This thought process makes some sense to me, though I have doubts about the implementation. Is fiber really the best fit for rural areas? What about some kind of wireless solution? Surely that would be more cost-effective?
Another subtle problem with rural wireless is that you have to point to a well-connected area, so the cheapest way to service houses that are on the side of the hill facing away from town is often either fiber or putting the tower in orbit. Otherwise, you'd need to run a fiber trunk line to the middle of nowhere to support the ISP side of the wireless link.
Fiber has the property that you can run it in parallel to existing power lines, so any house with power already has an existing right of way to piggyback on (owned by the wrong person, but imminent domain exists). Also, you can choose to trench the fiber or run it on telephone polls.
If I were running things, I'd look into strategically deploying fiber-to-the-home to areas with good line of sight to poorly connected areas, and then use a hub and spoke network topology where the spokes are directional wireless and some of the homes are hubs.
That creates a second order problem though: You need backup power at the hubs or parts of the network will go down when the electricity goes out. Long term, it's probably cheaper to maintain a network of buried fiber than it is to maintain the electricity infrastructure needed to keep the towers online.
When it comes to broadband there is a question of “how good is good enough” but for the people living in Hamilton Montana (population 5k) I hope there can be a path to infrastructure investment that leads to high quality internet in town.
During high demand, our local WISP has 1 second latency to the internet because their wired backhaul link is terrible. It's less than 1ms to the tower. Starlink to the internet is in the 20-30ms range, and it's bouncing into orbit and back.
[1] https://www.eff.org/deeplinks/2020/06/why-slow-networks-real...