Eg if it is $10m to LEO and $15m to HEO you will pay the extra to get to HEO and put a satellite designed for long life - but if it is $1m to LEO you can build a cheaper satellite and launch it 5 times over the following decades.
I like the idea of having internet on my sailboat as much as the next guy, but it takes some serious cognitive dissonance to convince oneself that 42,000 pieces of nearby junk is better than a handful of pieces of far away junk.
No. First, The whole reason they need to maintain a fleet, rather then just having it sit there for decades, is that they're so low. The vast majority of planned Starlink satellites from that number are from Phase 2, and are V-band VLEO sats with orbits around 340 km, which is really low. At that altitude, natural decay time is measured in weeks at best. To work they'll need both active thrusters providing regular boost and aerodynamic low drag design (and maintaining orientation for that will itself require fuel). Should they actually lose all control through malfunction or a collision, they and resulting debris will deorbit very rapidly (they'll both have no boost and be less aerodynamic).
And second, "through malfunction or collision", because it's not as if SpaceX (and other LEO satellite operators) doesn't have, and indeed are required to have, plans for controlled deorbit at EOL. Most of the satellites can be expected to get deorbited in a controlled way as planned. Making sure everything burns up has been one of the things slowing Starlink development, it took some work to ensure the optical links would properly go for example.
SpaceX is not interested in leaving them up there. I mean, FFS people, who would be hurt more than SpaceX by Kessler Syndrome!?
>I like the idea of having internet on my sailboat as much as the next guy, but it takes some serious cognitive dissonance to convince oneself that 42,000 pieces of nearby junk is better than a handful of pieces of far away junk.
Your dismissiveness towards hundreds of millions of underserved people and challenging use cases and ignorance of orbital dynamics does you no favors here.
This is pompous and elitist nonsense of the highest order. People have other more pressing concerns than fast internet. You know...like water, food and cheap power.
Also it's worth noting that communications networks is but one application for LEO satellite constellation technology.
With some gyroscopes, I think they only need electricity for it, not actual reaction mass.
Imagine thinking that a product that costs $500 up-front and $99/mo thereafter and uses ~100 W constantly is designed for "hundreds of millions of underserved people".