security or privacy, for that matter, which are related, but not necessarily the same.

Fred Goldstein: Absolutely, and why it doesn’t provide reliable quality of service, it wasn’t designed to, there was a telephone network at the time, and it was explicitly designed not to compete with it. You have all these variations where the Internet’s being used for things it wasn’t designed to do. One can certainly posit other internets, and there are other internets. Private companies and industries, just to give an example of -- We’ve heard of the SWIFT Network, that the Society for Worldwide Interbank Financial Transfer recently got hacked. It’s an extremely secure network, but a bank in Bangladesh got hacked, giving hackers the ability to use SWIFT to steal money. The SWIFT network is, in effect, a private Internet of the banking industry worldwide that uses its own techniques to be extremely secure. One of the parts of that is it’s very closed access. Airlines have closed Internet. When you have multiple companies exchanging data with one another, that’s an internet. The “big I, public Internet” is the one that benefit from worldwide access. That’s where you put public things. Work with public safety clients who run a private Internet. The fire and police departments don’t want their traffic exposed to the public. There are very carefully controlled gateways for logging in, VPNs between the two, but it’s a private internet. Many, many public safety agencies around the country, and around the world, and other governmental agencies build private internets. Sure, it’s not that there is one internet, but there is one “The Internet,” which is the big one that is the most insecure.

Christopher Mitchell: To imagine a second big Internet, or a second very large public Internet is almost inconceivable, because if you are trying to build something like that, at some point, I think you’d want to interconnect to “the larger Internet,” “the original big Internet,” and at that point you’d be part of “the big Internet.”

Fred Goldstein: Depends on how you connect. Thing is, if you were to build another internet, it wouldn’t be using the same wide-open policy and wide-open protocols. You’d connect, but you’d probably connect through gateways that gave secure, limited interconnection. Essentially by definition, when you can exchange raw IP packets, you’re on the same Internet. If you want to have another Internet, you might want to be able to exchange email, but you may have to go through a gateway that would be a secure process. I think security is really the key feature, but also it could be that the other Internet is using different protocols

Christopher Mitchell: Let’s talk about some of my grammar usage, bugaboos that drives me nuts, and people who work with me I think, and annoyed with me ranting about it, which is, when people talk about how we want the government to implement policies for “fast Internet,” or we have problems with “slow Internet,” it always drives me nuts. I always want to make it very clear, we’re talking about Internet access, because “the Internet” itself is not fast or slow, it, as you described it, it’s something else. If this agreement to exchange traffic, and that in and of itself is not fast or slow. When we talk about fast or slow, we’re talking about the telecommunications network that are supposed to be a different level and just transporting bits. Am I getting that right?

Fred Goldstein: Precisely. Telecommunications’ facilities have a speed. The Internet is not a thing, the Internet is a phenomenon. The Internet is a phenomenon that results from these agreements to exchange traffic. A phenomenon doesn’t have a specific speed, there are physical phenomena that do, but the Internet doesn’t. The individual components have their own speed, and that’s why your end-to-end speed is so variable. Because, to get from your ISP, to wherever you’re going, you don’t know what path you’re taking. The path may be variable, it may be crossing 5 different providers’ networks. The Internet wants to be able to route around damage automatically, and the way it does that means you can’t tell for sure how you’re getting from one place to another. Even if you trace route, it can change, and the 2 different directions are taking different paths in many cases. How can you predict end-to-end? That’s the beauty of it, it was never the promise. The Internet never promised to speed. The Internet nearly promises “best efforts” connectivity. “Best efforts” by the way in quotes. I call that BESQR, “Best Efforts Scare Quotes,” required service. Not making a guarantee. That’s what a lawyer would call best efforts. It’s just, we tried to get packets, and we try to get them from end to end, and if it works great, if it doesn’t, we’re sorry. That works very well most of the time. You can’t guarantee speeds on the Internet. If you did, you would be providing more like telecom, because that means you’re reserving facilities, reserving capacity. But, the Internet doesn’t reserve capacity.

Christopher Mitchell: Let’s talk about where the FCC may have gone wrong in terms of its net neutrality rules. Because, it seems to me that the goal of many net neutrality advocates is to maintain this separation between the telecommunications’ facilities, and the content, and the Internet effectively. You’ve argued persuasively that the FCC didn’t do it in the best manner possible. Where did it go wrong?

Fred Goldstein: Well, they went wrong because they didn’t understand the layered model of the Internet. I don’t mean the layered model of the TCP stack, or the TCP lay, or the yada yada, because, TCP stack is not a layered model. TCP/IP is not an actual layered protocol stack. It’s a single integrated protocol entity from application protocol down to the IP layer. That’s its architecture. It’s not layered internally, because there are IP addresses visible in the application layer, and at the API, the sockets layer. That's a mistake in the architecture, but that means it’s not a layered architecture.

Christopher Mitchell: I think just a few people would benefit from talking about why layers are important. In my mind, layers are important because then they are separable, you can change them around with different approaches that might be better, and you can change one layer without impacting another. Is that why layers are just important for understanding this?

Fred Goldstein: Well, it’s important because it allows us to distinguish who has what role, and what is the proper way to regulate something? If you have layers, you can understand how a layer behaves, and regulate it appropriately. If you don’t understand how the layering work, it’s very hard to regulate it appropriately because the actual phenomena that are the Internet don’t line up with the FCC’s view of how the phenomena works.

Christopher Mitchell: Okay, so please continue then.

Fred Goldstein: Okay, so, essentially the way it worked, and the way the Internet came about, was that there was this distinction between enhanced service and basic service from the computer inquiries, going back especially to CI2 in 1980 that said the phone companies had to make the basic telecommunications available to everyone. They could offer DSL, but other ISPs could as well. That meant that the Internet was competitive. If you didn’t like your ISP, you could change your ISP. No one thought of regulating ISP behavior in 1997, nobody talked about regulating these services because they were all clearly computer services, enhanced services, information services, pick your term, that ran over telecommunications. You had competition, market forces, the ISP could change, there was no net neutrality in AOL. It was a timesharing service. Literally a time-shared computer service where web browsing was one of the things that their terminal could do when they first offered this in the 1995 timeframe.

Christopher Mitchell: Right, and anyone who wants to build a competing service could just order that from the telephone company, and basically get the same deal to get the same lines and everything else. As you pointed out in our previous conversation, the FCC rescinded that, understanding that rule in 2005.

Fred Goldstein: That’s right, there were 10,000 ISPs more or less. It’s hard to count there were so many. About 10,000 ISPs in the United States around the turn-of-the-century. What the FCC and its war on ISPs did is shut the vast majority of them down. What’s left, the one corner left alive is in the wireless space. The wireless ISPs don’t depend on Bell's circuits. I do work in that space myself, I do work with a wireless ISP association has which has about 800 active members around the US. Especially in rural areas, where there are no DSL or cable competitors, and the wireless ISPs step in. In the cities, the available radio spectrum isn’t really enough for wireless ISPs to offer total competition. They can offer some competition to some customers, but really, because of density reasons, the wires are needed. The urban and suburban ISPs mostly went out of business, and competition is really just left in the rural areas. Yeah, there was no thought of regulating the Internet because you regulated the telecom. In the TCP/IP literature, going back to 1978, there was a break between the network and the Internet. IP (Internetwork Protocol) is what ran above the network. You had network, meant the wire. You had the network link coming from your Bell company with a modem. In 1978, it was a 9600 per 2nd modem. That was expensive. That was telecom, and it was understood. When Ethernet came out, that was the network. Internet was the payload of the network. Here is a distinction, the addressing. Telephone numbers are the name space of the telephone network. Ethernet has its MAC addresses. Lower layer networks, X.25 network of the 1970s and 80s, they were popular in Europe, had their own address space. The underlying network had an address space, and an underlying protocol, it was not IP, an IP was in the payload of that network. A very, very clear carrier versus payload distinction. The Internet was the payload.

Christopher Mitchell: Right, so if we think about that for a second, you might say a bus stop is an address. As a person, I get on the bus, and the bus doesn’t really care who I am, or what my goals are, or anything else, it’s goal is just to get me from place to place, and then I get off the bus, and I have my payload with me, and I go about my business.

Fred Goldstein: That’s right. If you change buses, bus A doesn’t know you’ve gotten off, and then got on bus B to then go off to complete your journey. Of course, in the real world, you sometimes have to take more than one bus. In the Internet, you have to take many networks, it’s an internetwork. You’re going across many underlying networks. The point is, the original concept of network was neutral, but it wasn't the Internet. You were regulating neutrality because it was common carriage. And, common carriage is not allowed to peek at the payload. Common carriage isn’t allowed to discriminate among providers. It worked very well until 2005 when there was real competition. Now, there were issues, cable companies were never common carriers, cable companies did not have to make their underlying wire available because they had never been phone companies. That was a very controversial ruling the FCC made, it was legally justifiable. The Supreme Court held it up on narrow grounds, that it’s a little questionable, but the FCC’s the expert agency, it’s not outside the possible reading of the law. Perfectly rational there, but the phone companies were always meant to be common carriers. The people who wrote the Telecom Act of 96, which was written at the beginning of the Internet era, knew full well that there was this distinction between carriage and content. The phone companies offered carriage, and the ISPs were content. The FCC, in the 2000s, took that away. Now, the trouble there is that once there was no competition, once the Bell company said, “We’re not making our DSL available to competing ISPs.” And cable never did. The competing ISPs went from having one source of bandwidth, to 0. The ISPs were cut off, and you had a duopoly. Two ISPs in most cities— the phone company and the cable company. That’s not real competition, that’s very little competition. Well, when you have very little competition, all sorts of abuse can happen. The Bells were really looking at doing some things that were really pretty bad. They wanted to charge by the application, the way CompuServe did 20 years earlier. They wanted to be able to discriminate bits that were used for video at one price, bits that were used for voice at another price, bits that were used for web browsing at a third price. That was their model.

Christopher Mitchell: I think just for people’s edification, one of the problems that I typically have with that is, not just that some applications might be discriminated against, but it actually, when you start doing it, it makes it very hard to be innovative, because you have to start asking permission, and the company has to make a determination of which rate are we going to charge you for this new thing you’ve developed? It gets messy in ways that I think are very anti-innovation.

Fred Goldstein: Precisely Chris. It’s very anti-innovative, they wanted to milk the money, rather than allow innovation. That was a real threat. Now, the obvious answer was, allow competition. Competition brings innovation. The FCC in 2005 didn’t want competition. The trouble is, the term network neutrality was coined in 2005, and instead of being applied to what was the “network” in old terms, people saw the Internet as the network. People wanted to regulate IP transport. People wanted to regulate ISPs, and let’s face it, the Bell companies and cable companies were calling themselves ISPs. The FCC, later, politics made this necessary, came out with the rules to regulate ISPs. Now we have this topsy-turvy world where they deregulated the telephone companies who were still largely a monopoly, and regulated the ISPs who were the ones the law was meant to not regulate. They’ve stifled progress. They decided that the progress that needed, the rules for “network neutrality” are, in fact, rules of regulate the Internet that are designed to allow certain applications to work without discrimination. Primarily video. The FCC really wants to make sure Netflix can compete with cable, and this goes back to the stove pipe system at the FCC where the media bureau regulates cable, and are absolutely afraid to do anything to open up cable systems per se. The media bureau regulates cable. And, since they don’t want to do anything there, they want the wireline bureau to open up the Internet to carry TV in competition with cable. It’s a new open model cable being done over a medium that wasn’t designed for it, the Internet. The Internet was designed for variable rate, “best efforts” data applications, not designed for high-bandwidth streaming. The FCC’s neutrality rules were really all about making sure ISPs didn’t block Netflix, and any competing, Hulu and other, video services.

Christopher Mitchell: Right. I think that’s because that’s where the concern is right now, because there is a lot of investment in those areas, and it’s a hot topic. If this had come about at a different time, it might have been YouTube more specifically, or it may have been, in the future, it might be a social networking platforms or something. To some extent, it’s hard to say, but I think right now there’s a sense that the video services are so desirable that that’s where we are concerned about harms. I think it’s really worth pounding this home, because I think a lot of people have forgotten, and it’s a little confusing. What the FCC did was basically they made a lot of rules that govern how ISPs have to behave. And what you would say is that ISP should be able to behave however they like, and the FCC’s job should be making sure that anyone that wants to create an ISP has the ability to order the relative -- Rather than having to lay their own fiber, they can order the service from the telephone company effectively. A physical service that they can then use to create services.

Fred Goldstein: That’s right, that’s what should’ve been done is that the network, meaning the wire in the ground, should have been neutral. Yes, they are worried about TV competition for programs, but that’s one application. By turning the Internet into TV, into open cable, if you go back about 60- 55 years ago, Newton N. Minow was the FCC chairman, coined the term “vast wasteland” to describe programming on TV in the 1960s.

Christopher Mitchell: I don’t know if he’s still around, but I would love to know -- I wonder if his head has just exploded at this point.

Fred Goldstein: Well, yeah. In fact, I actually coined the term, when the FCC was about to revoke computer 2, that what the Bells wanted to do was turn the Internet into “fat waistband”, not broadband, but a “fat waistband”. What they’ve gotten instead now is a “vaster waistband.” Because they’ve got more TV, but they are not really making the wires available to innovate on anything, but the video content and social network content, the innovation that created video and social networks is restricted because the transport is still someone else’s IP. It’s monopoly IP service, and IP is a -- I say, it’s a very old protocol that doesn’t do everything that people might want to do, but without access to the underlying wires, you can't do anything about it.

Christopher Mitchell: Now, we see municipalities. My passion are doing a lot of things, a lot of different things. Some of them are building open infrastructure, some of them making dark fiber available. Some of them are making the network available at level 2 or level 3. We see other networks like in Ammon, Idaho, where they are using Software Defined Networking to try and make different slices of the network available. Then we have some municipalities that will say, “Well, we provide open IP, and we consider that to be open access.” You’re saying that, based on our history, that, that should not be considered open access. That IP is not good enough, that if we really want to solve these problems, we need to have more wires that are open so that different ISPs can use them in their own ways, and compete, and just innovate and all of that. Is that right?

Fred Goldstein: Exactly, exactly. The IP Internet, IP services make many, many decisions as how they operate, how they prioritize, how they set parameters that impact the performance. That should be competitive. The physical networks should make it available as a “layer 2 service”, or a “layer 1 service”. Now, there are some cases, and the very small cases, and the wireless ISP case, it’s sometimes very hard because these are very limited, very small networks, and very limited bandwidth. As soon as you have fiber, or even in some cases wireless, it’s possible, and not even difficult, to make it open. A fiber network should always be operated as an open network. Most of my work nowadays is with municipal networks. Rural municipal networks. I do recommend that they be open, but most of the people there don’t know what I’m talking about.

Christopher Mitchell: When you say open, what do you mean specifically?

Fred Goldstein: Well, I mean, they shouldn’t just be designed around IP. That there should be the ability to offer lower layer services that, not just the one ISP service, but other services besides the ISP should be able to use the physical network.

Christopher Mitchell: What kind of a change would one make if you originally have a design in which you’re going to be the only one offering “Internet service” to home users and what not, how do you design it differently to make sure it’s open?

Fred Goldstein: Well, fundamentally, you have to have the ability to offer a service at a lower layer than IP. It could be, for instance, Ethernet. If it's an active Ethernet. Ethernet is carrier Ethernet and an active Ethernet our services operated below the IP layer, and they are almost inherently open. With a passive optical network, it can be done, but you have to offer service through the MAC layer, it requires designing for that. In the case of wireless, and some I suppose with PON, one can offer VPNs, of a layer 2 VPN type services that, again, would be below the IP layer, and let users do what they want at the higher layer, not be on your IP service. It may mean offering quality of service options other than best efforts. It may mean offering committed information rate services to those willing to pay for them. Because that way, they may be able to offer premium services. I think about audio services, I think about being able to do music in a high fidelity. It’s not huge bandwidth, but it’s sensitive to delay and packet drops. That’s not something that runs well. People do listen to streaming music, but it sometimes doesn’t work. Think about a band jamming at a distance. The delays on the Internet and buffering, make that pretty hard to really work. What if someone wants to offer a low latency, low loss service by which musicians in different places could still feel that they were playing together because the total latency was under 50 milliseconds, and the fidelity was good? That should be possible, but you can’t do that over the public Internet, it works or it doesn’t, but there’s no control. That’s the innovation I would like to see available, and make that happen by making the lower layer transport services available. It was the rule until 2005, so pretty much all the equipment can do that. Or, at least there is equipment that can do that for any medium.

Christopher Mitchell: I think it’s worth just thinking about that for a second. If you are a company that was specializing in making this service available. Right now, theoretically, if you wanted to do that, and you and I wanted to purchase this service, that company might have to literally build a fiber connection to our houses. Whereas, in the world that you are arguing we should live in, the world that we lived in before 2005, that company could have called up the local phone company, and said, “What is the price to get this service to that home?” They would have had a simple rate so they could get from it, and they’d be able to set up their service. I think we’re trying to move back to that world, and in my mind, it’s going to start, to some extent, with these municipalities, and really innovative, but small companies that are also doing this with some rural areas that I think are really progressive thinking about the future of the Internet. We’re going to make this available where one could get that level of access without having to figure out how to build a whole new network.

Fred Goldstein: That’s right, and that’s what should be. Think about it, you’ve got a piece of fiber, it has almost infinite capacity. There’s enough capacity on a strand of fiber going to a home to make everyone happy. Yet, the US Telephone Association, the trade group that represents the Bells, talks about constantly— the only real competition means facilities competition. Meaning, pull your own fiber. They want to keep the fiber to themselves, and if you want to offer a high layer service, you have to pull new fiber. That’s ridiculous, that’s like saying we've got to level the streets, level the houses, and put in a new highway in order to have Toyota compete with General Motors. That’s the way the Bell companies are thinking, and they’ve got an FCC that’s sitting there saying, “Yeah, I guess that’s okay, we should have lots of different companies pulling fiber.” That’s crazy.

Christopher Mitchell: The FCC, it's action, I think we would both agree, has mitigated some of the harm that the monopoly providers could do, but it’s also clear I think that if the FCC had gone with that direction, they would have had just tremendous pushback. I think that the fighting we saw on the net neutrality rules that they proposed, we would have seen a maybe tenfold increase of lobbying, and frustration and anger from the telephone companies if the FCC had moved in that direction.

Fred Goldstein: Maybe, maybe not. Because in fact, that was the rule until the 2000s. That was the rule that was assumed when the Telecom Act of 96 was written. All we’re really calling for is to restore the Telecom Act only 20 years ago, to go back to what it intended, and follow the rules there. The Bells are spoiled children. They will fight back because they are very lawyer heavy, very lobbyist heavy, but that’s the role of the government to regulate. The case is very strong, that’s what should have been, that there should be open networks. They aren’t providing open networks, and the Internet suffers. They’re trying to regulate the Internet at a layer that isn't designed to work that way. They are using the wrong tools, like I say, the arm without an elbow. Because, without the layer boundary between the underlying telecom network and the IP network above it, the flexibility of that bending elbow, the flexibility is missing. Vertical integration is always going to cause tension, and what they’re doing now is regulating the vertically integrated networks to make sure they are safe for Netflix, but that doesn’t mean they are safe for the next thing to come along. They are really harming the potential for future innovation and not really helping bring about competition where it’s needed.

Christopher Mitchell: I think you’re right. Well, in fact, I would say that I am strongly in agreement with you. I wanted to make sure people had a sense of, I think, why the FCC might not be moving this direction. Even if the FCC did, I could imagine Congress being much more vociferous in overturning it. That’s because I think the telephone companies would go to Congress and say, “It’s not fair.” Which is what everyone says to Congress, I’m sure. They would say, “It’s not fair, the cable companies can make all of this money and be closed, and we are just like them, and we want to be just like them, and we should be treated like them.” Just so people are aware of their political argument that I think is used against this ruling. Now, the other thing that you and I haven’t talked about I think is just that, it is so preferable to have bright line simple rules as a regulator to enforce, rather than very complicated ones. You’re proposing that we go back to a very simple rule that would be easy to enforce, whereas, what we have now is a more complicated rule, which is more difficult to enforce, and it’s great for lawyers perhaps that have these -- The firms that are representing the companies and what not. It’s really bad for the rest of us.

Fred Goldstein: So true. A simple bright line I think would be much easier to enforce, and much easier to regulate. I don’t think Congress would change it. The Telecom Act of 96 assumed just that kind of bright line. The Telecom Act of 96 did have a distinction, the FCC stretched the law to get rid of it, and Congress is so locked in now, they would not be able to overcome their own inability to act just on behalf of the Bells. I think the law wouldn't change. The FCC would be acting within the regulatory authority to open up the networks. They are just afraid because they don’t want to deal with the lobbyists, and they don't frankly -- many of them don’t understand it anymore. People who were there in the days of computer inquiries are gone. People who understand the importance of layered networks are gone. They're looking at Internet from the top-down without understanding the dynamic complexity of the underlying phenomena. They think it’s simple, but it’s black magic. When David Copperfield goes onstage in Las Vegas and make something disappear, it looks easy, but he did a lot of prep work for that to happen. The Internet is magic. The Internet is based upon illusion, and the illusion is important. That’s the key. The FCC really think they are sawing a lady in half and putting the lady back together. They don’t realize there is a mirror there, and net neutrality is all about ignoring that there’s a mirror. Go back to the old rules, we’d have neutrality, we'd have neutrality because if ISP 1, didn’t let you see your Netflix, ISP 2 might. They might charge $5 more because your average use is higher, but they'd make it available, and everybody would make money. Maybe somebody who doesn’t watch Netflix would get a $20 a month service, and somebody with a heavy video service habit would have to pay $5 a month. That’s what it takes, that’s what it costs. Instead, we have this one-size-fits-all model that almost going back to the old telephone monopolies. Where the telephone network was regulated, and everybody had the plain black phone, and paid an extra luxury tax for a colored phone, but the phone was a phone. The FCC is trying to regulate the Internet like a phone. They just don’t understand how the Internet works. You’ve got to really focus on what you can do easily. The FCC understands wire. They don’t understand IP, they shouldn't be the ones to regulate IP. By the way, that doesn’t mean the Federal Trade Commission, if it’s not the FCC’s common carriage, the Federal Trade Commission has the right to regulate abuse of monopoly power by non-common carriers. The FCC has snatched that away from the FTC. There’s, right now, a big question about privacy on broadband networks because, again, this was the FTC’s expertise. They are the expert agency for privacy, but the network neutrality ruling declared that these broadband Internet access services, are now common carriage, and therefore, they are the FCC’s exclusive bailiwick, and the FTC lost their authority to regulate the privacy aspects of these networks. Now, the FCC is coming in from scratch trying to apply what little they know about privacy, which basically has to do with telephone bills, (it’s really their job), and they are trying to apply the rules for telephone bills to the Internet.

Christopher Mitchell: Right, I seen a lot of concern because, I think, this is a significant area where we need to pay good attention to. It’s not clear to me that the FTC was doing a good job, but that doesn’t mean that I think the FCC is necessarily going to just come in and solve all the problems. We’ve run long once again, and I think we’ve covered a lot of topics really well, and I really appreciate your time for both of these episodes.

Fred Goldstein: Well, thank you Chris, I’ve enjoyed being here.

Lisa Gonzalez: That was Chris and Fred Goldstein, principal at Interisle Consulting. We have transcripts for this and other community broadband bits podcasts available at Email us at with your ideas for the show. Follow Chris on Twitter, his handle is @CommunityNets. Follow stories on twitter where the handle is @MuniNetworks. Thank you to the group Roller Genoa for their song “Safe and Warm in Hunter’s Arms” licensed through Creative Commons, and thank you for listening to episode 216 of the Community Broadband Bits Podcast.