Starlink performs amazingly well in its use-case.
Satellite is OK if there isn't anything else. On average, it doesn't beat 4G/5G.
Starlink sucks in high service density areas, that's not news [0]. It also averages 100Mbps and tops at a theoretical 500Mbps, whereas 5G averages 250Mbps and tops at a theoretical 1000Mbps. Latency is also considerably higher for Starlink (~60-100ms), than for 5G (~1-50ms).
[0] https://arstechnica.com/tech-policy/2022/09/ookla-starlinks-...
The economics are boarder-line but it should be still be given a fair chance at success.
This is not a technical or economic issue, but a regulatory one, and it is not new either.
> Starlink satellites are small and cheap to send to orbit. Its still more expensive than wiring an urban apartment building with fiber but it is absolutely the only way we have to get these people connected.
This is absolutely not true. Quote [0]:
> We assume that the territory of the continental United States can be served by 5 percent of the satellites in the constellation, and in this case, the constellation should consist of 4,229 satellites.
This only accounts for rural US population.
As of August 2023, the total number of satellites in the constellation is ~4500 [1], with an estimate cost of ~$500,000 per unit, including launch [2].
A 5G tower costs roughly a fifth of that [3], while providing higher bandwidth and dramatically lower latency.
Again, this is just another example of corporations sucking from the government, and giving nothing in return.
[0] http://www.satmagazine.com/story.php?number=1026762698
[1] https://www.space.com/spacex-starlink-satellites.html
[2] https://ts2.space/en/how-much-does-a-starlink-satellite-cost...
[3] https://www.steelintheair.com/5g-cell-towers-in-2023-top-que....
I don't know that, but it is still more expensive than deploying 5G towers, as I have explained in my previous comment.
> Do you think we should bother to connect these people or not?
I never said rural areas don't deserve broadband service, I'm not sure where you got that from.
I'm saying that their best chance to get fast, widely available broadband services, is to force telecom companies to deliver on the promises they made when they received billions in subsidies.
I'm also saying that Starlink is not the solution, because it is more expensive, slower, and less reliable than 5G, and unless a significant investment is made, it will get even less reliable over time for a number of reasons.
Also, we seem to be forgetting that Starlink is not even profitable as of 2023, and will need hundreds of millions of dollars in new investments to finally break even. Since, as you said, rural internet is not profitable enough to invest money in it, unless Starlink gets some federal money, their chances of providing a sustainable, long term service to rural communities, is next to nil.