The default answer in the cable industry for solving "underserved" communities seems to be run fiber - but that's prohibitively expensive, especially for very rural communities. I believe "underserved" is defined as sub 32 MBps down and sub 6 MBps up. There's HFC which alleviates things somewhat but that still implies noise problems and signal amplitude loss for each home's coax tap.
It's nice to have Starlink as an intermediary solution for rural people while the "fiber everywhere" approach takes slow adoption.
Is fiber really more expensive to run than coax? I'd imagine that most of the cost is the cost of digging a trench or stringing out cables under the power lines. If you take a close look at the outdoor plant for a cable system there are a large number of large powered boxes that also cost money.
Contrast that to a fiber system which has very few active parts.
I'd imagine that coaxial cable still dominates in many places because it was installed in the 1980s and 1990s. When the cable company wants to upgrade the system it is cost-effective to replace the uppermost nodes of the network with fiber and still leave coax for the last half mile.
I know a rural "cable" company that today runs fiber optic plant in communities with favorable economics. Their primary product is an analog cable signal that is exactly what goes over the coax just modulated on a light beam. It's an economical and reliable system now and it is straightforward to upgrade to GPON or more advanced technologies in that future.
The question isn't _new_ runs of fiber vs coax, everyone just runs fiber for that. It's the cost of upgrading existing runs. HFC (hybrid fiber coax) networks have been the norm for a long time. When the original coax was run it was significantly cheaper to run it even accounting for inflation. Like all major construction projects (which is what a coax replacement is) costs in the US are spiraling upwards: https://www.bloomberg.com/news/articles/2021-12-08/why-build...
It might not be so much that the fiber itself is expensive. It might just be the conduit or other delivery mechanism that can be rather costly to run. But in general, fiber is absolutely a viable option for rural communities, usually if the project/roll out is being handled by some local organization (Co-op or municipality or as you sated a smaller cable company). I wrote my thesis on community owned broadband programs in Ohio and did a fair amount of research on the topic.
The one entity that is not going to invest in new plant is an entity that has an existing plant.
For instance if there is a phone company with a copper plant that can charge $60 a month for DSL it is obviously unaffordable for them to upgrade to fiber so they can charge $70 a month.
Some other entity might find the new plant is worth investing in but it still fights against the old plant. For instance the phone company is making crazy profits on $60 a month DSL and can probably still make profits on $40 a month DSL. This puts a downward pressure on pricing for a fiber service and also means the fiber plant is going to bypass many possible subscribers because there are plenty of people who will put up with a 100x drop in performance to save a few dollars.
works an interesting case study that was fundamental to the Democratic party becoming advocates for free trade. In particular, the steel industry in the US in the 1970s still depended on open hearth furnaces that were obsolete in every way to the basic oxygen furnaces used in Germany and Japan.
From a marxist perspective, capital is the "master class", maybe even the capital itself (as opposed to the capitalist), embodied in those furnaces. Instead of replacing the furnaces, the US steel industry pursued protectionism which was harmful to the competitiveness of other sectors of the economy: inferior and expensive steel is an ingredient for inferior and expensive cars, buildings, etc.
To analogize, however, the problem is that the copper phone network exists. So long as it exists there are strong incentives to keep using it. It might not be a problem of encouraging investment in fiber as much as forcing disinvestment in copper.
Do people even dig trenches anymore? I assume there is already a services pipe that runs along most city roads that has all of the various utilities in it. Am I wrong?
Of course! Around here each utility generally has its own pipe, and often they take different paths. There’s a mix of underground and above ground. The neighborhood is roughly 80 years old.
It's nice to have Starlink as an intermediary solution for rural people while the "fiber everywhere" approach takes slow adoption.