Somehow I suspect SpaceX did the math on how many users it can support and the cost of replacing satellites, before betting the company on the project.

Yeah, like every start-up clearly did before them (example: Uber). /s

What?? Noooo... they would only have to launch indefinitely something like 10 rockets a month to keep replacing aging nodes, I am sure that is financially viable way to compete with a copper/fibre cable -_-

The tech works, its just completely non feasible. Geosync satelites can give you global access with handful of satelites.

Trans-sea cables give you cheap and quick connection speeds.

Starlink is middle of the road solution that solves both problems are immense costs.

I can only imagine high frequency traders being able to afford it vs utilise the advantage of the latencies.

It’s certainly not going to require 10 starships/month. And they’re aiming to get those down to <$10M/starship launch.

HFTs use direct, point-to-point terrestrial connections. They rarely need to send trades out of <middle of nowhere, that does not have a fiber link>.

They've already optimized the hell out of their latency, to a ridiculous level.