An Incremental Approach To Broadband Mapping

The debate over how best to map broadband continues to rage on, as evidenced by this most recent Wall Street Journal article.

Yet what's troublesome is how much these discussions focus on what Connected Nation's doing wrong rather than exploring the right way to do broadband mapping.

This is especially worrisome as I'd argue that no one has fully cracked the nut on the best approach to mapping broadband.

And given that we're about to spend hundreds of millions of taxpayer dollars on trying to tackle this challenge, I want to attempt to push this dialog forward in a more constructive manner by proposing a new approach to broadband mapping.

The key to my proposal is that instead of trying to have one entity attempt do all the mapping in one fell swoop we focus on an incremental methodology that starts with establishing a baseline of data and then encourages others to add to it.

More specifically, I want to see us start with the most basic question of, "Where's broadband?" Produce maps like Virginia's that show where broadband--or symmetrical service over 768Kbps--is available on as granular a level as possible.

Then make this data available to the public to build additional layers of data on top of.

For example, with this baseline in place I'd want to go and round up data from all the FTTH providers to show where their networks are available and what speeds they offer at what prices. Doing this could then inspire their DSL and cable competitors to step up and show what they have to offer rather than risk creating a vacuum in which customers flock to fiber.

Another layer of data could be actual speeds and/or quality of service, with data drawn from user speed tests like BroadbandCensus.com, a network of committed network monitors with more sophisticated monitoring tools, or any other mechanism. In this way customers could start to know more about if they're truly getting the service they're paying for, and ultimately do so in a granular way that can identify stronger and weaker parts of the same network.

Along these lines, someone could create a layer of data showing the usage restrictions like bandwidth caps of the various broadband networks on a regional basis.

Arguably even more important than these supply-side metrics would be the opportunity to layer on data related to adoption and usage.

By layering on data showing takerates for broadband we can identify areas where it's available but not being adopted so we can target demand-side programs.

Also, if we start collected more hard data about the actual usage of broadband we can figure out which communities are gaining the most benefit from using broadband in order to learn from them about what's working so we can share those lessons with other communities.

Building on this idea, we could also start layering on data about demand for bandwidth comparing it to how much supply is available, with areas where there's a gap between demand and supply clearly showing which markets are underserved by market forces alone.

Trying to collect all this data at once would be a monumental challenge. Not every mapping entity has expertise in all these areas, and by spreading out the work it can be done more efficiently.

Also, by taking an incremental approach with multiple entities layering data we can leverage the best ideas from every state. Instead of trying to be prescriptive in saying that there's only one right way to do mapping an incremental approach will embrace the fact that we're still learning what the best way to do this is. In this way each state can be encouraged to innovate in finding new layers of relevant data to add and new ways to make the layers of data more relevant to each other.

And perhaps the best argument in favor of an incremental approach to broadband mapping is that we can't have these maps be static. They're going to need to evolve over time. We don't need to just map broadband once, we need to be getting a baseline set today and then having it evolve so we can track our progress over time.

I can't say that I have all the details figured out for how this incremental approach to broadband mapping will work. There's still a lot to be hammered out, especially things like making sure the maps are standardized in such a way so as to be interoperable and so that the state maps can be collated into a larger national map.

But I do feel like we're going to be better served by taking this mapping process one step at a time rather than throwing our lot all in with one model or another.

Let's start with the most basic question of, "Where's broadband?" but instead of making that the endgame let's have that be the starting point upon which a robust, multifaceted, user-supported, incremental approach to broadband mapping can be built upon.