Recall that you won't be able to use starlink on any mobile phones Yet.

In Germany it's 1200 Euros per year...

What does your figure show? It is missing a y-axis and a title/caption. I assume the x-axis represents years CE. Something is increasing. What is it?

@Gerrit, that graph illustrates that 99.9% of many nations have 3g web access.

@aliential 20 years ago, your cellphone would have consumed hundreds of watts and would have needed a room and a desk to support it. See how time changes the rules?

I live in the US and I don't have access to the internet if I drive out just 30 miles out of Seattle into the mountains. That's where Starlink would come in - allow seamless Internet browsing absolutely anywhere and at any time, as long as there's antenna nearby.

This answer started out fine and then utterly cratered at the end with a bunch of complete nonsense.

it will have nearly zero upload speed, and 99% of developed populations already have 3g everywhere, and poor nations can't afford starlink. . Wrong. Wrong. And completely wrong. There are large parts of the US where starlink is a vastly superior option even now before the network is finished. Poor countries aren't going to pay the same price as US. Upload speed is fine.

@J...: It’s a matter of transmit power, antenna size and you need a directed beam. Pretty much impossible to do in a mobile phone.

@Michael Yet. Probably never with the satellites in the sky now. The sats in the sky now have a four year life expectancy. Newer ones will replace them. Technology improves continuously. THz has big potential.

@J...: Only if our understanding of physics changes. What’s the range of THz communications? How does it perform through clouds or rain?

@Michael Garbage through the atmosphere, to be fair. But even now there's nothing stopping Elon from dropping Starlink towers everywhere to provide 4G/5G service from space. Once it's up and working it's only a matter of time before eyes start shifting to the mobile market.

Ahh, spectrum pollution, the pollution that nobody cares about because it doesn't kill baby turtles and give kids cancer. Years from now they'll will look back and say "WTF were those morons thinking?" But we are complete idiots, and are going to crap in our own nests just like we always do.

@barbecue Spectrum pollution kills by worsening weather forecasts.

The atmosphere is opaque to THz/submm radiation due to the high water vapour absorption, therefore this cannot be used to provide telecommunication from space (except on high desert mountains, huray?)

@gerrit Only if you actually engage in long-term thinking, something we tend not to do as a species.

@barbecue Good that those satellites last only 4 years. All burning up in the upper atmosphere, dramatically increasing pollution in this sensitive part of our atmosphere that has little interaction with the rest. It's probably fine — what could possibly go wrong?

@eps, ok the first statement i read recently, it's 20mbps upload, which is about 10 times slower than optical fiber, i said it's slow, it's not too bad though! The other statements are correct: 99.9 percent of us can surf 1080p for cheaper using phones, and poor nations are fast developing nationwide 4g coverage which costs 5$ per month. selling 500,000 tesla economical cars versus 40 million expensive skyweb boxes is not the same! Starlink needs 40million customers to break even at 360 per year PLUS the 100/500 equipment cost.

@gerrit satellites will burn up completely in the atmosphere

@J... Yes, of course they burn up — that's the problem. They might leave a lot of pollutants there that go on to destroy ozone. So far the problem has been negligible, but it might not be if we increase the mass burning up in the mesosphere or stratosphere by a factor 1000. Burning is not a safe way to dispose waste. See the link that I posted, and sources therein.

@gerrit Agreed, but it remains a potential future problem rather than a real problem at the moment. It seems to be on the radar of the relevant people and ongoing technological development seems to be pacing such that we can reasonably expect that solutions will be forthcoming before it becomes a serious problem. Everything humans do is disruptive in V1.00, but we're also quite good at refining away those disruptions. Unlike CFCs, which we walked blindly into before we figured out it was a problem, this seems like one where we're planning sufficiently into the future to steer around.

@J... Given our record in the past, I do not share your technological optimism about our future.

@gerrit What can I say, I guess I'm an optimist. :)