This is the transcript for episode 371 of the Community Broadband Bits podcast. In this episode, Christopher speaks with Sascha Segan, lead mobile analyst at PC Mag. They discuss our recent fact sheet, Pocket Guide to 5G Hype, and Sascha offers some corrections and helpful nuance about 5G technology and regulation in the U.S. and abroad. Listen to the episode, or read the transcript below.
Sascha Segan: No technology is really going to solve what is a regulatory and market structure problem more than anything else.
Lisa Gonzalez: Welcome to episode 371 of the Community Broadband Bits podcast from the Institute for Local Self-Reliance. I'm Lisa Gonzalez. When large Internet access companies begin discussing advances in technologies they're developing, it's important to maintain perspective and not get swept away by the inevitable hype. Over the past two years or so, 5G has been on everyone's mind, and big providers haven't done anything to curb unrealistic expectations. Recently we released our Pocket Guide to 5G Hype to help manage some of the claims that have been floating around that are a little overblown. Shortly after the release of the pocket guide, this week's guest Sascha Segan from PC Mag, contacted us to let us know that he disagreed with some of the content of our resource. Since Sascha has been covering mobile technology for what Chris describes as forever, we wanted to bring him on the podcast to explain what he disagrees with and why. In this interesting discussion, Sascha provides some great education on the intricacies of 5G millimeter waves and the technology and marketing campaigns surrounding them. He and Christopher also talk about fiber deployment and what it will take to bring high quality connectivity to all of the U.S. Now here's Christopher and Sascha Segan from PC Mag.
Christopher Mitchell: Welcome to another episode of the Community Broadband Bits podcast. I'm Chris Mitchell with the Institute for Local Self-Reliance in Minneapolis. Today I'm talking to a very notable guest, Sascha Segan, PC Mag lead mobile analyst. Welcome to the show.
Sascha Segan: Thanks for having me on. Hi.
Christopher Mitchell: You know, we were talking for a minute and I realized I've read your name a million times, but did I pronounce it correctly?
Sascha Segan: It's actually Sascha Segan, but a lot of people pronounce it the other way because they think of Carl Sagan, the astronomer, and it kind of connects.
Christopher Mitchell: I was thinking of Peter Sagan, the famous bicyclist. I just finished the Tour de France, you know, a week or two ago, so . . . Anyway, you are someone that's been around — I mean, so a little bit of context: I've been working in municipal broadband for 12 years. People think of me as someone who's been around since the beginning of time, I think, because many of them have only been in it for a few years. When I got started in this, you were someone that I thought had been around forever, and I think that's because you've been around forever. So, why don't you tell us a little bit about how you got into this?
Sascha Segan: Yeah, so I've been a technology journalist since 1997. I've been online pretty much forever, and I've always been interested in wireless. I've reported on things like Ricochet, which was this really primitive, interesting peer-to-peer wireless Internet system in the late nineties, and just how wireless technologies are transforming our society, transforming our minds, transforming the way we work, the way we live, the way we socialize. So come around 2004, PC Mag, which was at the time PC Magazine, was hiring their first dedicated writer on smartphones, and I had already been freelancing a bit for PC Mag before then.
Christopher Mitchell: Those are the Nokia, I'm guessing, right? I mean, back in those days,
Sascha Segan: Yeah, there was some Nokia Symbian, but actually the big players at the time were Palm and Microsoft. It was Treos, if you remember those, and Microsoft Pocket PC. And so I thought, yeah, this would be a great, growing, new area of computing to get into that's probably going to change the world. And so, I started there 15 years ago, and lo and behold, it did change the world. And so, I got to see the first iPhone and what that did. I got to see Android. I got to do the 2G to 3G transition, 3G to 4G, now 4G to 5G, and you could say that mobile kind of does dominate how people do computing in the world as a whole in 2019.
Christopher Mitchell: Right. That actually gives me the opportunity to reuse a joke that I love, which is "That makes you OG."
Sascha Segan: Yeah. Yeah. I actually had that as part of my Twitter bio at one point
Christopher Mitchell: Oh nice. I'm glad I'm not the only one who saw the — I used that just at random in a conference, and it killed. It was the best laugh line I've ever had, so it's etched permanently in my memory. So you I think saw a tweet that we put out about 5G. We put out something called a fact sheet, a 5G hype pocket guide. And I suspect it's because Gigi Sohn had retweeted it and you saw it, and you called us out on some factual — you know, I don't know if I'd say errors, but I'd say we glossed over key details in some ways that — I agree with your criticism. I think it's legitimate, and we're going to talk about that. But you know, you came back with some very powerful insights about what you really expect to see and some of the technology issues. And so, we're gonna talk 5G. We're gonna start by talking about the fact sheet I think, to try to get at some of these issues. You know, there's a number of engineers who listen to this show, and I'm sure that they cringe every time I simplify anything, which is something I frequently do for some of the audience. But let me ask you this, and I think this is actually where I really appreciate that you, you know, reached out on Twitter to say is because you actually read it. And I think a lot of people, you know, look at the headlines and then move on with their lives. And let me just ask if you would agree with sort of the bolded text, which is that we still need wires. Maybe not a hundred percent of the time, but I think you would agree with that, right?
Sascha Segan: Oh no, we absolutely do need wires. Wired broadband has so much more capacity even than the best 5G system. It's so much more reliable. I mean, we use the phrase backbone in the Internet world, and wired broadband is the backbone of all networking and will continue to be the backbone for the foreseeable future. 5G relies on fiber intensely. You can't have 5G without extensive widespread fiber. So these are two technologies that 5G is going to be able to reach some places where fiber can't reach because of physical or economic reasons, and at the same time, 5G is not going to be able to exist without fiber supplying the basic backbone, the big pipe that goes into the 5G radios.
Christopher Mitchell: I agree. I would actually — let me ask you, if I could push in on something you just said regarding fiber not going places. When you make that statement, I think you're probably thinking of a horizon of 5 to 10 years. You know, I think if we think longer than that, I expect that we will see fiber eventually reaching everywhere as long as we don't live in a horrible hellscape from climate change, you know, and things like that?
Sascha Segan: Yeah, I somewhat disagree because I do think there are two areas where there are real problems installing fiber and they aren't technical problems. They're in general economic and political problems that I don't see any time horizon necessarily curing. And one of them is truly rural installation, and that's a situation where it just costs costs a lot of money to build it out. The companies involved do not see the ROI, they don't see the return on investment, because the structure of our economy is about short term return. They don't see short term return in laying fiber way out into the farmlands. I don't see that necessarily changing. And then the other area is actually in some urban areas, the situation with property ownership and who can give Rights-of-Way across certain properties, who can allow for digging under streets, who can allow for access to conduits can often be so complicated and so politically fraught and so economically fraught that no change in the fiber technology is necessarily going to fix that. You literally have to go around the outside, and that's where wireless may come in to help solve that problem.
Christopher Mitchell: Without getting lost in this rabbit hole, I think that's why, in fact, I actually think we'll see fiber to almost every rural household before we'll see it in every urban area. I think the cable monopolies and that Right-of-Way issue will be a challenge. I'll send you our rural fiber map. I think that you may be interested in checking out how far the rural fiber's already gotten from the co-ops. But I do want to move on to "5G won't fix the broadband market" — that was another one of our bolded ones — "5G won't solve the digital divide," and then "there's no 5G race." So with our bold headlines, any quibbles?
Sascha Segan: So 5G will help the broadband market.
Christopher Mitchell: Okay.
Sascha Segan: The problem with the broadband market, once again, is more economic than it is technical. If we had, for instance, common carrier regulations on fiber, if we had a requirement for unbundling, if we had, what is it, UNEP the way we had in the late nineties —
Christopher Mitchell: Unbundled Network Element . . .
Sascha Segan: Platform.
Christopher Mitchell: Platform. Thank you.
Sascha Segan: And that's actually what they have in the UK. And so in the UK they have, as far as I know, two main fiber providers, but there's a requirement for UNEP. And so as a result, your average household has seven or eight different choices of competitive Internet provider running on those fiber providers, and that creates a really vigorous competitive, interesting, powerful market there that really helps UK Internet consumers. The problem we have in the U.S. is there's 15 percent of the population, that 15 percent rural population, who are going to say the problem is that we don't have wires going to our vicinities, but for the other 85 percent of the population, the problem is not that there aren't wires. The problem is that the wires are owned by powerful monopoly companies who absolutely control them and don't let anyone else play.
Christopher Mitchell: Right. Their control over their wires is only dwarfed by their control over Washington D.C. it feels like.
Sascha Segan: Exactly, exactly. And so, once again, using this idea of going around the outside of that monopoly, the wireless carriers are going to start distributing Internet access via 5G in some metro areas probably mid — like, they'll say later this year, but it's really mid-2020 and it'll grow beyond that. And it's not going to be a panacea because they are just introducing one or two more players that once again aren't common carriers, don't allow for a wide variety of virtual network packages. And in wired broadband, where you've seen one or two more players enter, like places where Fios competes with cable, it's been a little better, but it hasn't been transformative. And I think as we see Verizon and merged Sprint/T-mobile come in with 5G broadband offerings, it's going to be a little better, but I don't think it's necessarily going to be transformative because you really do need three, preferably four players in a place offering similar options to get that kind of competitive ferment that you really need to break things open.
Christopher Mitchell: So what I'm expecting, and I agree with everything that you just said, is a dynamic that's not unlike the third pipe discussion we had with Wi-Fi 15 years ago. When I look back at that, which actually was a little bit before my time, but I've talked with a lot of people who, you know, very active in that. One of them, a friend of the show, Travis Carter, who runs U.S. Internet — for people who listen to the show regularly, drink — because you know, he hq`s a very pithy way of saying a lot of things. One of the things I think about is the way wireless works really well, perhaps in places like Sacramento, certainly in places like Tucson. You know, I've been through Sacramento. I don't remember exactly the level of trees and things like that, but I'll tell you here in Minneapolis, in St Paul, you know, I think Wi-Fi citywide was really failed from two major issues. One was speed. You know, it just could not deliver the speed to the home, and I think that this 5G crushes that. It's going to be amazing speed. But the second piece is reliability. And my concern is that people will be excited about it, but as soon as they're watching, you know, whether it's the MLS championship, the Super Bowl, or whatever, and they're their Internet connection's wavering because the wind's blowing the wrong direction, they're going to go right back to cable. That's kind of what I'm expecting.
Sascha Segan: Yeah. And that has been my — like, I've been very disappointed so far in a lot of the performance aspects of the first wave millimeter wave 5G devices. I feel like the carriers and the manufacturers have made a lot of promises about millimeter wave, I don't believe that those promises have played out on any axis. They promised 1,200 to 3,000 foot radiuses. I've seen 600 foot radiuses. They promised certain levels of performance. I've seen devices that overheat constantly. They've promised beam forming to get through windows if not walls. I still see 60 - 70 percent signal loss through windows. So I'm very, very frustrated with what I've seen there so far, but can we go to my biggest quibble with your fact sheet?
Christopher Mitchell: Sure. Let's do it.
Sascha Segan: So my biggest quibble with your fact sheet, and this is something that a lot of people do around the country, is they say 5G when they mean millimeter wave and they say millimeter wave when they mean 5G. And those two things are not actually the same.
Christopher Mitchell: So what do we mean when we say millimeter wave? Just to be very clear as we get into this.
Sascha Segan: What are we talking about when we talk about millimeter wave?
Christopher Mitchell: Yes. Specifically.
Sascha Segan: When we're talking about millimeter wave, we are talking about a new, very high speed, short range, high frequency network technology that is going to require thousands of new small cell sites probably on lampposts and the sides of buildings placed either every 1,200 feet or every 2,400 feet, depending on who you believe, that is easily capable of multi-gigabit speeds but has trouble with rain, trouble with trees, trouble with windows.
Christopher Mitchell: And what's the frequency range here?
Sascha Segan: The millimeter wave in the U.S. tends to be from 24 gigahertz up right now into the high 30s, and it can really be anywhere from — I think the bands considered millimeter wave are anywhere from about 8-ish up to about 100. But we're really talking the 20s and 30s of Gigahertz in the U.S. now.
Christopher Mitchell: Sure. And so this I think is where I told you I'd set a little bit of context for us, and let me ask you one more question before I talk a little bit about the context that we were thinking of when we released the fact sheet. And my understanding is if we're talking about the super fast speeds that we're talking about for 5G — which generally means, let's just say, competitive with and better than modern cable in urban areas, so we're talking about multiple hundreds of megabits and gigabit plus — that that's only millimeter waves.
Sascha Segan: No! No! That's not true.
Christopher Mitchell: Okay, so there you go. So correct me then because that is my understanding.
Sascha Segan: Okay. It's not true. That is just about the way that the FCC has decided to assign frequencies in the U.S. A hertz is a hertz. A millimeter wave hertz can transfer the same amount of data as any other hertz on the spectrum. The issue is that there were large blocks of unassigned frequencies up in millimeter wave because it was considered trash bandwidth that no one could properly use, and so because it was open and unused, the FCC found that really easy to auction. And the carriers said, "Okay, we're going to start installing 5G on here." But, in Europe for instance, you're seeing 5G start to deploy on what's called the mid bands, which in Europe are typically between 3 and 6 gigahertz.
Christopher Mitchell: Okay, and this is something David Burstein also just jumped on Twitter on Friday or Saturday to point out. He also had this same issue with me.
Sascha Segan: Yep. So in Europe, and South Korea actually, the regulators opened up several hundred megahertz of spectrum between 3 and 6 gigahertz. Now this spectrum has similar propagation to existing cellular frequencies. It has definitely similar window penetration capabilities, okay wall penetration capabilities. It's like the high bands of the cellular network, but it's definitely much easier to work with than millimeter wave. And you see carriers like a Sunrise in Switzerland offering unlimited home internet packages for 40 - 50 euros a month on this network with speeds of between 300 and 500 Megabits a second, which is not bad at all. And when you consider that the vast majority of people signing up for cable or even fiber are signing up for packages that are capped at either 100 or 300 — they're not buying those expensive gigabit packages — that is definitely competitive. The problem is the U.S. Government did not open up enough of these mid band frequencies here in the U.S. to make that a viable nationwide business plan, and I'm very frustrated with the FCC for doing that.
Christopher Mitchell: Let me ask you a quick question about the technology. Is this premised in some sense on having a lower penetration? Because this is, I think, one of the larger questions about wireless that I worry about, particularly fixed wireless in the U.S., is that there's a sense that you can deliver a lot, but it's premised on not having a lot of penetration, you know, using up the spectrum.
Sascha Segan: Unfortunately, I think that's something that hasn't been tested yet. The carriers all promise it'll be fine, but we have not seen — Because we haven't seen any real, widespread millimeter wave consumer premise equipment for home broadband yet here in the U.S., nobody's really been able to test that. Maybe it's a concern, maybe it's not a concern. There is a feature in 5G which can really help here which hasn't been implemented yet, which is called network slicing. Network slicing, which is coming probably late next year, let's a 5G network segregate bandwidth for certain kinds of users. So, the 5G network could say, for instance, home users get guaranteed bandwidth first and all of the cell phones on the street get what's leftover. And that'll definitely help. Verizon plans to have a homogenous, mixed-use 5G network where it's the same radios serving both homes and mobile users, and they need that network slicing so that they can guarantee a certain QOS, quality of service, to the homes while there are crowds of mobile users wandering in and out. And they're not gonna have that QOS until network slicing happens.
Christopher Mitchell: Okay, so that's very helpful. So if I was to sum up, our criticism of 5G is overly broad and focused on not so much the technology but a combination of the political decisions that have been made in the U.S., in which we are not likely to see those mid band spectrums anytime soon but if they were available we could see this very high quality speed without it being —
Sascha Segan: Okay. So wait though, because one of the things that Gigi and me and most importantly Jessica Rosenworcel are pushing for heavily is that there is the possibility of opening up a lot of mid band spectrum in the U.S. It's called C-band. C-band —
Christopher Mitchell: Right, I'm signed onto those comments.
Sascha Segan: Okay. There we go. So C-band runs from 3.7 to 4.2 gigahertz. That makes it 500 megahertz, right? And 500 megahertz divided between what's inevitably going to be three wireless carriers in the U.S. at the end of all this gets each of the wireless carriers a pretty decent chunk there. They can carry some serious traffic on there. Unfortunately, C-band is currently occupied by a constellation of foreign satellite firms and the U.S. Navy, and figuring out how to free up c-band and whether the satellite firms should just be paid off out of the public purse or whether some of the spectrum should be confiscated or whether there should be an auction or whether companies should be able to just buy it from the satellite firms and how much the military should be required to get to give up, that is all in discussion right now. I think those are the comments you're following, right?
Christopher Mitchell: That's right. Michael Calabrese at New America as well as Public Knowledge are two of the places that I think people could go to learn more about them. They've been major commenters and focused on that. So very good point. Now I wanna I wanna talk a little bit about 5G on even lower spectrum, which is what we're expecting to see from Sprint for as long as it's around — you seem to not expect it one way or another to be around much longer in its current form — and then T-Mobile. Which is to say that I think in coming years, we will see people in rural areas having phones that say 5G on them, but probably not experiencing what they would experience if they took those same phones into a major urban.
Sascha Segan: Right. So as I said earlier, a hertz as a hertz, right? And T-Mobile intends on implementing 5G on a 30 megahertz sections of their 600 megahertz spectrum. Now 600 megahertz is very low band. It's old antenna TV channels, so it's got great penetration, it goes for miles and miles and miles, it's terrific for rural, but it's only 30 megahertz. And so, you can't transfer that much data in an individual channel on this spectrum. It's not going to be that fast. It's going to be as fast as, you know, decent 4G. So why is T-Mobile even bothering is the question, and I think a lot of that has to do with other promises that 5G has down the road, like promises involving latency, like promises involving network slicing and guaranteed QOS. There's a lot of stuff where these other features tend to combine into things like industrial automation and agricultural automation and massive numbers of agricultural sensors — because 5G networks are able to address millions of devices per square kilometer — and a lot of stuff which could be really good news for rural industry and rural business, rural health care, things like that, but don't at all address the question of home broadband.
Christopher Mitchell: So let me ask you — if I'm on a farm for instance, and in North Dakota, more than 70 percent of the land mass has fiber today. 50 percent of South Dakota that's settled with people has fiber. I have Wi-Fi set up intelligently through my fields and things like that. Do you actually see a purpose for these sorts of wireless technologies, or would I just do everything using, you know, like, Wi-Fi and presumably at a lower cost than having to pay the wireless providers?
Sascha Segan: One of the things people are talking about with rural 5G is the future of private networks. And with a private network, you would as a farmer license some of this bandwidth for your own use. Where I was talking to — who was I talking to? — John Deere about this. And so this becomes a situation where — so you have fiber to the edge of your property because you're a farmer, and you have the option of either setting up one 600 Megahertz 5G tower on that fiber line at the edge of your property to manage all your, you know, 25,000 agricultural sensors and fleet of robotic autonomously controlled threshers, or you can grid it with 5 Gigahertz Wi-Fi by every thousand feet or so and have to manage all of those poles and all of those routers and all of those units and that whole mesh. And so, there does become a trade off of you're a lot more autonomous if you're using Wi-Fi, but the solution may be a lot administratively simpler using low band 5G.
Christopher Mitchell: Great. There's very few people like you that ask these questions of people who can answer them without a lot of hand waving and in generalities that aren't super helpful. As we're running out of time, I wanted to note something that I think our conversation has covered well enough and that's that, you know, we say that all the 5G antennas would need to be connected by fiber. You know, we said that even though I'd previously interviewed Doug Dawson, a consultant who's very good, writes the POTs and PANs blog, and he had mentioned that he thinks the low earth orbit satellites are going to be connecting a lot of these rural towers. And so, I don't think we have to spend a lot of time on that, although I would defend what I'm saying. And this is where the context comes in a little bit, I think, because you're providing very good context of 5G from a technical perspective. When I think about the sales job that's been done on local elected officials, 5G — I'd say "5G" maybe in quote marks, which is to say like super fast speeds that are going to be, you know, allowing you to do everything you want to do, no congestion, just, you know, basically gigabit-type speeds. And I would stand by that I think — and actually I think I'm going to amend the fact sheet to include this context explicitly so people understand it — that even though that is 5G and it may hit some of the technical specifications, it's kind of like 5G in the sense that, you know, if you go back to, I think it was 4G when T-Mobile came out with what they called 4g, but it was really just 3G with, like, one or two bells and whistles on it. Tell me I'm wrong. Or do you think that's pretty close to being accurate?
Sascha Segan: What Verizon is installing right now is definitely an order of magnitude faster than their 4G, but it is so early and so — it feels like they're using us all for a public beta. It doesn't actually work in a lot of the ways they say it will work in the future, and so it's different from when T-Mobile installed that fake 4G, the technology wasn't even theoretically capable of being world changing. This technology is, but Verizon is no where near the actual practical capability of what they're promising. They are dramatically over promising what they're able to deliver today. But if there's one thing that I want to bring home that I think that you and I can agree vigorously about and that I hope we back each other up on Twitter about a lot, is that no technology is really going to solve what is a regulatory and market structure problem more than anything else. That the digital divide is not going to be solved by a new wave, it's not going to be solved by a "G," it's not going to be solved by magic satellites in the sky, when the real issue is a lack of regulatory backbone, when the real issue is a small providers and community and municipal providers getting beaten down by unjust laws designed to maintain monopolies. This is an economic structure problem, not a technological problem, that we have in the U.S.
Christopher Mitchell: Oh, absolutely. I'm totally with you on that. If you have a couple more minutes, there's one other piece that I feel like I don't know who else I'd turn to to get this answer. You mentioned that 5G plans don't have caps, and I have to admit, I didn't do as much research on this as I probably should have in fact checking it. My expectation was that, you know, caps are fundamentally an irrational way of dealing with congestion anyway. I think caps are mostly just about patting the revenue, and so in my mind, of course it's going to have caps because Verizon and AT&T want to figure out how to pull money out of bandwidth hogs like you and me. So I assumed they would have caps, but you're saying that they don't have caps.
Sascha Segan: I mean, at the moment they don't. The question is what happens when people start using a terabyte a month, and even there, you see the cable and fiber providers are putting in caps. Like, you've been following the whole Comcast saga. So, will there be caps, but will the caps look more like the cable company caps than the cell phone caps? That might be one way to look at it. You know, definitely every network has limited capacity. One advantage that fiber carriers have of course is they can pull another line if they need to.
Christopher Mitchell: Or potentially even access additional light waves within their existing lines too.
Sascha Segan: Right. With wireless carriers, you either have to wait for better order modulation to be available or you have to buy more spectrum, so it's definitely easier to congest wireless than it is to congest fiber. But right now, it's looking like the high frequency 5G systems have capacities similar to cable.
Christopher Mitchell: Great. Thank you so much for coming on. This has been great, and I'm really just glad to have a high level conversation with someone that's been following this so closely. So, thank you for your time today.
Sascha Segan: Thanks for having me here.
Lisa Gonzalez: That was Christopher and Sascha Segan from PC Mag. We have transcripts for this and other podcasts available at muninetworks.org/broadbandbits. Email 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. The handle is @muninetworks. Subscribe to this podcast and the other podcasts from ILSR, Building Local Power and the Local Energy Rules podcast. You can access them wherever you get your podcasts. You can catch the latest important research from all of the initiatives if you subscribe to our monthly newsletter at ilsr.org. While you're there, please take a moment to donate. Your support in any amount helps keep us going. Thank you to Arne Huseby for the song Warm Duck Shuffle, licensed through Creative Commons, and thank you for listening to episode 371 of the Community Broadband Bits podcast.