Analyzing FCC Data Collection with Derek Turner from Free Press - Community Broadband Bits Podcast 380

Derek Turner: We don't succeed if we're operating from a basis of fantasy. We have to operate from a basis of fact, and I think the facts are on our side.

Lisa Gonzalez: Welcome to episode 380 of the Community Broadband Bits podcast from the Institute for Local Self-Reliance. I'm Lisa Gonzalez. The Federal Communications Commission has faced growing criticism in recent years about the accuracy of the data it collects and uses to determine where in America people have access to broadband. In recent months, the FCC announced that they would establish a new approach to collecting the data and asked for input from stakeholders and interested parties. In addition to the Institute for Local Self-Reliance, the nonprofit Free Press submitted comments. Today, Derek Turner from Free Press comes on the show to talk about the problems with the old data collection techniques, the FCC's proposal, and his organization's recommendations. Christopher and Derek talk about the Form 477, which is the instrument that Internet service providers use to report where they offer broadband access. They also discuss why Free Press believes that this form, while not perfect, shouldn't be scrapped as many other commenters have suggested. Derek and Christopher also get into what they expect in the long term from data that is more granular and where challenges may occur. Now here's Christopher with Derek Turner from Free Press.

Christopher Mitchell: Welcome to another episode of the Community Broadband Bits podcast. I'm Chris Mitchell with the Institute for Local Self-Reliance up in Minneapolis, talking with Derek Turner, the research director for Free Press. Welcome to the show, Derek.

Derek Turner: Hey, thanks for having me. I'm glad to be here.

Christopher Mitchell: Well, I'm really glad to have you. I feel like you're a person who speaks often through words on the page, and so I'm excited to talk about some of these issues. Tell us about Free Press.

Derek Turner: So a Free Press is a nonprofit. We were founded in 2003 with the aim of increasing the public's role in media and telecom policy debates that largely happen in Washington, D.C. and around the country with the public not really having a voice. And you know, we believe that meaningful engagement in our public life, our democratic process requires that people have equitable access to communications technologies, like broadband and services, you know, like wireless services, but also a diverse, independent ownership of the media platforms that use those communications technologies to speak to us. And of course, we need journalism that holds leaders accountable and tells people what's actually happening in their community. So, media and telecom policy in D.C. I do a lot of work at the FCC and in Congress, mainly around our telecom issues, like equitable broadband access, cheaper broadband, faster broadband.

Christopher Mitchell: You seem to be a numbers person from what I can tell. I mean, you've dug deeper into some of these databases than I think most people have who call themselves researchers on it. So, where does that come from?

Derek Turner: My first life, my first career, I was a medicinal chemist. And I kind of got fed up with the way the media was working, and decided I should go back and maybe apply my skills — instead of trying to help people have better lives through chemistry, have better lives through better media. And so, I got a degree in public policy from the University of California - Berkeley where I specialized in quantitative analysis and tried to apply my scientific background to public policy. So yeah, I do have, you know, a bit of experience in econometrics and quantitative analysis, and I try to apply those skills when I can. And of course, that means I need and want as much quality data as I can get my hands on, but I also think that kind of data and good analysis, whether or not it comes from me or someone else, is really important for policymakers to have. So that's why we at Free Press have been working at the FCC and in Congress on this issue of broadband data since I started working here about 15 years ago.

Christopher Mitchell: Yeah, I think that you guys do a really good job. And that's actually one of the reasons I wanted to bring you on is not just because I think you're mostly right, but because I respect the way that you'll tell other people that agree with you that they're wrong about some details, even if they may be right in general. That's one of the discussions that you and I have gone back and forth on a little bit because, you know, I still don't have a full sense of all the data that's available, data that I think you're regularly using, and as we're trying to figure that out, I've really appreciated your candid assessments of what's out there and how people use it.

Derek Turner: Yeah, that's a friendly description of some of the back and forth we've had, and I really appreciate that. But no, I think ⁠— I love all the folks in the advocacy community, and I think that, you know, we don't necessarily need to speak with one voice. But I think when we all have a common understanding of the information that's out there and we're looking at it fairly and with clear eyes, I think our work is much more effective against those out there who oppose some of the same goals that we have and who aren't, you know, shy about hiring for profit economists that are a little bit sketchy and they're certainly willing to spin the data in their favor. So, you know, I think it helps us if we all have access to good data and have a clear understanding of what that is. So . . .

Christopher Mitchell: Well in the discussion about good data, I think you might be ⁠— and this is not unusual for Free Press, which is very happy to take a contrary position in ways that put it outside of norms perhaps ⁠—you might be the only person willing to defend Form 477 in an era in which everyone but everyone is talking about how terrible our mapping is. And so, I think your position makes perfect sense, so I want to unpack it.

Derek Turner: Sure. Yeah, I do recognize that from the outside our position does look a little incongruous to where a lot of the criticisms are. But you know, I think as we get into this a little bit, you'll sense that I have a little bit of a sort of fatherly pride or personal pride in the data collection efforts at the FCC, in that they used to be really, really, really bad, and the criticisms we're leveling now, compared to where it was, it shows how far we've come. And so, you know, I think that I sort of come from position of we've made so much improvements to what the commission is doing — they certainly can do a lot more, but let's not throw the baby out with the bathwater. And let's realize how, not lucky, but how much hard work by a lot of people other than me, you, and other folks way back when had to go into getting us to where we are now. And it doesn't mean the job's over, but it does mean we should recognize, you know, the good things about the FCC's data that are worth preserving.

Christopher Mitchell: So let me see if I can describe this in a way — and I welcome your correcting me, but I think the way most people think of it is Form 477 is a form that Internet service providers have to fill out twice a year in which they submit, among other things, information about their service. And it's done at the block level, which is to say a census block, and it involves both information about how many subscriptions they have at different levels as well as what kind of service that they could offer in a reasonable period of time to someone ⁠— to at least one premise within that block.

Derek Turner: That is it. That is Form 477 as it existed basically a few months ago. Now, they're making some changes to it, which we can go into, but that's basically correct. The idea behind this data collection effort, which began in 1999, a long time ago, two decades ago, was essentially we were at the dawn of the broadband era, and the commission had a statutory duty under the then 1996 amendments to the Communications Act to promote competition and to promote the deployment of advanced ⁠— what they call advanced telecommunications capability, or in other words, broadband. And so the commission began collecting data from providers in 1999, but they used a really bizarre way of going about this. They basically said, "Hey, providers, tell us if you have at least one subscriber in a zip code." And if they did, then the FCC counted that zip code as covered. And so, what that ended up doing, it ended up that, you know, at a time when everyone knows their only viable choices really for broadband would be their local phone provider or their local cable company, the FCC was telling everyone, "No, 99 percent of the population is covered and their average number of available providers is eight." So just a wildly overstated methodology here. And then on the subscription side, they only collected data on the number of subscribers at the state level. So you know, you could see changes at the state level and somewhere like Wyoming or D.C. — they're going up, they're going down — but you don't really have any idea of where that was or, you know, any kind of granularity that. So that's how they started out, and over the years, particularly in 2008, with a lot of input from the public interest community, they'd made a major expansion to that to go down on the deployment to the census block level — actually that came later, but to go away from the zip code methodology to go down to the census tract level per subscribership. Congress got involved in 2009 around the Stimulus Act, told the commerce department to go and collect better data, and they decided that the census block level was where this data would reside. And you know, I think that made a lot of sense at the time. We at the time had been advocating, "Hey, ISPs know every single address where they offer service. Why can't we just get access to an address level database?" And they fought us tooth and nail on that and said, "Okay, well we'll go up to the block level." A census block level can have anywhere between one person, but on average it's about, you know, 50 to 80, almost up to 100 in some really dense areas, of folks living in a block, and that ended up working for what we were trying to get at there. We felt that that was a really good approximation of deployment and that it came with the bonus of now we can finally use the census demographic information to figure out, well wait a second, carriers, where are they deploying? Are they deploying to richer areas? Are they deploying to areas that don't have diverse populations? We were able to know the income of those areas. We were able to do all kinds of complicated analysis that we just simply weren't before, and so so that was quite welcome. But, as we've seen, the Form 477, we initially thought served multiple purposes. It does serve multiple purposes. It's to help the FCC know where broadband is and isn't, but it's also to talk about competition and analyze competition. But it turns out in very, very rural areas, census blocks can be gigantic, and you will have a situation in some areas where a block is considered covered, even though only a tiny corner of that block's covered and maybe most of the homes that are on the other side of that block in a rural area aren't. And so, that is where the FCC has come under a lot of heat lately, that they're overstating the level of deployment, particularly in rural America, and that's leading to inefficient targeting of their USF subsidies.

Christopher Mitchell: And that has led many people to just saying Form 477 doesn't work. So you've gone through some of the history, so let's talk about this. You know, first of all, almost everyone talks about the deployment data, so tell me a little bit more about the subscribership data that, that you've used in several of your filings recently.

Derek Turner: Sure. So when the commission first adopted Form 477, it was both to look at where broadband's deployed, but it's also to look at where it's adopted because if you build it and they don't come, that indicates maybe you've got a competition problem or that maybe you've got an education problem ⁠— whatever, we don't know. So they collected data on subscribership at the state level. That was way too high level, so they went down to the census tract level, which is, you know, a few hundred to maybe low thousand group of people area size. And 2008, that's when ISPs started reporting their number of subscribers at the census tract level, but also what type of technology they use, what's the speed of that technology, download, upload. And that was actually something we fought very hard for because in the original zip code methodology ⁠— for example, I used to live in Northern Virginia, and my zip code showed up as having three cable providers. And it's just because Cox would serve one part of the of the area I lived in and zip code I lived in, RCN was in another part, and then Comcast was in another part. Obviously, cable providers don't compete with each other, so by taking it down to the census tract level, you were able to see, wait a second, here are their actual number of subscribers to the cable provider, here are the actual number of subscribers to the DSL provider, and this is the level of competition we have. It's a duopoly, and now we can start to correlate that with other data that we want the commission to gather, like price, quality of the service, things like that. And so, that was the impetus behind collecting subscribership data. Now, the commission has done nothing with it since then. It's a really vastly untapped resource for the public because the commission not only doesn't analyze the data itself, they don't release the full data to the public on subscribership data the same way they do on deployment data. So yes, in our recent comment you're alluding to, this data, even though they don't release much of it, it can glean data points such as this. So for example, satellite broadband, you know — technically the entire United States is covered by satellite broadband, but in all the areas I've ever lived in, I've never gotten a flyer from HughesNet saying, "Hey, would you like to sign up?" even though I get all kinds of mailers all the time from my cable and DSL company. But it turns out, only about 1.7 percent of all the fixed subscriptions in the United States are to satellite, and it's the same way with fixed wireless. So —

Christopher Mitchell: Well, let's stick with satellite for a second just to unpack that. And so, I mean, one of the things that your comments talked about is that the satellite companies know that they don't compete everywhere, right? They could serve anywhere, but it would be — like, none of my neighbors are going to sign up for satellite unless it's a horrible mistake that they've made, right?

Derek Turner: Yeah. I mean, there's no way that a rational consumer would willingly choose a more expensive service that has severe limitations on it in terms of data caps and latency when there are other options out there available to them. And the satellite companies know that, which is again, why they don't market to most of the United States.

Christopher Mitchell: But what really I find amazing is it shows what a market rejection that is because you have a lot of people living in rural areas in which they cannot get anything else, and they've decided it would be better to have nothing. And that might be because of cost pressure or it might just because of the quality of the service that they have rejected that service that is available, and frankly, as bad as it's been, I mean, the satellite today is way better than it used to be. It still has limitations, but it's not three megabits anymore in most cases.

Derek Turner: Yeah, it's definitely improved. They've launched the new satellites that's made it for those that do have the ability to pay the quite exorbitant prices and who absolutely need it and who can't fall back on mobile, which is what a lot of folks have done when there's nothing else available — you know, it's a service of last resort for those people. And thank god it's gotten better, but certainly it's not comparable to any of the other wireline and fixed services that are out there. And again, this is a problem at the FCC with their analysis because if you just look at their patting of themselves on the back of their reports, they'll say, "Oh, you know, the average U.S. resident has maybe three or more fixed options available to them," and they're counting satellite in that when that's not actually the case. It's not actually available. It may be deployed, but it's not available.

Christopher Mitchell: So let's talk briefly about fixed wireless. What are the numbers around fixed wireless regarding how much of a claim there is to service versus the amount of subscribers?

Derek Turner: Yeah, I live — where I'm at now in Southern California, I show up in the FCC's data as having a couple of fixed wireless options available to me. But I've gone to their websites, and it's not clear that they're actually marketing their services to me. But yeah, again, they cover, according to the FCC's data, about 37 percent of the U.S. population, which is a pretty substantial size, but they only account for about 1 percent of all the FCC's reported fixed lines. So again, it is a solution for some people, but it appears that it's not actually a viable product for a lot of folks. And that can be for different reasons. Maybe they don't know about it, but probably it's more likely that where it is available, again, it's kind of this sort of almost last resort service to where people who don't have many other options are willing to go. And it doesn't have to be that way. You know, there are some very innovative fixed wireless projects in urban areas where it's not a last resort service. They're actually offering better ⁠— at least better priced service and maybe without the data caps compared to their DSL provider or their cable provider. But again, those are few and far between compared to what most of the average households in the U.S. actually can go out and purchase.

Christopher Mitchell: I think it's worth noting, like you said, Monkeybrains, Common, NetBlazer are the three wireless ISPs — I think Starry — a number of people quite like them, and so there's those. Your data's also, I would expect, 18 months old for that, and so I think fixed wireless has come a long way over those 18 months. But even so, I mean, maybe it's tripled. Like that would be an incredible jump, and it would still only be 3 percent.

Derek Turner: Right, exactly. Yeah. The most recent data on subscribership from the FCC is the middle of 2017, so we're two years behind there now. And yes, I would certainly think that fixed wireless has made a jump, and certainly a lot of the ISPs you just mentioned are putting out faster and faster speeds, and you know, I see more coverage of them in the media. I think awareness of them as options — particularly Starry has had a big push in the media from where they're operating. But again, I would like to look into it where I live. Certainly, I don't want to give my business to the cable monopolist if I don't have to, but it's just not a viable economic alternative compared to even the duopoly providers that I have available to me.

Christopher Mitchell: So as we wrap up talking about Form 477, I think there's two points. One is, you went into this in great depth in an email chain regarding the data from Microsoft. People, including myself, have said incorrectly that the Microsoft data shows that Form 477 data is horribly wrong, and you've convinced me that you're right, so tell me how I'm wrong.

Derek Turner: I don't do this with any glee, but I do think it's ⁠— you know, look, the folks at Microsoft are trying to get the FCC to pay attention to a number of policy issues, and one of those is the need for better data on where rural deployment is actually happening. And to do that, they have put forth some data that they use that basically is their statistical analysis of people doing, you know, windows updates in the middle of the night, ao they're able to actually, because Windows computers are everywhere, have a good sense of the actual deliverable quality of people's broadband connections. And basically what they're coming up with is a number that shows that about half the country is not using home Internet at the speeds the FCC considers to be broadband ⁠— so 25 Megabits per second downstream, 3 Megabits per second upstream. When you hear that number and then you compare that number to what the FCC says is available to people — 93 percent of the country has that level of broadband. So you hear 50 percent or is it 93 percent? Well, the difference there is Microsoft is measuring what people actually subscribe to, and some people for some reasons, a variety of reasons, are willingly choosing a slower tier than 25/3 in areas where 25/3 is in fact available. And if you look at some of Microsoft's subsequent filings, they kind of clean this up a little bit in their language and say that their data actually does match the FCC subscribership data. So I think they're playing a little bit of a game of trying to stoke a lot of upset politicians, fear in people that money's going out the door where it shouldn't be and they're using that number — which is a very interesting number. I don't want to downplay that number. That number is interesting for a number of reasons. For one, it tells me that there are a lot of people who could potentially buy what the FCC considers broadband, but they're not now. Why are they not? Are they not because it's too expensive or are they not because they're not interested in it? Are they not because they're not being marketed to because they live in a neighborhood where carriers don't deem those customers as valuable? That's a really interesting policy question to me that I think the Microsoft data can help elucidate and make clear, so . . .

Christopher Mitchell: There's, like, a whole slow food movement, but I don't think there's really a universe of people who are thinking, I really like my transactions on the Internet to take longer than they otherwise would. So presumably there is a reason there that they are not taking those higher speeds.

Derek Turner: Oh, sure. For example, my mother-in-law lives right outside of New York City in an area that's covered by Fios, Verizon Fios, it's covered by Optimum, Cablevision's network, and they have gigabit, essentially, cable available to her. They have gigabit fiber available to her, but because she was on an old modem that was a DOCSIS 2 modem, she wasn't showing up in Microsoft's tests as having that. And then I, you know, went there one day and said, "Hey, why don't you just call them up and say, 'I recognize you guys have faster speeds available. Can you send me a new modem and upgrade my plan?'' And they did without any cost to her, and so now she shows up in the data As covered. So there's issues like that. People that could afford it just didn't know about it, or there's people that could be on plans that they've been on forever and just haven't upgraded. To me, the more interesting thing is people who are looking at the tiers of prices and saying, "You know what? I am going to choose a 20 Megabits per second or 15 Megabits per second plan." Again, those are going away as progress marches on, but why are they choosing that? Is it price, or is it just because they don't actually need the faster speeds? I think that's a really interesting question that the FCC hasn't really decided to tackle because anything that relates to competition or looking at the market where people actually have broadband, it's decided that is not a issue of importance to them, unfortunately.

Christopher Mitchell: One of the things I want to make sure we highlight is that this truth, which you're right about — it's one of those weird things which everything can be true. Like, Form 477 is flawed, and yet, you can pull interesting, good, accurate data from it that matches up with the fact that I would say that we do need to find ways of improving access. But also from my perspective, I have a sense that there are a lot of people in rural areas that are paying often Frontier or Windstream or someone for a 15 or 20 megabit DSL connection, and they're only getting a few Megabits per second. But one of the things that I've found in talking with you about it, is that it seems like there's a lot of those people, but still they're not a large percentage of the American people.

Derek Turner: Yeah, that's right. And you actually raised a really good point there because we kind of had the FCC sniffing on this trail for a little bit, and now they've sort of gone away from it. And Sascha Meinrath and your research and other people have actually highlighted that cable used to have an issue as a shared network where they were advertising a certain threshold of speeds and they weren't able to deliver it. They've gotten a lot better with their technology, doing node splits and obviously DOCSIS 3 helps, opening up the megahertz size of the plant they're able to access. But DSL does have this inherent distance limitation, and they're a lot more susceptible to congestion when people are oversubscribing their networks. It's not as easy for them to go in and do a node split for example, like a cable company has. DSL, you know, about only has about a third of the fixed line subscriptions today. So maybe that's because it's a diminishing platform, it's not getting a lot of attention, but yeah, there's strong evidence that a significant portion of DSL customers are not getting what they pay for. So they may show up in the Microsoft data as not having that available to them when maybe they are buying that level of service and they're not actually receiving it, so that's again an issue that I would hope that the FCC's data could help us identify as one to look further into. Unfortunately, all the debate today is about rural and where we're spending the money. And that's important, but it's not the entire piece. There are rural folks that do have connections, and they're just a poor quality, and I'd like to help them as much as I'd like to help the folks who have nothing available to them whatsoever.

Christopher Mitchell: The FCC is about to improve the data collection. I think in their proposed rulemaking, they're suggesting that the new data collection, which will involve these polygons ⁠— and I don't want to spend a lot of time talking about this with you today, although we may come back when we get a better sense of what they're actually going to do. They have a sense, I think, that they want to sunset the Form 477. And you — I think I read all the comments in that docket, and you were the only one who really raised a concern with doing that.

Derek Turner: Yeah. You know, it's really unfortunate because we aren't the only folks who have actually made use of fully accessible public data at the block level. A number of other academics have used it. Certainly, it comes up in regulatory merger proceedings and things like that where industry is all too happy to have their payid economists go in and use the data. But yeah, so the commission has said, "Hey, we've got a new system here where providers will submit to us detailed polygon maps, and we'll use that to build a mapping database here that people can look at and decide whether or not broadband's available at a very granular level in their area." Well, great. Maps are useful for some folks. For researchers like me who want to do complicated analysis and demographic analysis, I need actual numbers. I need to work with something that can go into a database. And so, what we told the FCC is, "Hey, basically today, the way the submission works" ⁠— like providers, the ISPs you work for, I doubt they know what census blocks they operate in, in the normal course of their business. So what they do is the FCC has given instructions on how to actually take a shapefile, a map that you produce as an ISP because you know where your stuff is, and then transform that shapefile map into a list of census blocks using some open source software. What we told the FCC is, "Hey, if people are still just handing over maps to you instead of them doing the conversion to census blocks, can you do that and continue to release this database friendly information?" Because that is incredibly valuable to look at what's changing over time and to do all the kinds of, you know, demographic analysis that we like to do, that the states themselves have used. For example, California where I live has used this information to do a lot of demographic analysis on deployment. So please keep doing that in addition to gathering more granular information, because again, I can take a map of a particular census block that's partially covered and code that as a partially covered block. I can do things in a database with it that are still useful, and so that way everybody wins. You get your nice pretty maps that are very granular and detailed, and researchers like myself get access to the data that we need to track broadband deployment competition in this country.

Christopher Mitchell: But also the subscribership information. I mean that seems like it's just been forgotten in this discussion.

Derek Turner: And it wasn't clear to me and I haven't been able to nail it down — I don't think that the FCC is proposing to get rid of the subscribership information. If they are, they're doing that in a very sneaky way because they did not make it sound as if that's what they were up to. That would be a grave mistake because it's been 10 years now, but we did have this thing called the national broadband plan. A lot of folks have forgot about it, but it cost a lot of money to produce, a lot of good folks worked on it, and it produced a very big and detailed report. It had a number of recommendations in it, but in that list of recommendations were, "Hey FCC, take this subscribership data you have and analyze it better for competition. And not only that, let outside researchers come and look at it." Okay, maybe it's confidential to to know exactly how many subscribers a given carrier has at a very detailed level, but researchers access confidential data at the FCC all the time. So the national broadband plan team said, "Hey, let outside researchers access it." The commission dropped the ball on that totally. The national broadband plan team said, "Hey, go collect granular pricing information," because supply and demand, one of the key variables in supply and demand, which is what we're kind of getting at with Form 477 deployment in subscribership supply and demand, is price. You need to know how price impacts both the willingness to deploy and also the willingness for people to adopt. They've been told by the national broadband plan team, "Go collect price." They've even themselves sort of temporarily recommended that, "Hey, we will go collect pricing data," but for some reason in 2013, all that was dropped and it's kind of in the rear view mirror now. We've forgotten that that's actually one of the major points of going to the trouble of collecting this level of data is so we can actually use it to measure competition. And so if that's going away, I think I'll have to go in my closet and get my pitchfork out because that is unacceptable.

Christopher Mitchell: As you were saying that I'm just thinking, well probably it is an honest error, but at the same time with the agencies we're seeing under the Trump administration you have to be concerned that it may be something that just ends up not, you know, being — perhaps not even intentionally, but ends up being a way to drop key data that is available to the public right now. In my mind, the maps will be superior because from them we can derive the census blocks — and I agree with you the FCC should be doing that — but it will also give us this granular information in ways that we will be able to see, you know, at the which houses and which businesses and that sort of a thing are included. So, to me that seems like an improvement. I've had some pushback from people saying that I'm insufficiently concerned about the addresses. In my mind, I don't know what I'd do with a whole bunch of addresses. I know what I can do with maps. I feel like I can turn that into addresses to some extent. So as a researcher, you know, is the FCC more or less getting this right with the polygons and then assuming that they would take your suggestion to turn that into this census blocks as well?

Derek Turner: I think so. In most of the country ⁠— you know, 80 percent of the country is in an urban area ⁠— in most of the country you're not going to have the same issue of one part of a very small geography is covered and another is not. That's largely a rural issue, you know, so yeah, they're going to ask for these very detailed polygons. When providers are going to turn over these so-called polygons, they are going to be very detailed maps of where broadband is and isn't. It'll be useful on the other side of that, if the FCC goes forward with what they've annoyingly called the fabric part of this, which is essentially a separate detailed map of every single building structure, you know, an apartment building, a business, and deep knowledge of that, they're going to layer those two over one another. And so that essentially mimics what would be an address level-specific deployment database. You know, in the past we've asked the FCC to collect address level information. We've gotten a lot of pushback from the carriers as to why that's not as good as an option as we might think it is. They say, "Well you know, we don't actually keep our records the same way, address by address. We use GIS. We use billing addresses sometimes if if we use addresses at all." So I don't know the validity of those arguments. I think the carriers historically are just unwilling to give up any information unless forced to, so the fact that they're willing to hand over these polygons is fine, and that with the fabric information should give us an approximation of address by address level. I just hope at the end of the day that they understand and embrace the utility of having database level information so that either they themselves do these conversions to very granular census block level database information, or if they give it to us, they give us some ability to have that analysis done quickly because I've played around with some of the open source software on the mobile side — the FCC doesn't produce the data in the same way they do on fixed — and it's a bit of a hairy process to convert all these thousands of shapefiles into census block information. So the FCC is a big powerful government agency; they should be doing that. But I do think that at the end of the day, what the FCC has proposed will hopefully solve most of this overstated deployment issue, and then once that's solved, I guess I'm a little naive here, we can get on to talking about solving the competition problems, okay? We finally solved, you know, knowing where broadband is and it isn't. We're paying to get it into those areas where it isn't. Now, can we also focus a little bit on those who do have broadband but only maybe have one provider available to them who's charging them a ridiculous amount of money for it, or they're potentially not delivering on the service speeds the customers are paying for, or they have data caps that they're using anti-competitively. Those are the kinds of things I think the FCC has the data on to start enabling that kind of scrutiny and just historically have just been reluctant to do that.

Christopher Mitchell: Yes. Well, I'm relieved to hear that because that's what I thought. Nice confirmation bias for me, I'm not gonna question you further on it.

Derek Turner: Great.

Christopher Mitchell: But also just cause we're running out of time. So Derek, I really appreciate your time. I appreciate the work that you've done and that you're willing to correct allies and made sure that we're all moving in the right direction of not just making a good rhetorical argument, but one that actually is moving us toward having the data we need and using it correctly.

Derek Turner: Hey, I do it with love, and I'm also happy to be on the receiving end of correction as well because again, we don't succeed if we're operating from a basis of fantasy. We have to operate from a basis of fact, and I think the facts are on our side.

Christopher Mitchell: Yeah, that actually reminds me the last thing I wanted to note. In my office we were working on the FCC data on Pennsylvania when the new data set had come out that had Barrier as a company that claimed all of Pennsylvania. And we were having this moment of, like, argument internally when it happened because we were like, well, this is obviously not true. It's just not even physically possible. And so, we were trying to figure out how to rejigger the maps, so I'm really glad that someone took the time to actually tell the FCC rather than just arguing about it internally and passing it off as well. That's what you get today.

Derek Turner: Yeah. I mean it was — the FCC, at least this FCC chairman, should have a little bit of shame. Unfortunately, he was more than willing, and his staff, I don't know if they were afraid of him or what, but obviously as you mentioned, when that data came in, it was clearly a red flag that this was way overstated. All of a sudden — you know, the trajectory of data lines usually go up and to the right at the same slope. Well this one just kinked right up, and you know, something was wrong. And it was immediately obvious to you, and it was immediately obvious to me. We wrote a letter to the FCC about it and said, "Hey, maybe you shouldn't be patting yourself on the back here because it's in large part based on this erroneous data filing that you should have caught." And that's a big issue. The commission has quality control of their data, and they said, "Okay, we're going to look into it." It turns out Barrier's probably still overstating in their corrected data where they were, and in the most recent data set, they're not even in there at all. So they went from being nothing to deploying huge amounts of broadband, both fiber and fixed wireless, to then just slightly less, and now they're gone from the database. So clearly that identifies an area of quality control the FCC needs to work on that makes folks like me who are out there defending Form 477 look a little ridiculous because if they can't assure the quality of their data, then how can anyone rely on it? But I think at the end of the day, hopefully in the future as FCC leadership changes, maybe they will emphasize the need not only to collect the right data, but to make sure that what's being handed over is not terrible.

Christopher Mitchell: If they're not going to bother getting it right, they could do it more quickly. You know, why do we have to wait two years for something that's wrong?

Derek Turner: Exactly. That was one of the things that came up in the recent proceeding is the FCC actually said, "Hey, what should we do about chronic filers of bad data?" And we said, Hey, maybe one mistake is fine. You know, ask for some help. Two mistakes — well, what's going on there? Three, four, or five — well, maybe these people are being incentivized to be lazy if there's no consequences, you know. If you don't do your homework once, maybe your teacher will be nice, but if you repeatedly stop, then maybe she'll give you a failing grade. But unfortunately, I don't know if you've seen the record, almost no one supports the idea of penalizing chronically bad filers, and I think that's really unfortunate because what's the incentive to actually take the time to get this right if there's no consequences for just handing over bogus data?

Christopher Mitchell: Yeah. Well I'm with you on that and thank you again. It's been a wonderful conversation.

Derek Turner: Sure. Thanks for having me.

Lisa Gonzalez: And that was Christopher with Free Press research director, Derek Turner. We have transcripts for this and other podcasts available at muninetworks.org/broadbandbits. Email us at podcast@muninetworks.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 podcast from ILSR, Building Local Power and the Local Energy Rules podcast. You can access them anywhere you get your podcasts. You can catch the latest important research from all of our 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 380 of the Community Broadband Bits podcast.