How I'd Spend NTIA's BTOP Billions On Deployment

There's a lot more to NTIA's BTOP program than just deployment--including money for demand stimulation, public computing centers, and broadband mapping--but for now I want to take a moment and outline in broad brushstrokes how I'd spend the roughly $4 billion they have that can be allocated to deployment.

First I'd set aside $1 billion for best-of-breed testbed projects.

These projects will be selected based on the level of connectivity being deployed, the number of unserved homes getting connected, the amount of local buy-in based not just in dollars but in programs set up to make use of the networks and how much demand already exists for what it will deliver, and the potential for these models to be scalable if successful.

I'd look at who's successfully built sustainable networks in the past first, and I'd analyze applicants to favor those projects that are truly shovel-ready.

I also wouldn't necessarily worry about having these projects reach the largest number of people at the lowest cost as the point with these funds aren't to get everyone connected now they're to allow multiple models to establish themselves that we can learn from to inform our decisions of where to invest next.

As a fiber guy, my preference would be to focus on those projects that bring future-proof networks to unserved areas so we're not having to subsidize deployment again in a few years. To keep things simple I'd look at funding the ten best $100 million projects, or if there are a number of smaller ones with merit perhaps five $100 million and ten $50 million projects.

Then I'd take another $1 billion and focus it on projects that can get large numbers of unserved homes online as quickly as possible. For example, one possibility might be to peel off $100 million and use it to get as many Native Americans on Indian reservations setup with satellite. With that money 100,00-200,000 homes could be installed, possibly with computers too, and have their service subsidized for the next two years.

In general I'm not a big fan of satellite as I think it's too expensive, doesn't have enough capacity, has too much latency, and won't be able to scale to meet future demand for bandwidth. That said, it's going to take time to deploy big broadband networks, especially in areas as rural as reservations tend to be, so I'd consider using satellite in this instance so we can get people online at any speed ASAP, though I'd be very open to other wireless and even wireline technologies as well. The point with this billion is that it be used to get large numbers of unserved online as quickly as possible.

Then I'd take the last $2 billion and use it to start wiring and interconnecting schools, libraries, and healthcare facilities with fiber. By getting these community anchor institutions upgraded to 21st century connectivity we provide a way for everyone to get access to it even if it's not in their homes.

But I'd want to be smart in how we build these networks so that we're leveraging existing assets as much as possible and so that the networks we're subsidizing are designed in such a way so as to be easily used to extend fiber further into communities.

The simple truth is that T1s aren't sufficient to support the connectivity needs of these community anchor institutions. They need fiber. And by focusing a good chunk of money on these networks we can get a lot of good work done that will both afford more people access to big broadband while also laying the groundwork for easing future deployments.

So let's review:

- $1 billion for testbed showcase projects
- $1 billion for getting unserved online quickly
- $2 billion for wiring community anchor institutions with open fiber

The last point I'll make about this issue today is that I'd be careful about worrying too much about making sure every state gets an equal amount of funding. While there is language in the stimulus act mandating that every state get at least one grant, that doesn't mean they all need to get a grant for deployment.

The reason I say this is that I don't think it's wise to artificially restrict who's eligible based on their geography. For example, what if one state has two great projects that are certain to be successes and another state doesn't have anything comparable? I think we should be more worried about funding the right projects than about treating all states equally.

And in fact by doing this we can hopefully help inspire states that are lagging in their planning to get their act together. I don't want anyone to feel like they're entitled to get money but instead believe we should reward those that are truly ready to make a difference in their communities wherever those communities may be.

This doesn't mean we should ignore the rule saying every state should get a grant, just that we shouldn't worry about every state getting a grant for deployment.

So there you have it, a very rough, very broad look at how I'd spend that $4 billion on deployment in a way that can give us the greatest bang for our buck.

What do you think? How would you spend this $4 billion on deployment?