When SpaceX first announced Starlink in 2015, the reaction from telecommunications engineers was largely skeptical. Low-Earth orbit (LEO) satellite constellations had been tried before (Iridium, Teledesic) and had failed commercially. But Starlink did something its predecessors couldn't: it iterated fast, launched cheap, and actually delivered on latency.
Why Starlink is different
Traditional geostationary satellites orbit at 35,786 km — so far away that round-trip signal latency is 600ms or more, unusable for video calls or gaming. Starlink's satellites orbit at 550 km, producing latencies of 20–40ms — comparable to ground-based broadband. The tradeoff is coverage: at that altitude, each satellite covers a much smaller area, requiring thousands of satellites to cover the globe. SpaceX has launched over 5,000 so far.
The real-world impact
Starlink has provided internet connectivity to remote communities in Alaska, the Amazon, and rural sub-Saharan Africa that had no viable alternatives. Its most dramatic deployment was in Ukraine, where SpaceX provided terminals within days of the 2022 invasion, maintaining communications infrastructure that proved critical for military coordination. This wasn't in anyone's original product roadmap.
The astronomical and regulatory concerns
Astronomers have raised legitimate concerns: 5,000+ satellites create visible streaks in long-exposure astronomical images, and the eventual constellation size (SpaceX has FCC approval for up to 42,000 satellites) poses collision risk and complicates orbital management. The ITU's spectrum allocation rules are under intense pressure as every major tech company (Amazon's Kuiper, OneWeb, China's Guowang) races to claim orbital slots.