There are [still] too many locations in the National Broadband Map
For many months, states have raced to add locations to the Federal Communications Commission's National Broadband Map, trying to maximize their allocation of broadband funding from the Infrastructure Investment and Jobs Act (IIJA). But in a few short weeks, the National Telecommunications and Information Administration will allocate funding and then my prediction is we’ll never hear about “missing locations” again. Instead, we’ll focus on the opposite problem: there are too many locations in the National Broadband Map. Overall there are 113.3 million Broadband Serviceable Locations (BSLs) according to the Broadband Service Location Fabric which underpins the Broadband Map. Because one “location” can represent more than one housing unit, such as in the case of an apartment building, there are 158.4 million “units” according to the Fabric (BSLs explicitly include businesses, so the Fabric “units” is both housing units and business locations). The Census, which also tracks the number of housing units in the country, reports there are 140.5 million housing units. The broadband Fabric is an estimate. It’s a model. It uses the available data — parcels satellite imagery, tax records, etc — and makes a guess at where a “Broadband Serviceable Location” exists based on that data. Some error, both leaving out some locations and including too many, is expected. This is particularly true since the definition of a “Broadband Serviceable Location” is itself squishy.
There are [still] too many locations in the National Broadband Map