This is episode 213 of the Community Broadband Bits Podcast. Fred Goldstein, author of "The Great Telecom Meltdown" and "ISDN In Perspective," joins the show to discuss the history of the Internet with Christopher Mitchell. Listen to this episode here.
Fred Goldstein: Anything a computer did, anything you did over the phone was enhanced, unless it was plain voice.
Lisa Gonzalez: This is episode 213 of the Community Broadband Bits Podcast from the Institute for Local Self-Reliance. Welcome. I'm Lisa Gonzales. If you're listening to this podcast, you understand the importance of Internet access to education, economic development, and all around quality of life. Have you ever thought about the nature of the Internet? In this episode, Fred Goldstein joins Chris. Fred is a principal of the Interisle Consulting Group and shares his extensive knowledge of the history of what we now know as the Internet. Fred gets in to the technology, policy and regulatory actions that brought us to where we are now. This is the first of two interviews with Fred who has written two books about telecommunications, and who, in addition to his work at the firm, is a columnist for TMCnet. Now, here are Chris and Fred Goldstein, author and principal at Interisle Consulting.
Christopher Mitchell: Welcome to another edition of the Community Broadband Bits Podcast. This is Chris Mitchell, and I'm here today 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: I'm glad to have you on. I've been corresponding with you for many years, definitely more than five, and I've always been impressed that your level of knowledge and history with the nuances around the Internet and past telecommunications regulation. For people who aren't familiar with you, can you tell us a little bit about your background and why they should listen to what you have to say?
Fred Goldstein: Sure. I've been the telecom area for over 40 years now. I started off ... As a kid, I started off in ham radio over 50 years ago and was in broadcast engineering college. I did the college radio station. Help get an FM license for them, the chief engineer. Did a little bit of a broadcast work. Then, later, became involved professionally in telecommunications consulting in the 1970s. I worked at a regulatory oriented consulting firm where we watched how the states and FCC regulated telecom. Mostly this had to do in those days with telephone rates. That was the big money. There's still a lot of rented PBXs and other terminal equipment. A lot of Centrix service. It was very voice oriented. Though there were of course some data circuits. I then went to work for Digital Equipment Corporation as an internal telecommunications consultant and worked in their network products group. I had also spent a few years as telecommunications manager at Bolt, Beranek and Newman in the late 70s when they were essentially running the ARPANET. I got to know the ARPANET people. Many of the luminaries of the Internet world were at BBN and I was fundamentally in charge of the phones, but I did pick up a lot of the ARPANET culture and network culture which Digital was carrying on in its own way. Digital was a very networked company even in the early 1980s. Its internal network technology, I have to say from an applications perspective was well advanced of the Internet today.
Christopher Mitchell: It sounds like you're very involved with computers, with DEC. Even if you were focused on telecom, you're with companies that were starting to really pioneer the computer revolution.
Fred Goldstein: Exactly. Really, my focus became the Nexus of computers and telecom. In the late 1980s, one of the big pushes was ISDN. I was Digital's representative on the ANSI Standards committee for ISDN Frame Relay and what became ATM networks. As these technologies were being developed, I was essentially in liaison with the telecommunications industry. We on the computer side of the house were a minority. We already had the Internet. It was not open to the public, but we were using the Internet in those days. Companies were networked in those days. We were networked better than most. Other companies had networking. We needed public networks to give us the interconnection we needed. The Bell companies were really having a problem with this. They had so much narrower focus notions. There was some tension there, but that's what the standards process is about. We did make some progress. Then in the 1990s, as the Internet became public, the perspective of the world changed. Everybody started noticing this technology, that there were a lot of issues that had happened in the meantime. One of the key things, of course, I was working in the field in the early 1980s when two very critical regulatory and legal changes happened to the industry.
Christopher Mitchell: Let me guess. Is this computer inquiries?
Fred Goldstein: That's right. That was the number one. People think divestiture. That was number two, in importance. The computer inquiries are what made the Internet possible.
Christopher Mitchell: I think briefly, because this leads into our question of what is the Internet. Why are the computer inquiries important? For people who are just wondering what is that. What is it?
Fred Goldstein: Okay. This goes back 50 years, 1966. Computers were starting to get used inside the telephone network. The first computer controlled switching systems were being used. The original 1ESS. Companies were connecting their sites together. They were using computer terminals, what was called teleprocessing in the IBM world. Banks were coming online, and a lot of businesses were making some use of telephone and computer together. Modems did exist, but you had to rent them from the phone company. They were an expensive business luxury. Time sharing computers. I first saw a Dartmouth Time Sharing in 1965. Over that next decade, it spread very widely until personal computers rendered it obsolete. The idea of every school having a computer terminal was catching on in those days. Computers had to use the phone. The FCC then had to draw lines. When is the computer part of the network? When is the computer a re-user of the network? That was really the fundamental question.
Christopher Mitchell: Why is it important to make that distinction?
Fred Goldstein: It's a very critical distinction, because when the computer is part of the network, the phone company owns it. They're providing a computing service, versus they're just carrying the traffic and letting you own the computers and do what you want and the phone company using its natural monopoly to provide what it provides best, which is just raw transport.
Christopher Mitchell: Great. Okay.
Fred Goldstein: They were trying to capitalize. They had transport. What if they could provide services with their computers? They could use computers to do more than simply run the network. The computer inquiries between computer inquiry one, 1969, divided it in three. There was pure telecommunications, and pure computing, and range of hybrids in the middle. When a phone company wanted to do something that was hybrid, they were supposed to go to the the FCC to figure out what to do. The first computer inquiry, the FCC, really didn't solve the problem. It essentially said, "Come to us when anything isn't clear." That didn't work.
Christopher Mitchell: Just to be clear. Why would the FCC have to involved? Is it because Bell might use its monopoly to disadvantage the companies that wanted to do whatever they wanted to do?
Fred Goldstein: That was certainly a big part of it. Also, remember, this was 1969, just happened the Carterfone decision that legalized privately owned terminal devices, your own telephone set, your own modem and answering machine. These had just been authorized, but the terms, the original terms until the mid-70s were very difficult. It was not widespread. The idea was that the telephone network was a closed shop. Since the terminals belong to the phone company, computer terminals had this interface, RS232. The modem was the phone company and the mouthpiece where you talked was RS232. The phone companies wanted to extend what they owned versus what the customer controlled. There was always this tension, because the network was a monopoly. The more the network did, the less you could do for yourself.
Christopher Mitchell: Great. As you were saying that, I just couldn't help but think about the movie War Games with the receiver. I could say, I think for many of us, that was our first exposure to it who didn't grow up with it.
Fred Goldstein: Great. Joshua. Yes. There was a lot of this time sharing going on, and the computer inquiries made this possible as did the Carterfone ruling in 1968 actually which allowed attachments to the phone network. During the 70s, we had ... Dial-up modems were clearly allowed. Computer II was a ruling that came out around the end of 1980. It was absolutely seminal. It really changed the nature of the network. It said that the telephone companies were only allowed to provide under tariff on a month to month rental basis, or service basis "basic service". A basic service was one that wasn't "enhanced." Essentially, an enhanced service was the content. Anything a computer did, anything you did over the phone was enhanced, unless it was plain voice. People in the computer business have no idea that long distance telephone calls were an absolute monopoly of the Bell System even tough independent phone companies interconnected interstate through AT&T Long Lines in the 1970s. Competitors started ... MCI and Sprint started dial-up competition in the late 70s. The way you accessed it, if you had an MCI account, MCI started this. It was called Execunet. It's not as if the FCC knew they were giving permission. MCI was called a law firm with an antenna on the roof for good reason. MCI figured out. They filed a bunch of tariffs for private line services and worded them in such a way that they were able to piece together a long distance telephone call service. Pick up the phone at one end, dial your local MCI access number, get a second dial tone, enter your account number, get a third dial tone, dial the number you want to call, and the call then travels across MCIs network. They pick up a local company phone in the distant location. Dial out and complete your call. Long distance competition. The FCC hated that, and spent three years litigating it. Finally, in 1979, coming out with a rule for it that MCI legal, but it also meant they had to pay something. They didn't want computers to be on the same rule. Anything that looked like MCI Execunet just dialing voice was subject to one set of rules that everything else they knew of they called an enhanced service.
Christopher Mitchell: Basically, you have MCI, which I just want to note for people. MCI led to so much interest. There's so many great companies that were started by people who came out of MCI. I think it's often unremarked on just how impressive MCI was and how it's impacted our current lives. MCI basically develops a service that looks and feels like long distance, but technically is not. The FCC says, "Well, we don't want to continue this, but we're going to set up this other regime for computers." Is that more or less accurate?
Fred Goldstein: Essentially, they said, "MCI, you can continue to do this if you pay a contribution." Because the assumption was, in those days, long distance subsidize local. Long Lines kicked back lots of money to the local phone companies. They made MCI kickback money to the local phone companies by creating a special set of rules for the lines used by long distance companies to complete calls. They call those exchange network facilities for interstate access. That's still the rule, so-called access charges are still on the books and carriers still have access circuits. They're called switched access circuits. That was created in 1979 as a way to make sure that MCI and Sprint who's doing the same thing, and others were coming along, copying them, were at least making a bigger contribution to the local phone companies. That was really the idea. Just as that was happening, the computer inquiries said, "You can hookup anything." These computer networks were not local. Computer networks were interstate, non-local. Phone company people who wanted that to be treated as a long distance call. Many times, that came up. The FCC very wisely said, "No. Those are enhanced services. Those are not access lines. Those are exempt lines. They're exempt from access charges, because that's an enhanced service." That was the beauty of Computer II. It also said this critical if the phone company offers an enhanced service, they have to offer it through a separate subsidiary. Cannot be tariffed. Has to be a separate subsidiary. Any basic services that are used by the enhanced service have to be made available on the same terms to competitors. All the private lines, data lines, all the least circuits had to be made available to competitors. That's what opened up the market so that the Internet and ... First, it was online data services, like CompuServe and Timeshare. Those sorts of services were all enhanced services. They could order phone lines and the phone company couldn't deny it. They cried in the beer all the way to the bank, giving them least lines, giving them dial-up circuits. These companies build huge networks, paying the phone company for basic service.
Christopher Mitchell: Great. Ultimately, leading to families like mine to having two phone lines in my home.
Fred Goldstein: Absolutely. Come the 1990s, the second phone line. Talk about crying all the way to the bank. There's a boom in dial-up, because that was the only way to reach the Internet for most people. Yes, the phone company sold a lot of phone lines that way.
Christopher Mitchell: Let's tie this back in. You mentioned AT&T breakup was the other important point. Let's just cover that, and then tell me how old that led to the Internet.
Fred Goldstein: Sure. The AT&T breakup essentially changed the nature of the industry from having one huge company that owned 82% of the U.S. telephone lines to having 7 companies that were the local Bell companies, and AT&T was left with the Long Lines business and its manufacturing businesses that were all competitive. You had a competitive AT&T competing on an equal playing field with the other competitive companies like MCI and Sprint. You also had all the terminal equipment going to AT&T that had been the Bell companies. The terminal equipment business was totally deregulated and very competitive. The Bell companies were left in charge of the "natural monopoly", the local networks only. This did crate a lot of competition in the areas that were left open to competition. It left the Bell companies with more power than people realized. Having that monopoly, that was the squeeze. They could charge those access charges to the long distance companies, and they tried to get the rule changed to charge access charges to the enhanced service providers as well. That was nicknamed the modem tax and was stopping congress in 1988.
Christopher Mitchell: Okay. We have this system with the 7 local Bell companies. How does that lead to the Internet? Why is that important for the Internet?
Fred Goldstein: What's happened to the Internet was not so much that there were 7 Bell companies. What was more important was that Computer II said that the Bell companies had to provide their local facilities as common carriers to anyone who asked. They couldn't turn down a competitor. The Internet existed using the NSFNET backbone, but that was not really not opened to the public. It had an acceptable use policy. It was not allowed to be used for commercial purposes. In 1992, that rule has changed. Commercial ISPs popped up. The online service providers began a multi-year process of morphing into ISPs. AOL was not an ISP in 1992, but they existed.
Christopher Mitchell: Right. I remember that quite well. I remember when they started adding Internet access. People that I knew would be confused and think, "Why would I go on this Internet thing? I have everything I need from American Online." Which was basically chat rooms and maybe like a little bit of browsing opportunities.
Fred Goldstein: That's right. They added e-mail, I think in version two. Web browsing in version three. It was still a terminal to host system. It was ... You were dialing in as a computer terminal. Your PC rent a program that was simply displaying material that their computers in Virginia were doing all the computing, and all the actual Internet access went through Virginia. AOL and their competitors morphed into ISPs, but the idea that there was the glory days of totally neutral transport didn't exist. AOL provided the applications AOL wanted to provide. Application by application, it was added. You may remember net news. Still exists. The NNTP net news. AOL added that as did some of their competitors. You had this morphing overtime, and this is all done because everyone could rent phone lines and use the Internet and it was not treated as a long distance call, much to the chagrin of the telephone companies. It blew their business model. They did sell a lot of phone lines in the 90s. The Computer II rules also said, at those days, if the phone companies had DSL, which came out in the late 90s, they had to make the DSL available, again, to anyone who wants to use it. Not only did Verizon Online had access to Verizon Internet. Those days, of course, it was called Bell Atlantic. Bell Atlantic had a DSL service, a retail service, but the Bell Atlantic telephone companies had to make their raw DSL packet transport available to other ISPs. Earthlink was a big user for instance. There were many local ISPs could use DSL.
Christopher Mitchell: Let's make this very plain. I think that I have a sense, and you can correct me if I'm wrong. Now, I lived in Rochester, Minnesota at the time, and I think ... I can't remember. I think it was Infonet was the ISP there, a local company. The Bell companies, as you were saying, I think they were frustrated because my family was paying money to Infonet to connect to the Internet. Even though we also paid the telephone company for the lines that we're using and Infonet paid the telephone company for the lines they were using. There's also this margin that we were paying them. The Bell companies wanted that margin. Now, we had the choice of many different ISPs, but the Bell company would have preferred if we only could go through them, if they had the power to do that. Is that correct?
Fred Goldstein: You said something very important. You had a choice of many ISPs. That's right. The Bells wanted to be your ISP so that even though they hated the Internet with a purple passion, they'd rather be the one to take some business and do a crappy job of it, because they hated it. They didn't want third parties to be making that money. They didn't want ISPs to exist. They waged a war on ISPs. This really was political, because the Clinton administration and the FCC under Chairman Kennard and Chairman Hundt were pro-competition pro-ISP. The minute the new FCC came in after the Cheney-Rove regime took power in 2001, they put in a new FCC under Chairman Powell that was anti-ISP and really took its orders from the Bells. They wanted to take away your choice of ISPs and put the ISPs out of business. One of the things they did was say, "One by one, you can't ... If you're an ISP, you can't use fiber. Only the copper DSL." Because they wanted to encourage fiber, was one excuse. That if the Bells had a "regulatory holiday" from ... Having to share the fiber, maybe they'd upgrade their network as they had promised 10 years earlier to do anyway.
Christopher Mitchell: I have an example of that, which is when I started here at the Institute for Local Self-Reliance, we got our Internet access from a local company called ipHouse, which is a terrific company that still does lot of data hosting and that sort of thing. We could only get the slowest DSL available. We couldn't get the faster service. If we had gone to Quest at the time, I think that was before CenturyLink bought Quest, then we could have gotten a faster service that's only available through Quest. That was because Quest was denying the faster DSL service to this local independent company. Now, if the FCCs reasoning was correct, by allowing Quest to do that, Quest would have had an incentive to deliver fiber throughout the Twin Cities metro area, but they didn't do that. They just basically delivered a slightly faster DSL with poor customer service. In many ways, we screwed and we ended up on Comcast, because the last thing we wanted to do was to get a slower service. If we couldn't have a good local service, we were going to get the fastest service possible. That's what we did.
Fred Goldstein: Of course, and cable was the winner. That's the thing. Cable was offering a better service than DSL. The phone companies didn't really want to invest too much in fiber, but they used that as an excuse. An original rule was they couldn't share fiber, but they still in 2001 had to share the DSL, but they were sharing the low speed DSL. Come to 2005, the FCC ... This is critical. The FCC dropped the rule. In 2005, they literally revoked Computer II, repealed it and said that the underlying transport facilities do not have to be made available to competitors. The Bells could offer a vertically integrated information service. Information service, being essentially the new name from the Telecom Act of 1996 for what had been called an enhanced service. They're almost the same thing. The Bells could offer their Internet service vertically integrated, meaning that you didn't have the access to buy the bit transport as a competitor. They only made the retail available, not the underlying wholesale. That shutoff access to the competing Internet service providers. No more competing local ISPs. They didn't shut them all off right away day one. They made all these noise to the FCC how not requiring them to do it would give them an incentive to commercially upgrade their facilities. After all, they wanted to compete with cable. In fact, they shut off the ISPs. Over the next few years, they basically grandfathered existing ISP contracts on to the lowest rate plans, and those lowest rate plans became obsolete. Then they phased those out. Then they made this ... In some cases, an offer that an ISP could resell the Bells own ISP and take over the billing as a sales agent. It's like DISH network where you can still be an independent DISH dealer, but your'e selling DISH. You could still be an independent "ISP seller" but you're selling SentryLink, or selling AT&T. That's not competition. That's just agency. That's gone away. A meaningful competition.
Christopher Mitchell: You had made a point that the Bell folks really hated the Internet. I think we just need to explain that for a second. This gets down to something that has shown up occasionally on muninetworks.org with the discussion between people who might call themselves Bellheads and people who call themselves netheads. Just tell us why the Bell folks were so angry and distrustful of the Internet.
Fred Goldstein: It goes back to how the telephone company business model works in the U.S. Has nothing to do with charging based upon cost. If you're a real business, you want to do cost accounting. Figure out how much it cost to deliver a product and mark it up by some amount to make a profit. If you're a Bell company, the rules are different. They were on an accounting rule setup for monopoly utilities that said they took in money and they spent money, but there was no connection between the products. They could lose money in one place and make an absurd profit in another. Long distance was setup to be profitable so that they could have low rates for residential service. They used to brag everything is there for grandma. In fact, when Carterfone came out, again, they cried in their beer profusely, that if you're allowed to buy your own telephone set, grandma can't afford a phone, because the charges they mark up their profits on luxury, unnecessary items, like color telephones that aren't black. There was a charge for that. Princess phones. Light up dials. Right. Anything but a plain black rotary dial telephone was marked up. Touch tone was marked up absurdly as a service. Rotary dial service is basic. Touch tone could be $4 a month extra for the right to ... It cost them less to provide. They actually saved money with touch tone, but they marked it up because it was as luxury. Long distance was the number one luxury. AT&T Long Lines, and later, all of the divestiture, all the competing long distance companies had to pay those minutes of use, switch access charges, for every minute they delivered or picked up. Both ends of a call. If you were calling from Minneapolis to New York, then in Minneapolis, the long distance company would pay U.S. West or Quest. In New York, they'd pay NYNEX, Bell Atlantic, or Verizon, depending on when the company's name changed. The point is, that the long distance company always lost. The business model was long distance subsidizes local. Along comes the Internet, which has no notion of long distance. That's very threatening. Especially VoIP, the idea of voice over IP. Even sending an e-mail, even just doing any activity that you could use the phone and not be paying them a share of what they consider the value of the call was threatening to their monopoly business model.
Christopher Mitchell: Right. Although it's worth pointing out that the trade off for that, it's not just said AT&T was screwing the country. There was also the benefit that everywhere had affordable local phone service, more or less. Everyone had service. These are the trade offs we've made a long time ago. I think, certainly, AT&T did very well with them. It developed a certain culture and a certain power that they sure love to have. That level of control.
Fred Goldstein: That's right. In fact, the Bell companies, even today, do things that do not maximize profit. They simply do things for the sake of control, even if it loses money. They price services above the profit maximization point, because they'd rather not do it than lose the control they think is implicit in providing some services. It really is that monopoly DNA, the frog and the scorpion comes to mind.
Christopher Mitchell: Yes. No, I absolutely agree. Yes.
Fred Goldstein: There's a lot of scorpion left in their DNA. Yes, it's true that the cross-subsidies helped, but there are ways of doing that, and it's done today, that don't inhibit technological progress. Their model was to hold up technological progress in order to preserve the business model intact. Of course, they could have made money changing their business model. Their costs did go down because electronics ... The telephone network cost much less to run now than it did 50 years ago. Stringing wire still cost a lot of money. Stringing fiber is expensive. The electronics, the inside stuff, dirt cheap. A tiny fraction of what it cost back in the days when there was just thousands and thousands of relays needing lots and lots of maintenance and huge jobs to install.
Christopher Mitchell: That's where you also get into. They have all these real estate, which was paid for by a monopoly, by regulated rates. Now, they own all of these buildings in major cities that they really ... They can lease out. They can do whatever they want with, because the big computers they used to have making all these magic happen, it's reduced to one or two racks here or there it seems like.
Fred Goldstein: That's right. You can put a huge modern switching system in two racks. That used to take a whole floor or two of the building. Before that, remember, they had cord switch boards a century ago. Those took lots of rooms for dozens of operators in a major city to be plugging in cords. They still got those buildings, and they're starting to realize that they ... It's a big market for data centers. These buildings had wonderful air-conditioning, and massive power systems, and redundant power, and back up systems, and they're starting to realize that maybe this is useful real estate. They've been very slow to that party to even recognize the value of their buildings and use the empty space.
Christopher Mitchell: As we're running out of time, we're going to come back and discuss this more fully. In a brief synopsis Fred, just tell us what the Internet is and we'll discuss this in a later show.
Fred Goldstein: Sure. The Internet is not the wires. The Internet is an agreement. A voluntary agreement among network operators to exchange traffic for their mutual benefit. In other words, there are many Internet service providers. They're not supposed to be rules like the telephone network as a single worldwide network. An ISP is what it wants to be. ISPs make their own deals for interconnection bandwidth. ISPs offer the services they choose. That's why it works, and they're a layer above telecommunications. ISPs are computer operators who rent wires. Now, the trouble is, the FCC, by revoking Computer II, allowed the telephone companies and the table companies to claim to be ISPs and claim that their wire is the Internet, but it's not. Louis Pouzin invented Internet networking in France in 1972 as a way to interconnect networks when the local phone companies across Europe were building incompatible data networks. The idea was you can relay a packet across one network, and then from another machine computer, relay across the next network when two networks connect to one machine. That's Internet networking. It's a layered concept, and it's a voluntary concept. It's the antethesis of telecommunications, which is a regulated service that was designed around to monopoly utility model. They're both necessary. The Internet is the payload of telecommunications, and the FCCs model and the current "network neutrality model" totally misses that. In so, doing both telecommunications and the Internet are damaged. It's like an arm with no elbow.
Christopher Mitchell: With that, I think we're going to cut this show off. Thank you so much for coming on. Definitely, I hope people would keep an eye on the feed for when you're back and we can flesh that out more.
Fred Goldstein: Thank you so much Chris. I enjoyed being here.
Lisa Gonzalez: That was Chris and Fred Goldstein, principal at Interisle Consulting. For more, checkout the firm's website at interisle.net. We have transcripts for this and other Community Broadband Bits Podcasts available at muninetworks.org/broadbandbits. E-mail us at firstname.lastname@example.org with your ideas for the show. Follow Chris on Twitter, his handle is @communitynets. Follow muninetworks.org stories on Twitter where the handle is @muninetworks. Thank you to the group Roller Genoa for their song “Safe and Warm in Hunters Arms,” licensed through Creative Commons. Thank you for listening to episode to 213 of the Community Broadband Bits Podcast.