This NewScientist (12 December 2020) article more or less says it with its title:
Google’s AI can keep Loon balloons flying for over 300 days in a row
Huge stratospheric balloons that act as floating cell towers in remote areas can stay in the air for hundreds of days thanks to an artificially intelligent pilot created by Google and Loon.
Which means Internet coverage.
But does it mean that Elon Musk’s Starlink project can be abandoned? It’ll depend on the cost, bandwidth, latency, coverage, and dependability of Google’s Loon balloons.
The reason I don’t like Starlink is that it pollutes lower Earth orbit with thousands of satellites. These interfere with astronomical projects, and when they burn out, they have to be replaced at a fairly high cost. And I’m assuming that they’ll burn up rather quickly once they become inactive; if they don’t, then they enter the category of space junk, which is a far more important category than most folks realize. Getting and keeping satellites in orbit is already something of a challenge; adding more space junk makes things even more difficult.
Meanwhile, Loon balloons are presumably salvageable. However, they do use helium – a resource that is becoming more and more valuable. That may weigh against them, and I’m not aware of a more economical replacement.
