While I was aware of the project, I was not aware of the details. I decided to look into it when Starlink was referenced in the Spinlauch thread.
The numbers are so far off I think Starlink is basically a scam to relocate money from investors (and governments).
This is one of those things where there is a thin line between amazing progress and amazing failure. Musk has previously described the cash chasm that Starlink must past through before reaching profitability. For success he must get large revenue streams from Starlink, which requires large constellations of satellites. Nothing technically impossible about this, but getting it done before the cash runs out is a real problem. Apparently he was counting on Starship to get this constellation into place quickly and relatively economically. Again, doesn't appear to be technically impossible, but requires solving a great many problems in a hurry. Faster than anyone has solved them before. He has already done many of the things, but there are a lot to do. Some of which (regulatory issues for example) are not really anything he has a lot of control over.
Besides what I quoted above, I came to conclude that the numbers just doesn't add up. The COO of Starlink "Ms. Gwynne Shotwell, has outlined that the total addressable market (TAM) for her company's Starlink satellite-based internet service is $1 trillion."
https://wccftech.com/starlink-is-playing-in-a-1-trillion-market-crucial-for-spacexs-core-aim-outlines-executive/I suspect the server farms will likely not move away from the reliability of land-based connection they currently use. The current board band market is more suitable to estimate the Starlink's market potential.
According to BusinessWire.com, 2021 global boardband market is merely $271 billion.
https://www.businesswire.com/news/home/20210823005338/en/Broadband-Internet-Services-Global-Markets-Report-2021-C-Band-Ku-Band-Ka-Band---Long-term-Forecast-to-2025-2030---ResearchAndMarkets.comAccording to Wikipedia (2019 latest year shown), 86.6% of the developed world is already internet connected. In the developing world, 47% is internet connected.
https://en.wikipedia.org/wiki/Global_Internet_usage
Observably, there is more internet connected by now, but let's use the smaller 2019 but more generous (for Starlink) assumption that $271billion at 86.6% (developed) and 47% (developing), the
growth capacity is merely 14% in the developed world and 53% in the developing world. Even if you
pretend the entire world at 53% growth capacity, you get $415 billion USD. That would be the entire board band market: fiber, cable, satellite included; merely 415 b$USD.
Looking at these numbers, I think $1 trillion market is a pipe dream, even $415 billion (entire board band market with fiber+cable+satellite) is a pipe dream. In a city, satellite is not very suitable. They are left with mainly suburbs or edges of cities. In my opinion, if they get as far as 1/3 of all board band, at $138billion that would still be a very generous dream but still a dream.
They are estimating up to 50% back haul traffic and 10% local.
https://en.wikipedia.org/wiki/Starlink#2014%E2%80%932017This left me thinking they are betting a lot of B2B traffic. I have no hard-data on how executive in those individual businesses will decide, but I suspect much of those will actually go to or stay on existing land-based ISP back bone instead of via satellite with less reliability and slower speed. I think this is the uncertainty that they are exaggerating, and exaggerating the growth of internet in the poorer developing world. Exaggerated to a point I think it smells scam.
Does it look like a money relocation scam to you?