Mapping Data and Dollars - Community Broadband Bits Podcast, North Carolina Bonus Episode Three

This week is episode three of the new podcast project we're working on with the nonprofit NC Broadband Matters, whose focus is on bringing ubiquitous broadband coverage to local communities for residents and businesses in North Carolina. 

The ten episode podcast series, titled "Why NC Broadband Matters," explores broadband and related issues in North Carolina.

logo-nc-hearts-gigabit.png This week, Christopher and his guests explore mapping in our episode titled, "Broadband Mapping Means Money: Understanding How Data Drives Decisions.”

He talks first with Brian Rathbone, Co-Founder of Broadband Catalysts, a consulting firm that works with communities, non-profits, corporations, and governments to expand broadband Internet access. Brian and Christopher dig into federal mapping data and talk about some of the challenges in obtaining accurate data.

Jeff Sural works as Director of the Broadband Infrastructure Office for the North Carolina Department of Information Technology. He and Christopher take the mapping conversation to the state level. Jeff describes the work of the Office and explains why it's important that the state have the most accurate information possible. He explains state methods that involve citizen input about Internet access to help them get a more accurate picture of connectivity for residents and businesses in North Carolina.

We want your feedback and suggestions for the show-please e-mail us or leave a comment below.

This show is 54 minutes long and can be played on this page or via Apple Podcasts or the tool of your choice using this feed, at the Community Broadband Bits page, or at the NC Broadband Matters page. We encourage you to check out other "Why NC Broadband Matters" content at the podcast feed so you don't miss future bonus content that may not appear in the Community Broadband Bits Podcast feed.

Transcript below. 

We want your feedback and suggestions for the show-please e-mail us or leave a comment below.

Listen to other episodes here or view all episodes in our index. See other podcasts from the Institute for Local Self-Reliance here.

Thanks to Shane Ivers for the Music: What's The Angle? by Shane Ivers - Creative Commons Attribution (4.0) license.

Image of the American Community Survey Map courtesy of the NC One Map uses U.S. Census Bureau Data shows the percentage of households in North Carolina with no Internet connection.

Geoterm
Transcript

Brian Rathbone: And so I think we've made progress, we'll continue to make progress, but right now the fact that there's federal dollars are being spent without sufficient understanding of the true on-the-ground service, I think is indicator enough that we need to continue to concentrate on this.

Lisa Gonzalez: Welcome to a special episode of the Community Broadband Bits podcast and our new podcast series, Why NC Broadband Matters. I'm Lisa Gonzalez with the Institute for Local Self-Reliance in Minneapolis, Minnesota. NC Broadband Matters is a North Carolina nonprofit. Their mission is to attract support and champion the universal availability of affordable, reliable, high capacity Internet access, which is necessary for thriving local communities, local businesses, and a local workforce to enable them to compete in the global economy. The group has created the North Carolina chapter of CLIC, the Coalition for Local Internet Choice. We are collaborating with NC Broadband Matters to present this series that touches on issues that while certainly affect folks in North Carolina also impact people in other states. Our third episode is titled Broadband Mapping Means Money: Understanding how Data Drives Decisions. You've heard from us and from other organizations about the problem with mapping data. Most grants and loans established to connect unserved and underserved communities are based on FCC data that overstates coverage. Today's guests are working to change that. First, Christopher speaks with Brian Rathbone, cofounder of Broadband Catalysts. They get deeper into the problem as it relates to topology and federal mapping. Then Christopher talks with Jeff Sural from the North Carolina Department of Information Technology where the state is working to improve the data they use to determine where folks need better Internet access. Now here's Christopher with Brian and Jeff to talk about mapping.

Christopher Mitchell: Welcome to another episode of the Community Broadband Bits podcast. I'm Chris Mitchell at the Institute for Local Self-Reliance and this is episode three in the bonus edition of Why North Carolina Broadband Matters series. We're exploring better broadband Internet access throughout North Carolina, and today we're talking about mapping. Our first guest is Brian Rathbone, the cofounder of Broadband Catalysts. Welcome to the show Brian.

Brian Rathbone: Well, hello Chris. It's a pleasure to be here and hello to everybody out there listening

2:30

Christopher Mitchell: Broadband Catalysts — it sounds like something I'm in favor of. What do you do?

Brian Rathbone: We're a broadband planning consultancy. In general, we work with towns and cities and boroughs. A lot of times we do it at the behest of federal agencies like the EPA, USDA, and the Appalachian Regional Commission. And we help communities plan for their broadband future, and in many cases, what we do is go into rural communities where availability is low and help those communities understand how to use their existing assets to lower the cost of deploying broadband in their communities. And from doing that, they can attract providers who normally wouldn't be able to make a business case if they had to build all their own towers and do all those sorts of things. The folks that work with Broadband Catalysts all at one point worked with nonprofits or with the Department of Commerce in North Carolina to do broadband planning, and after we all returned to the private sector, Broadband Catalysts has been a good home for our continued efforts in broadband planning.

Christopher Mitchell: So we're going to be talking about mapping efforts around broadband in this episode. We're going to talk with you a lot about federal mapping, but I'm curious about the work that you do and the way you just described it before we start that other stuff. Do communities in rural areas or any of the communities that you work with, do they often have a sense of what relevant assets they have and what they are?

Brian Rathbone: Some do and some don't. It really ranges widely, and the understanding of what those assets can be used for also ranges widely. And a lot of times the ownership of those assets is sprinkled about throughout the community, and it's not viewed as a whole. To give you an example, the community that I live in, in Rutherford County, North Carolina, the water towers are owned by the Broad River Water Authority, and they partnered with a nonprofit who applied for an Appalachian Regional Commission grant to put a fixed wireless on those water towers. Now there's also county assets and city assets that are used in that that really lowered the cost by using existing buildings that already had fiber in them and using that to provide the access up to the water tower, and you know, all those structures were already there. That was a significant savings to our community, but it really took some assistance in helping folks understand which pieces of infrastructure were actually relevant to broadband and what that value was and how they could use it to partner with potential providers who may not know they could get access.

5:18

Christopher Mitchell: That's interesting. I just, I'm always interested in creative approaches to solving these problems. So I'm guessing that's Monday through Friday for you.

Brian Rathbone: Indeed.

Christopher Mitchell: So let's talk about federal mapping. And when you're describing your critique of federal mapping, I think that listeners are really gonna enjoy the rigor with which you are going to present it in terms of looking at different types of technology and how the mapping, you know, uniquely may result in better or worse results based on each technology, but let's just start with a general picture of what's wrong with the federal broadband mapping.

Brian Rathbone: Well, let me tell you a little bit about where I started with broadband mapping and how that related to the federal map. Originally I was hired by the EMC Authority which was the North Carolina Broadband Initiative back in 2010, and I had 31 counties in Western North Carolina that I was helping to do broadband planning. And part of that effort was to work with at the time the NTIA to produce some of the first broadband maps at that level of granularity to get them at least to the census block. And so as I was going up and working with these communities. I was looking at these maps and trying to understand the landscape I was going into. And initially, in a couple of the counties, we didn't have any of the data. They hadn't reported anything yet, and so I was like walking into a black hole. I didn't know anything, and it made it very difficult for me to be useful. And the more data I got, the more I found certain pockets and certain things that I could analyze from that data that was really useful. The problem I saw was even back when we were doing data collection for NTIA, most of the data was collected at the census block level. We did collect some data at the address level, but it was problematic in and of itself, and a lot of it we ended up rolling up to the census block level to have a homogenous dataset to look at — you know, lowest common denominator sort of thing. And so when the FCC took over that responsibility from NTIA in 2014, I believe it was, there was a different dataset employed, and now we're using the FCC form 477 data that is submitted to the FCC by providers that follows a certain number of rules that basically the providers are supposed to indicate census blocks where their technologies are available that they could deploy within a reasonable amount of time. I think it's 10 to 12 days or so. And so that's fine as long as everybody in that census block has service, but that's almost never the case. When you look at the FCC information in general, if a single household has been indicated to have access to a technology from a provider then the entire census block is lit as served, and that dramatically overstates the coverage. Everybody else in that census block may be unserved, but because there's that one, by our mapping standpoint that says it's served. And the reason that's so critical to all of our communities out there is those maps are used to determine eligibility for federal and other funding. And so if you're looking at a census block and you're saying up that census block already has DSL and cable, we can't provide federal funding to overbuild those networks, therefore that census block is not eligible for funding. That is really a difficult situation if only a small percentage of that census block really has service. Or if you're just one of the households out there that, "Hey, you know, what about me? You know, my neighbors have service and I don't," that's where the whole situation becomes very difficult. And since it's allocating billions of dollars, it really is critical for us to find as many ways to evaluate this as possible so that we can deal with the weaknesses in whatever datasets we're using.

9:17

Christopher Mitchell: Now, one of the things that people sometimes say as a criticism is that this data is all self-reported. And I think that sounds good, but I just wanted to do a quick check in with you on this because it strikes me that's the only kind of data that we're going to get because no one else knows where the service is available. And so I think ultimately our problem is not that it's self reported in the critique of some folks, but rather that it is done at this level that is not useful but then also that is not rigorously checked. Right?

Brian Rathbone: Right. Cause even if we switched to — let's just say tomorrow the FCC says you have to give us address level data, and we have to determine whether it's subscriber level data, demand level data, or availability level data because subscriber level data is the people who pay for it have it now, demand data is people who want it and don't have it, and then there's the other bucket of folks who basically, you know, don't have access to it but you know, you are able to serve. So, it gets really difficult to figure out exactly where folks fit in that from a provider standpoint, and you're right that it's difficult to then go and ground truth that data. Only some of this infrastructure can be validated by sending someone in and looking at the infrastructure. You can generally see the cable for example, but sometimes that's buried as well. You can usually see markers indicating it, but it's very difficult to actually crack open the case and look to see well how many DSL ports are available for this neighborhood? Or that fiber that goes by my house, is there anywhere near here that you can get access to it? Cause they generally just don't cut into fiber. So those types of things get really difficult, and that's where it comes down to technology by technology that are challenged in mapping.

Christopher Mitchell: Let's dive right in. Which technology would you like to start with?

Brian Rathbone: I would like to start with DSL because it's the one that most people are most familiar with.

Christopher Mitchell: Yeah. Also the most frustrated with.

Brian Rathbone: Yes, yes. And it uses our existing copper technology. That copper has been in the ground in many instances for 40-50 years, and the quality of it as questionable. It's an older technology that many of us, you know, lament, although I have a DSL circuit at my house and I use it all the time. So you know, it's a critical part of our infrastructure, but its limitations are many and they're fairly well known, but they're a little hard to articulate. The quality of DSL is usually determined by how much copper is between you and the node. So if you live in a fairly new development, and the telephone company has a big box by the entrance to your development, the amount of copper going between that box and your house is fairly short. And because of that, you can get really good DSL service because it's distance sensitive

Christopher Mitchell: And by really good, I suspect you're thinking about like 45-50 megabits down, 10-20 up, sort of in that neighborhood, right?

12:16

Brian Rathbone: Yeah. And that technology would qualify as what we call VDSL or very high speed DSL. Anything above 10 down, 1 up, you're getting into that VDSL, and the range on that as much shorter. You can't have much copper in that. So if you look at something like AT&T's Uverse product that is fiber to your neighborhood and then copper from the node to your house. But if you live more than a thousand feet or so from that node, it's very unlikely that Uverse will be available to you without some other infrastructure being put into place to get over that distance limitation.

Christopher Mitchell: Right.

Brian Rathbone: When you get over 3000 feet, then 6 megabit DSL becomes a problem. So the farther you go, the worse.

Christopher Mitchell: Right, and I'm going to stop interrupting you all the time — he said, probably not meaning it. I wanted to note one other thing, which was that you don't always know how far you are because the copper doesn't run in a direct line from you to the box. And so, I just wanted to clarify that also for people who might be mentally drawing the line. The line actually — I mean it could go all the way around the neighborhood and back to you. And so you might be just 500 feet away from the box but have 1500 feet of copper between you and it. So this is one of the challenges.

Brian Rathbone: Indeed. And if you need to know how far away you are, you can call your phone company and ask them the distance of your local loop. They should be able to tell you that. But it is is extremely frustrating for folks who had existing copper that came from three miles away and they have a development behind them who have a node that they're very short. So it can be just — you can see the people who have broadband, but you can't have it.

Christopher Mitchell: Right.

Brian Rathbone: It's a lot to do with historical — how did they build it back in the days that that community was constructed? The older the community, the more the problems.

Christopher Mitchell: So let's go back to where we were before I asked you what you meant by the higher speed DSL.

Brian Rathbone: Oh, no problem. The other thing about DSL is the fact that DSL deployments generally have capacity issues, in that most of them use what are called DSLAMs or DSL access modules. And they'll have 24 ports on them in general. It might cost $50,000 or so to put a new one of those in and the service to feed it. So if you've got 26 homes in the neighborhood and they all want DSL, only 24 of them are going to get it and there's going to be two of them on a waiting list. And the reason is, those two subscribers aren't going to make up the amount of money it takes to put that second DSLAM in. You know, when I first bought my house, DSL wasn't available to me because the DSLAM was full, but eventually one of my neighbors moved and turned off their service, and I jumped on. And now, whoever bought that house, that poor fellow can't get service. And you can't see that in the map, and that's one of the problems I have with, I don't know how to map DSL well because you can't really see that. And the distance of your local loops and all those things contribute. DSL has its list of mapping issues.

15:20

Christopher Mitchell: I'll just throw in that when the FCC was taking comments about how to do mapping better with this new proposal, which will full deal with some of the critique that you're offering, some of us did include include that we needed to have that data on whether or not a technology could provide service to 100 percent of the homes if 100 percent were interested, we're interested, and I think that may get at that. So it may not show up on the map, but this is something that the FCC should be tracking.

Brian Rathbone: Validation of that will be extremely difficult. That's one of the problems that we've had is that there will be some questions with regards to the accuracy of that data without having some ability to go and verify it, spot check it, do those sorts of things. But I agree with the point that you're making, and it would be an improvement to have that data even if there are some questions as to its full accuracy.

Christopher Mitchell: So let's talk about cable. What's happening with cable?

Brian Rathbone: So cable is an interesting industry where they tend to be limited by the number of homes per linear road mile, where their infrastructure is either on the telephone pole or in the ground. So if you live in an area where there's fewer than 15 homes, if you drive a mile on the road and you pass fewer than 15 homes, chances are you're not in the cable industry's business model. So most of it's going to fall within that more populated area where you drive past a hundred homes per mile. That's where the cable industry is going to flourish. But there are some things about the cable infrastructure that are difficult to map as well, such as which side of the road is the infrastructure on. And this is something I saw working with the EPA and USDA and Appalachian Regional Commission's Cool and Connected program where we went to 18 different cities, towns, and boroughs. Many of them I would go at least on one or two streets, and I would see one side of the street with cable infrastructure and all the houses on the other side of the street had satellite dishes. Chances are all the ones on the right had really bad DSL, and all the ones on the left with the cable infrastructure had 100 megabit or better Internet. And interestingly enough, there was also a financial divide. You could see the difference between the affluence of the neighborhood that had the service and that that didn't. Those who don't have as much money don't tend to get it. And to cross the street, if one person on the other side of the street said, "Hey, I want cable and I'm willing to pay for it," it can cost them upwards of $10,000 to get that infrastructure across the street and connected to them. So a lot of times, once it happens, when the cable company comes in and they're deploying, if they deploy on one side of the street, the odds of them deploying on the other side of the street are very low, unless almost everybody on that side of the street decides they're going to subscribe. It's cheaper to do it upfront, so if you get missed when they deploy it, it's harder to get them to come back and do it later, it's expensive, and I saw that a great deal. And the other thing that I saw is in rural areas, if let's say you pass 15 homes per mile but most of those homes are a thousand feet from the road — rural areas, farming areas, places out in the mountains where folks are specifically going there to not be right on the road, there's a really high cost. I think 250 feet is the usual cutoff that most providers have quoted to me. To say, if you're more than 250 feet from the road, I have to charge you at least $50 a foot to connect to you. For me, I'm 700 feet off the road. There's fiber at the end of my driveway. I'm looking at a $3,000 bill to trench and get it up here. And that cost is prohibitive and prevents a lot of people from getting served. So that really kind of covers some of the difficulties in the cable area. Some of the other things apply as far as the granularity of census block, but I really think the side of the road and how far you live off the road, those are indicators that I don't know how you account for those things in the mapping at the census block level, at the very least.

19:34

Christopher Mitchell: Let's talk about fixed wireless. What's what's happening there?

Brian Rathbone: Oh, you can hear the can of worms opening. Fixed wireless is one of the most difficult things to map.

Christopher Mitchell: Well let me tell you, in Kansas they did actually require people to put in the polygons showing where they could have service. And it was remarkable, but the fixed wireless shows up in a perfect sphere around antenna sites.

Brian Rathbone: Isn't that amazing?

Christopher Mitchell: And so actually, I think the mapping worked quite well.

Brian Rathbone: And to be honest with you, I work with a lot of fixed wireless providers on their mapping, usually for things like grant applications. They say, "If I put a tower here, who am I going to serve?" That is such a difficult question to answer with any level of accuracy, and it depends on what technology you're going to use. So I'm going to list off four different fixed wireless technologies just so we kind of know what we're looking at in the spectrum of things. One is high frequency, unlicensed fixed wireless, like 5.8 gigahertz — a service that does not go through trees very well at all. The second is lower frequency but white licensed — free but first come first serve type of frequencies, like the 3.65 CBRS service. And then you start getting into things like licensed wireless such as what AT&T and other big providers have paid billions of dollars for access to their frequencies, their type of wireless. I guess I only looked at three there. I kind of lumped a couple together. But essentially those are the types of service. And with fixed wireless, that is line of sight only, trees between the home and the transmitter will block service. That's the situation I was in when I bought my house. There's 5.8 gigahertz service available five miles from me, but the trees on my neighbor's property blocked it. And then when my neighbor clear cut his property, I got 20 megabit service. So hooray for the fact that he cut down those trees, I guess. I got Internet out of it. That was a good sign.

Christopher Mitchell: You got a few years of Internet until they grow back.

21:48

Brian Rathbone: Exactly. But it's a really good indicator that, yeah, you can go a long distance. I know, you know, folks who are 20-25 miles from that transmission site that get service off of it. But if there's a sheet of paper in the way, you've got a problem. In areas where there's a lot of foliage, mapping line of sight based service is questionable at best. So then you look at, well what about near line of sight service? What about the stuff that's meant to go through trees? And that's usually when you're talking about a two and a half to five mile buffer ring around the tower, that there's a decent chance you're going to have enough signal punch through the trees that you can still provide service that's going to be, you know, 25 down, 3 up, at least aiming for 25 down, 3 up.If there's a mountain in the way, if there's building in the way, there's lots of things that can block this signal that can cause it to not serve everybody in the community. That's the number one thing about fixed wireless for me is there's going to be people who can't get it for whatever reason. Even with AT&T's fixed wireless, I was very disappointed in them recently because I get AT&T LTE service here over — I believe it's somewhere in the 750 megahertz range, and that works through trees over 15 miles. But then they started serving my area with their fixed wireless service. And when they came out to install it, the installer said, well I'll get up on your roof and see if I can see any towers. At which point I just had to laugh and cry at the same time cause I knew he couldn't. If you can't go through trees, I'm not getting it, unless you happen to be on that one mountain top five miles away and they're not. So even the folks that are playing in the licensed realm, a lot of times they're using these higher frequencies to do the fixed wireless, and they still can't go through trees. So when you have to take foliage into consideration, which you pretty much have to, in my opinion, no matter what, and topography — you know, if you're in the flatlands it's easier. But if you're in the mountains, sometimes that's an advantage for having peaks, and sometimes it's a huge disadvantage because communities in valleys are very hard to serve.

Christopher Mitchell: So let's move to Fiber-to-the-Home. What's the mapping analysis here?

Brian Rathbone: I think with Fiber-to-the-Home, it's better, although that 250 foot off the road limitation holds. You know, to pull fiber, there's a cost per foot, and the same with the cable industry, they kind of build usually 250 feet into their cost model. So that's just their normal expense, but when you're beyond that, it's usually $50 or more a foot depending on what they have to dig through to get to you. It can go up from there. But that's probably the biggest challenge on mapping Fiber-to-the-Premise. You know, if you're doing it at a granularity that's the census block, you're still gonna have that issue that there's going to be homes that don't make the business case that are going to get left out in your mapping because they're sitting too far from the fiber or for whatever reason it costs too much to get to them. In some cases, it's things like they're on the wrong side of the Blue Ridge Parkway. They're on the wrong side of that wildlife reserve. Those are types of things that I've seen throw a monkey wrench into fiber deployment.

25:04

Christopher Mitchell: Well, and I'm guessing that even if you have a maximally defensible fiber company that is saying, "You know what? We have an unlimited checkbook. We will build fiber to everyone," they probably don't even know if the house is within 250 feet of the road or not, right? I mean, this is an unknown as opposed to just something that's inconvenient.

Brian Rathbone: It's true. But I can say we can look back a little bit at what was done back in 2010 with the American Reinvestment and Recovery Act. We used — several counties in North Carolina got roughly $20 million to provide countywide service, and their goal was 97 percent because the last 3 percent doubled the cost of the network. And that's really difficult math to get past. So you're always going to end up with some subsection of the community underserved, and my feeling on it is one of the best remedies to this or at least supplementary datasets that can give us a better view of this is citizen sourced data. And that comes in two flavors right now that I'm using. One is speed test data generally gathered by third parties. I'm not hosting my own speed test at this point, but M lab has taken their data, they gathered and made it available on the Google cloud. It's just two or three years out of date is its number one flaw. Citizen sourced data is the data that I am collecting directly myself and that others collect directly themselves through surveys, essentially allowing the citizens to have a voice. There's been folks who said, "Hey, you're going to have accuracy issues asking people about their Internet," and they are correct, but nonetheless it is better than having no data. So, we do our best to kind of bridge the divide of, I have to ask somebody if they have Internet over the Internet. That part's difficult. We have to work with that. And how do I ask somebody who might not understand what their Internet is or what they have to respond with the information I need to accurately determine the level of service they either do or don't have. We have found that in general when you combine the FCC flawed data set with our flawed data set and any other available and probably also somewhat flawed data set, put them together, we end up with a much better view and we can use those things to offset them. We can find weakness in the existing infrastructure. And for me, if you can have folks report their location and we know what census block they're in and we know how many homes were in the census block as of the last census, then we can say yes, this census block is shown to have DSL, cable, fixed wireless, and fiber. However, 20 percent of households in this census block have reported themselves as underserved. At the very least, we can consider that census block underserved and contested, and it's worth looking at does that census block warrant investment even if it's limited in scope, maybe limited an area to an unserved area. That's where I think citizen sourced data really has a big value point. The other place that has a big value point is when you're applying for grants, and I'm going to take USDA's ReConnect program as an example. The more information you have about the number of farms, the number of students, the number of businesses that your infrastructure would serve, the stronger your grant application is, and a lot of this type of information can be gathered through a properly developed survey instrument that asks the right questions: What is your budget? Are you a farm that makes more than a thousand dollars a year? By collecting that data and mapping it, we can produce a data set that helps us get a more thorough understanding even though none of our datasets are perfect.

29:20

Christopher Mitchell: One of the things that I wrestled with as you're describing it — I mean, what you're doing is, is really good and powerful. Do you think if we just had access to — you know, if for instance, CenturyLink and AT&T had to make available where their fiber runs are, where are their copper runs are in some kind of meaningful way that was useful that you could specify, that we could dispense with a lot of this? Or do you think there's just no way around the fact that this is messy and we need to be creative?

Brian Rathbone: I don't think there's any way around the fact that it's messy. I think we're going to have to be creative. We're going to have to think about each technology a little bit differently. Fixed wireless being a great example of where we just need to rethink how we're doing that. You know, let's take for example, you said what if the, the incumbent providers told us where all their infrastructure was. This is something I've seen reproduced at the third party level where consultants will go into a county and do an infrastructure survey where they drive and they look where all the fiber is. You can tell what it is on the pole. You can tell who owns what's in the ground. You can drive the infrastructure and you can get an idea. And I always struggled with the fact of, okay, you told me that AT&T has fiber here — how many of those strands are in use? And so if we were to ask about infrastructure, I also have to ask about which of that infrastructure is lit, which is dark, which is maybe available for lease. And to be honest with you, those questions generally are viewed as intellectual property and trade secrets. And so it's almost like if you can't get all of the stuff, it's not as useful because I can find — I know there's fiber there, but I couldn't tell you necessarily that you could do anything with it.

Christopher Mitchell: Right. And I think they have a history of being overly broad in claiming that things are intellectual property, but I think I agree with what you're saying in that we could agree that at a certain level of detail there is a true competitive desire to keep things secret. I think I draw the line differently, but I think that there is a line, and what you're saying resonates with me that ultimately to have good maps, we can't just pretend that simply, you know, getting all the information that in a perfect world would be public, that would not be enough in and of itself.

31:42

Brian Rathbone: I agree with that, and you know, the truth of the matter is, you know we would all love to see this problem solved. I think the providers would like to see this problem solved and do it in a way that they can remain competitive and they can do the things that they do. Cause that's one of the things that I've learned through my time is that demonizing the providers does nothing for anyone. They have a very difficult job and they have difficult margins to make and just the burden of asking them to provide some of this information digs into their bottom line. So I do feel for them and I understand that it's difficult and I want to find a way to work with them and help them as well. So you're asking this question of how do we do something we don't really know how to do well, that the responsibility largely falls on the private sector who's self reporting or the community. And with the community, you know, with the maps I could show you of the data I've collected, it's really clear that the problem is awareness. We go to communities and we work there and that's where we end up with data. Everywhere else we have nothing because I don't have an effective way to communicate. Even when I go into a community, the first thing I say is can we survey through the schools? Can we survey through the hospital? Can we survey through the power company? Because I don't have a way to get to the citizens to allow them the opportunity to know they can make their voice heard. As soon as you give them the opportunity, they will come out in droves. And you know, sometimes you even just have to go out and find them in these remote communities, but it's really going to be a challenge. I think you're going to have to come at it from provider sourced data, citizen sourced data, speed test data, all the different ones, and then map them together to start to show trends, areas of weakness, areas where investment might be getting diverted away that shouldn't be that we can then go back to the FCC and these other agencies and say, "Okay, based on our current methodology, here's some areas that we think need to be reconsidered as to whether or not they're truly eligible." And can we carve out parts of census blocks that are contested and make them eligible in places where we wouldn't be overbuilding somebody else's infrastructure with federal dollars, but we would be entering a census block that is partially served. And most of the time when you have a partially served census block, it will remain partially served unless somebody else does something to change the environment.

Christopher Mitchell: Well this has been quite in depth and I really appreciate all your time with it.

Brian Rathbone: Oh, it's my pleasure, and I've enjoyed talking with you and others about this problem over the years. I do think that there's been a lot of good work done on the private sector side, on the nonprofit side, on the university side, working with Dr. Rick Bunch at UNCG on the radio wave propagation analysis and understanding how to deal with trees and things like that. So many different folks have put their mental energy to this that we've come a long way. If we look back to where we were when we started all of this the costs involved have come way down. The ability of folks to get to this data and use it themselves has gone up. And so I think we've made progress, we'll continue to make progress, but right now the fact that there's federal dollars are being spent without sufficient understanding of the true on-the-ground service I think is indicator enough that we need to continue to concentrate on this.

35:20

Christopher Mitchell: Well, thank you Brian Rathbone, cofounder of Broadband Catalysts. For the listeners, you're about to hear my voice talking to somebody else. Back in a second.

Christopher Mitchell: Welcome back to this episode three of Why North Carolina Broadband Matters series, where we're exploring better broadband Internet access in North Carolina, how to get there. Now we're going to be speaking with Jeff Sural, the director of Broadband Infrastructure Office — um, the director of the Broadband Infrastructure Office in North Carolina's Department of Information Technology. Welcome to the show, Jeff.

Jeff Sural: Thank you. Thanks for having me.

Christopher Mitchell: I feel like whenever anyone has a longish title, you always have to make some kind of a joke about how it fits on the business card.

Jeff Sural: Yeah, we're a government agency, so we use a lot of acronyms, you know.

Christopher Mitchell: Right. Well, and what does your office actually do? I mean, I think, you know, we have a sense of what information technology generally does, but within that agency, what does the broadband infrastructure office focus on?

36:26

Jeff Sural: Well, we focus on enhancing broadband deployment and adoption across the state so mostly on policies and planning, and also we operate or run certain programs. For example, a few years ago we published Connecting North Carolina, which is the state broadband plan. It lays out over 80 recommendations for the state government, local governments, and nonprofits, government agencies on how they can improve broadband, both availability and adoption. And then we also do some grant administration. We have a $15 million broadband grant program, and those funds go to Internet service providers that will deploy in in rural areas across the state. So we just had our first round. It was a $10 million pilot program, if you want to call it that. And our general assembly just reauthorized that program at $15 million a year for 10 years. So that's a big part of what we do. And then we have a technical assistance team that works with counties and local governments on planning. We have a community broadband playbook that that we host on our website. And then these individuals work with each of those local governments on a number of different different things like, you know siting vertical assets in the community. We have a survey, so they'll conduct demand aggregation campaigns. And so that's that's another big part of what the office does. We do have a public safety element. We administered the First Net program, that federal program and grant. Because we're a government agency, of course we have a vision statement, and that's that all North Carolinians should have access to affordable broadband services. And we believe this because we have seen broadband enhance a community's viability and livelihood by creating income opportunities, facilitating greater civic and cultural participation , educational opportunities, access to healthcare providers and other essential services. We've seen communities revitalized and changed when they have reliable, affordable, high speed Internet access.

Christopher Mitchell: So as you're listing all the things that the broadband infrastructure office does, I'm realizing it's probably quite larger than I imagine. How many people are there working for it?

Jeff Sural: Well, we have 10 folks. One is a as a part time temp solutions person, and then we have two individuals that work almost exclusively on the public safety side of our office. They were the folks that were funded through the federal grant to set up and educate policy makers and the governor during the First Net procurement. And if you're familiar with that process, each governor of each state had to either opt in or opt out of that program, which is, you know, the nationwide wireless broadband public safety network. And so they did a lot of work and data collection and so forth. They're continuing some of that work, and then we're trying to evolve that office into a public safety communications, technical assistance team, and looking at future technologies to help our our state agencies.

40:05

Christopher Mitchell: Yeah. We have mostly ignored FirstNet because it seems like it's overwhelmin, and I don't want to put you on the spot in talking any further about that. I want to focus on . . .

Jeff Sural: Yeah, yeah, yeah. We have four members of our technical assistance team. These are individuals that have decades of experience working for industry providers, broadband providers. And then we have some folks that work on data and research.

Christopher Mitchell: And so I'm curious, for all these different folks, how does mapping really play into how they do their jobs? And you know, how does your office ⁠— like, why do you care so much about mapping when you're trying to make sure people get the benefits of these high quality broadband networks?

Jeff Sural: The primary concern is money and funding. Both our broadband grant program that I just mentioned and several of the federal programs that fund broadband deployment, including some of the programs that are out of the USDA and out of the FCC, will rely on or, you know, define unserved areas based on their maps and based on data that they receive or that those entities received. Now, as you probably know, most of these programs and most of the nation looks to the FCC maps as sort of the standard map for, you know, broadband coverage both for wireline and wireless. The predecessor to this office had received grant funding through the US Department of Commerce, you know, National Telecommunications and Information Administration, to map the state's broadband coverage, and we did that back in 2014. But since that time, there haven't been any updates to that map and the data collection at the state level stopped. And that data collection was from the Internet service providers themselves. So since that time we've simply relied on the FCC maps, and for all the reasons that were previously discussed, those maps are inaccurate, you know, at best. If we're willing to focus money in areas that need better broadband, then we should have pretty accurate data. And we feel pretty strongly about this, especially at the state level because these dollars are taxpayer dollars and we want to make sure that they're used to their greatest benefit. The other important reason that we need accurate data and maps is for our technical assistance team, which does a lot of planning with communities. So we need to know where their infrastructure is when looking at helping those communities.

42:49

Christopher Mitchell: What tools do you have as a state? To some extent, there's a question of, well, if the Federal Cmmunications Commission is unable to produce accurate maps, what can the state of North Carolina do?

Jeff Sural: Yeah, and that's a great question. And sometimes, you know, we feel like we're the lightweights trying to take on the heavyweights. I think, you know, with the FCC data collection and their methodology, we think that could be pretty simply solved by simply having, you know, the Internet service providers provide more accurate data or specific data I should say about where they have coverage and where they have infrastructure. I think it's simply, you know, a policy decision that's been made to limit the amount of information that is collected. So what we're doing at the state level is we've kind of taken it upon ourselves and encouraged citizens to kind of take over and tell us where they don't have coverage or where they're lacking service. So we've been working with the counties, mostly at the county level, we've done demand aggregation surveys and some speed testing. We have a hundred counties in the state, and we've done this type of initiative with 40 so far over the last four years. We also have on our website a speed reporting tool, which we're going to upgrade here in the next few weeks. So it's not currently available, but we're going to pair that speed testing tool with a survey so we can gather more specific information from folks that that don't have a service. We use a couple of commercial mapping services. Those services allow us to locate, for example, where fiber runs are, lit buildings, central offices, cell towers, and so forth. So those are all key pieces of infrastructure that we like to know about. Finally, one thing that we found incredibly useful is our grant program. So when Internet service providers, big and small, apply, we ask them to be as specific as possible about which locations they want to serve in their proposed project areas. The legislation that authorized this grant fund allows us to declare portions of census blocks unserved. Well, our legislation allows us to go into that census block and do a deeper dive. In other words, so the grant program has a mechanism called a protest and if an Internet service provider believes that an applicant is going to serve an area that that's already should be considered served or designated served where they're providing service, then they can submit information to us. Now we make them and require them to provide detailed information. In the last round or our first round of our state program, we received approximately 15 protests, and in each of these protests we were able to use some pretty detailed data to drill down into these areas and actually look at almost every location within a particular area. That was helpful because now that gives us something to build upon.

46:28

Christopher Mitchell: You know, you've already described some of the challenges that you face, but are there other challenges you've faced along the way in terms of how you're trying to collect this data and make it useful for folks?

Jeff Sural: Yes, definitely, and part of it is on the collection side. It would really be helpful if we could simply receive information, preferably location specific information from the Internet service providers that are operating in the state. There's been some pushback on that. Some of that is push back based on the administrative burdens that would cause some of these companies. Some of it is based on the resources that we have here at the state level. It might not be available to collect that data. Georgia recently went through an exercise where they were able to get some very specific data from Internet service providers. Part of the reason why they were able to receive that was because the state law protected that information from their public records act. So, that would be another challenge. You know, we need to make sure that the information is protected. I mean, we appreciate and want to respect the confidentiality of these providers, but we also feel like there's, you know, a balance there or could be a balance there. The other challenge that we faced would be on the data itself. You know, it's not just as simple as looking at a Google map. We've all had this experience of typing in a location in our phone or in our Google maps and get into pretty accurate directions to where we want to go. But a lot of our GIS folks that are working on these issues really rely on very detailed information and data. And so for example, when we receive maybe an address location from someone and we go to plot it and they try to geo located on the map, sometimes those don't translate very well. So there are some technical challenges that we are facing.

Christopher Mitchell: Do you feel like this is something that in five years you're still going to be wrestling with or are we working this out?

Jeff Sural: Gosh, I hope not. It has been something we've been wrestling with. Well I've been here five years and I know that, you know, even before I got here folks were wrestling with this issue. It seems like a fairly straight forward fix, and we just can't seem to get there. What I would like to get to is a place where at least we have sufficient data and information to inform these funding decisions, you know, to basically to direct us to where the money should go. And I think if we do that then we'll be in a good place, and I think we can get there within the next five years.

49:17

Christopher Mitchell: Good, because I'm also tired of talking about mapping. Who isn't?

Jeff Sural: Yeah. As CAF II has sort of progressed and is now winding down, I think people and specifically lawmakers and policy makers are starting to figure out, wow, this didn't turn out, you know, as successful as we thought. And part of that has to do with the data itself and what we're collecting. So now that Congress is starting to get involved and raise awareness of the issue and is looking at, you know, passing legislation and they've been holding hearings ⁠— I think that's really really moving the needle. So I think with that sort of attention that we'll get things moving pretty quickly here.

Christopher Mitchell: I'm glad to hear that. You know, I think so too. I think the FCC's new proposals are better. I think some of the state experiments ⁠— Kansas and Georgia both have been thinking outside the box a bit in ways that I hope we'll all learn from. But, I want to end this by with where you begin it in some ways, which was to remind us that, you know, you're working on this issue, not just so people can have broadband Internet access for the sake of having broadband Internet access, but so that the communities are stronger, we have better educational opportunities, economic opportunities, and things like that. So you know, I think it's a reminder that winning the mapping discussion is sort of the first phase of a multi-phase approach to solving this problem.

Jeff Sural: It really is. And you know, as we're looking at, you know, the rural-urban divide that continues to grow and we're looking at, you know, the low income and high income divide you know, income disparities continue to grow, there are some equalizers out there and we feel like access to broadband is one of those. It's pretty remarkable to see the transformation of a company out in rural North Carolina, specifically a small manufacturing company in Madison County, which is far west in our state and a very mountainous county. But they were able to finally connect their facility to fiber and, and because of that, they were able to hire a couple more salespeople to work the phones and also to start communicating with their customers overseas and to troubleshoot any of the problems that the tools may have been having. They've transformed their business and become a global player, you know, and I think that says something about the power, you know, of high-speed Internet access. So yes, we've seen it transform small businesses and really allow for opportunities in a lot of our rural areas.

52:18

Christopher Mitchell: That's great. And, you know, that's what we're hoping for is that we can see economic opportunity spread much greater into the rural parts of our state because people can choose to live in cities. They should choose to live in rural areas, but fundamentally it should be, they should be able to thrive wherever they want to live.

Jeff Sural: That's right. Exactly.

Christopher Mitchell: Well thank you Jeff for coming on and spending this time with us to get us a better sense of both why this is important for the state but also what you're doing to deal with it and what in five years you will be able to stop doing.

Jeff Sural: Yeah, exactly. Well thanks for having me and thanks for raising the level of awareness around the mapping issue. And I'll be happy to stop talking about it in less than five years.

Lisa Gonzalez: Thanks for tuning in to the special Why NC Broadband Matters podcast series and for listening to the Community Broadband Bits podcast from the Institute for Local Self-Reliance. Remember to follow Christopher on Twitter. His handle is @communitynets. And if you follow @NCheartsGB on Twitter, you'll tap into all the NC Broadband Matters material. We want to thank Shane Silverman of SilvermanSound.com for the series music, What's the Angle, licensed through Creative Commons. And we want to thank you for listening. Until next time.