Three years ago, Georgia state officials knew they had a problem. The state needed federal funding to bring broadband to unserved parts, but no one knew where those gaps actually were. The US Federal Communications Commission maps weren't precise enough, and Georgia didn't have any of its own. So those officials decided to build their own maps.
The state legislature formed the Georgia Broadband Deployment Initiative and passed a law that made it easier for service providers to share private, detailed information without tipping off competitors. Within two years, Georgia had built a map considered to be one of the most granular in the nation. And it didn't have to wait for the flawed national broadband map to catch up.
"We took the approach of getting [data] more at an address- or a location-level approach to get a better understanding," said Deana Perry, executive director of the Georgia Broadband Deployment Initiative. "The publication of the map has been very beneficial," she added, in supporting efforts to get students connected during the novel coronavirus pandemic, to help expand access to low-income housing and to get people access to doctors through telemedicine -- all areas that are vital as the pandemic continues to ravage the country.
The effort by Georgia officials -- taking it upon themselves to solve a problem that's plagued the federal government for decades -- underscores how some are starting to move with more urgency to solve the broadband gap. It's a problem that affects millions of Americans, and is particularly urgent in light of a pandemic that has forced most interactions, from classes to weddings, to go online.
While the federal government works to allocate $20 billion on top of billions of dollars in funding already earmarked for unserved communities, there remains a lack of understanding of where the problems lie. The FCC's broadband map isn't detailed enough to pinpoint unserved areas, and that's left millions of Americans behind. Unless the data improves, service is unlikely to improve.
The latest FCC data from January, which includes data provided through the end of 2019, determined that fewer than 14.5 million Americans -- or 4.4% of the population -- lack access to fixed broadband, which is defined as download speeds of 25 Mbps and upload speeds of 3 Mbps. But Microsoft, which looks at how quickly people across the US download its software and security updates, in December said 157.3 million people in the US, or 48% of the population, don't use the internet at broadband speeds. And broadband data tracker BroadbandNow a year ago estimated that at least 42 million, or 13% of the population, didn't have broadband at all, double the FCC's declaration at the time.
The flawed maps present a big problem as the government tries to distribute broadband funding. If a census block is considered covered by the FCC map, it's not eligible for federal assistance. It's proved to be an issue as the US allocates billions through the Rural Digital Opportunity Fund, or RDOF, which the FCC has called its "largest investment ever to close [the] digital divide."
Thanks to $65 million in funding from Congress in December, the FCC now will require internet service providers to share more detailed data, giving a better picture of what areas are unserved by broadband. It will also have to open the map to public feedback, letting people flag when something is wrong and providing more data points on gaps. Last week, acting FCC Commissioner Jessica Rosenworcel launched a new task force to fix the data, saying "it's no secret that the FCC's existing broadband maps leave a lot to be desired."
But some experts say the new mapping parameters still aren't granular enough, and the new maps almost certainly will arrive too late to help people during the pandemic. States aren't waiting around for the FCC.
That includes Georgia, Maine and Pennsylvania, which are building their own maps, as well as many other states around the nation. They're drawing on speed test data, specific information from ISPs about what homes they serve, and other resources to find out where their gaps are.
"The states are the ones who are innovating on this," says Peggy Schaffer, director of the ConnectMaine Authority, the state's effort to bridge its digital divide. "We know we can't wait for the feds to fix it. We waited, we're done, so we're moving."
Those FCC maps are often too vague to be helpful. The agency collects what's called Form 477 data from internet service providers, which contain information about where ISPs say they can provide service within 10 business days. It's measured at the census block level -- the smallest geographic area used by the US Census Bureau -- and if only one home in a census block can get service, the whole block is considered served and isn't eligible for public funds.
ISPs have long resisted sharing address-level data with the FCC, and still won't under the new plan. Instead, carriers will report on access based on regions called "shapefiles." The polygon shapefiles will be overlaid on census blocks to depict the areas where broadband-capable networks exist. Until the FCC publishes its new maps -- next year at the earliest -- it won't be clear whether those shapefiles are detailed enough to help.
"We saw so many folks just talk about shapefiles breathlessly, as if it was this magic technology that was going to solve everything," said Josh Stager, senior counsel at the Open Technology Institute, the tech arm of the New American Foundation think tank that pushes for policy and regulatory reforms. "What matters is what's in the shapefile. What's the underlying data? How granular is it?"
Even if a service provider actually shares address-level data, like in Georgia, there's still a gap in knowing what areas are uncovered. The Enhanced 911 database is typically used for mapping, but it contains information like addresses for barns or various other locations that aren't the top priority for connectivity. Few places have been able to put together a picture of what are addresses that actually need coverage, at least not on a large scale. Georgia is one of them.
"The Georgia broadband map is the most granular in the nation," Gigi Sohn, an FCC staffer from 2013 to 2016 under Chairman Tom Wheeler and current distinguished fellow at Georgetown Law Institute for Technology Law & Policy, said in a speech in late January.
Georgia officials realized they needed to find a way to know what addresses actually were homes, not post office boxes or structures where people don't live or work.
During its first pilot, the Georgia Broadband Deployment Initiative program aimed to develop a statewide master address file by working with local governments. But officials quickly found "that would take a very long time to develop," said Perry, the program's director. Instead, it turned to LightBox, a commercial real estate data provider that has data on all addresses in the United States. ISPs provided Georgia with information on the addresses they served, and the state then matched that with LightBox's data to identify locations that didn't have broadband.
Georgia "has what nobody else has, which is a map of exactly what [does have broadband], and equally important, how many structures do not have connectivity," LightBox CEO Eric Frank said.
Those maps wouldn't have been possible without the participation of Georgia's ISPs. Service providers have long said they couldn't provide granular data because it would hurt them competitively and financially. But Georgia's Achieving Connectivity Everywhere Act, passed in 2018, requires that ISP data be kept confidential and not shared beyond its use in creating broadband maps. Competitors can't see a service provider's footprint, but consumers can see if their area is covered.
The law "gave them a level of comfort to want to participate" in building Georgia's granular maps, Perry said.
Ultimately, granular maps would help both parties. The state would know where gaps were and where to direct its funding. And the ISPs would get that funding to improve their networks.
While Georgia has address-level data, what the public sees when it visits its map is an aggregated level of information that only goes as deep as the census block. But the state tweaked the FCC's definition of what an unserved census block would be. While the FCC's maps up to now require only one address to have access to service for a census block to be considered served, Georgia says a census block is only considered served if at least 80% of the addresses have access to broadband.
What Georgia ultimately found was 507,000 locations, or 10% of the state's homes and businesses, lack access to broadband. In rural areas, it's even worse with about 30% of locations unserved. The FCC, in its January report, said only 6.2% of Georgia locations didn't have broadband.
More than a quarter million locations identified by Georgia were not considered by the FCC to be unserved, Perry said. "Those locations would not be eligible for any FCC funding," she said. So now the state is exploring what resources are available to get those areas connected.
Not all ISPs are comfortable sharing granular data with state and local governments, even if they're presented in an aggregated manner like in Georgia. In those places, governments are turning to another method of mapping to identify coverage gaps: speed tests.
For states that don't have complete data from ISPs, one of the best options for mapping is crowdsourcing. Consumers can share what sort of coverage and speeds they have, typically through speed tests. Maine, Pennsylvania and others have worked with Measurement Lab (M-Lab) for its open, free tests. When someone Googles "speed test," the box that appears at the top of the results is powered by M-Lab technology.
The aim when setting up M-Lab over a decade ago was to show the FCC a prototype of how it could quickly and inexpensively use standard scientific research methodologies to derive a national broadband map, said Sascha Meinrath, a broadband data expert who serves as director of X-Lab, a future-focused technology policy and innovation think tank, and holds the endowed Palmer Chair in Telecommunications at Penn State University. Along with his role at the university, Meinrath co-founded M-Lab with Vint Cerf -- considered one of the founders of the internet -- and other researchers.
While FCC maps in 2017 showed Pennsylvania was blanketed with broadband, more than 11 million speed tests conducted by Meinrath and his team in 2018 found no county where at least half of the population had access to broadband. And it's not just a Pennsylvania problem.
"Our data is showing that 50% of the populace doesn't have access to broadband connectivity," not just in Pennsylvania but in the entire country, Meinrath said. "Right now we're in this bizarre situation where even though we know -- we know -- that there's something dreadfully awry with our broadband [data collection], as a matter of national policy dating back 15 years, we simply refuse to collect the information that would explicate that," he added.
Pennsylvania ended up putting together a statewide map to help providers apply for RDOF funding last year. Some ISPs shared confidential data with the map builders at Penn State, which they didn't publish on the broader map, and the commonwealth also relied on building data and download speed tests. It kept thd speed tests off its official map because of pushback from the service providers, said Harry Crissy, a Penn State University Extension agent for economic resource development who has worked on the commonwealth's broadband mapping efforts since December 2019.
Pennsylvania published its map in April, and 13 service providers received $368.7 million in RDOF funds in December.
"We [released] it knowing there were flaws … but we did it specifically for providers because we wanted them to bid in that [RDOF] auction," Crissy said.
Aside from getting funding, the maps are helping other groups identify gaps in connectivity. In August, the Pennsylvania Department of Education turned to Crissy's team for help identifying homes that didn't have internet service. It needed to find ways to get unserved students and teachers connected for remote classes. "We turned that around in like four days for them," Crissy said.
Crissy and Penn State have proposed creating a consortium of land grant and research universities for not only collecting data on broadband speed and providers but also tracking what it would cost to connect all unserved addresses in a census block. Clicking a block on a map would say something like, out of 85 households, 23 are served, and it would cost $60,000 to bring service to everyone else.
"What we've demonstrated in Pennsylvania is that. collaboration can really work," Crissy said. "This is evolving really fast. Look at us in about 60 days, and you'll see what I mean."
Maine started looking at mapping in 2018, with state officials holding "broadband boot camps" to learn about the population's needs and to figure out the best way to put together maps that were better than those from the FCC. Eventually, it settled on pulling speed test data to find gaps in coverage. The state ran a pilot test in Oxford County in October to see how the process would work and went statewide with its plan on Thanksgiving week. Since then, nearly 17,000 people have taken its speed tests from over 13,000 unique locations.
"The speed tests do two things," Schaffer said. "One is they give us better data, but two is it gives the community something to do." The system automatically puts the speed tests on Maine's map, letting consumers immediately see some sort of result.
"Before we instituted this … they were just like screaming into the universe," she said. "Now they can do something about it."
Speed tests aren't always that useful for address-specific data, but paint a picture of what an area looks like. "It's not as important that it's accurate per household as it is you have a bunch of them in a period of time in an area that gives you a much better picture of what that area looks like," Schaffer said.
By combining Maine's data and information from the FCC, the state estimates about 11% of the population, or about 85,000 homes, don't have broadband, or nearly triple the size of the FCC's tally. Maine plans to connect 95% of its population by 2025.
Maine this month is opening its first grants for unserved areas based on its new mapping effort. One of the four possible ways to determine if an area can get funding is if speed tests show it doesn't have broadband.
Georgia, meanwhile, benefited from its data's granularity even before its map went public. The information it possessed let officials respond quickly when the pandemic caused the state to shut down schools in March. Knowledge of what addresses didn't have home broadband allowed Georgia to figure out plans for deploying hotspots for students or finding other ways to get them connected. And it allowed schools to give their limited supplies of computers and other devices to students who would actually be able to use them.
"Within a week, we were able to bring seven different agencies together and were able to quickly share data … to see where students lack connectivity and understand what was the best way to deploy the solutions," Perry said.
It's a model other states may want to follow as the nation waits for the FCC's broadband maps to improve.