When local communities apply for funding to improve local Internet infrastructure, grants and loans are often predicated on the need to deploy to unserved and underserved premises. Whether it’s federal, state, or local sources, Federal Communications Commission (FCC) data determining whether or not a region has access to broadband is often the data that funding entities rely on. In recent years, it’s become apparent that FCC data grossly understates the lack of accessibility to broadband. Finally in August 2019, the FCC called for comments as they reconsider how to collect fixed broadband data. The Institute for Local Self-Reliance teamed up with Next Century Cities and several other organizations with whom we often collaborate to submit both Comments and Reply Comments.
Fixing the Bad Data
We’ve covered this before, and the Commission has now decided to make changes. Traditionally, FCC data on broadband Internet access has been collected from Internet service providers (ISPs) that self-report on the areas they serve via Form 477. If a company has the ability to serve one premise in a census block they report to the Commission that they serve the entire block. Reality, however, often does not reflect such a high level of connectivity in one area.
When FCC data incorrectly determines that locations have the ability to subscribe to one or more Internet access companies, those areas lose eligibility for grants and loans for Internet network infrastructure. Sadly, these places are often caught in a strange purgatory between faulty FCC data and reality in which they can’t obtain funding to build out high-quality Internet access, and yet large Internet access companies don’t consider their areas a good investment due to low population densities.
For years now, the Institute for Local Self-Reliance and other organizations have worked to bring attention to the problem. A few lawmakers have pushed for change and several states, including Georgia and Vermont, have launched efforts to assemble their own mapping data to challenge the FCC’s. Whenever there is a conversation about improving the data, it usually involves taking a more granular approach.
Finally, in August, the FCC announced that they would take steps to improve broadband mapping and would christen it the Digital Opportunity Data Collection (DODC). As part of the process of establishing the new approach, the Commission sought input from interested stakeholders [PDF].
Working Together to Advance Data for Accessibility and Adoption
The Institute for Local Self-Reliance has worked with Next Century Cities, the Benton Institute for Broadband and Society, the National Digital Inclusion Alliance (NDIA), Access Humboldt, The Center for Rural Strategies, the Southern California Tribal Chairmen’s Association, and The X-Lab to submit suggestions to the FCC that address ways to improve the Commission’s data collection and handling techniques for the DCOD.
You can download the Comments here and Reply Comments here in order to read the full documents. We’ve summarized some of the most compelling points the group provided to the FCC.
The FCC proposed a definition of service availability that is more precise than the census block approach used now, and we supported that approach with an enhancement. Scaling down to a more granular data collection technique has been the recommendation of practically every entity that has criticized the FCC’s census block approach. NCC, ILSR, and allies suggest that the Commission implement a secondary category based on an ISP’s possible future service based on sufficient demand. We also recommended that the FCC allow Internet access companies to report buildings where they are prevented from offering service but would like have to the opportunity to connect subscribers. With a better idea of where ISPs are interested in serving subscribers and where those entities would be more inclined to deploy, the FCC can determine locations ripe for investment.
Keeping Track of Errors
What is more important than gathering data is making sure that the data is correct and that errors are corrected. We suggest that the FCC take steps to make data public sooner, including audits, to confirm data accuracy. As we’ve found when developing maps as resources, FCC broadband data is typically 12 or even 16 months old. Innovation in technology happens quickly and, as stewards of the data that researchers need to study the networks that innovation relies upon, the FCC needs to update its processes to likewise respond.
Our group also believes that the FCC could use crowdsourcing to confirm connectivity and report errors, which would speed up the process of sharing data, help root out errors, and reduce the inclination for errors.
When errors are repeated, we feel that entities that continue to make mistakes should face a series of consequences to encourage more careful reporting. There’s too many people relying on the accuracy of the data to allow repeated errors, regardless of the reason or source of the error. In our comments, we suggest that the FCC establish an error threshold as a standard instead of focusing on punishing the reporting of intentional errors. Because small and large Internet access companies have varying levels of resources to direct toward resolving errors, we suggest a transparent and sliding threshold scale based on number of premises served. This approach presumes that larger providers have more resources with which to fix errors and a greater ability to dedicate significant resources toward checking and double checking data prior to submitting it to the FCC.
Focusing on rooting out intentional errors is more complicated and extremely difficult to show, creating a system with almost no accountability. With no consequences, the propensity to commit errors continues.
We also suggest that local governments and states that collect their own data should be allowed to submit the information, and that it be assumed accurate. The FCC could use these valuable resources to compare to the ISP filer’s data.
Even if all these suggestions are adopted, however, there’s still a chance that something will go wrong and we suggest that there be a system in place for tracking complaints and resolutions. We support the notion that the Universal Service Administrative Company (USAC), which currently collects and delivers Universal Service Programs, would be a good candidate to create the system. Again, allowing local governments, tribal authorities, and states to submit bulk data about errors — because often their facilities encounter them — would provide invaluable information about reporting problems.
Include Pricing Data
The FCC has been called on repeatedly to collect data on how much subscribers pay for Internet access. In the past, we’ve suggested the Commission collect pricing data when we’ve commented on other proceedings and took the opportunity to once again point out why it’s important.
Primarily, the FCC is tasked with taking the steps necessary to encourage expansion of broadband to every premise in the United States by reducing barriers to that end. Research shows that affordability is one of the highest barriers to a large percentage of Americans. More transparency in pricing data would improve the usefulness of the FCC’s data.
Agreeing and Disagreeing
There were other organizations that commented and when we submitted Reply Comments we thought they had made especially important points that needed our support. Others stated incorrect assumptions that we thought should be corrected.
Several groups proposed measuring latency, and we agree. As innovation continues to advance and new applications become based in real time, or as voice applications become increasingly popular, latency is an important issue to subscribers. Hughes Networks Systems, that operates satellite Internet access, suggests that latency “doesn’t matter” to subscriber experience, but we find that people who subscribe to satellite vehemently claim the opposite.
When it comes to satellite Internet access, we agree with with Free Press’s suggestion regarding satellite and competition: satellite Internet access is a last resort. The Free Press suggests [PDF], and we concur, that by compelling satellite companies to provide list of zip codes where they actually serve premises, the FCC could better determine the state of competition in specific area.
In our Reply Comments, we noted that we agreed with the NTCA’s statement about taking a realistic view about toward where ISPs offer broadband was the best approach. Instead of allowing companies to report where broadband access “may” be available, they should be required to report where it “is” available. With a frank assessment, the FCC will not preclude Internet access companies from eligibility and will obtain a more accurate picture of the state of broadband coverage.
Read more from our Comments and our Reply Comments. We also encourage you to read the Comments from the Free Press and to check out other contributions at the FCC’s website.
