The move will immediately create a new, nationwide competitor to every fixed Internet provider across the country, whether it's a telecom behemoth like AT&T or a smaller, rural provider like Rise Broadband.
But the real question is: Should these companies be worried about Starlink?
After doing some digging, I think it's fair to say that Starlink doesn't pose much of a threat to the nation's established Internet providers, whether they're in urban or rural areas.
However, much remains unknown about Starlink – including whether the company has the financial wherewithal to go the distance – and so my initial take must include a big qualifier: Starlink doesn't appear to be a threat, at least not yet.
What we know
Starlink recently began asking potential customers for their addresses rather than their zip codes, which could indicate the company is getting closer to launching its service. Indeed, Musk hinted in April that Starlink could begin offering a private beta by July and a public beta by October.
Starlink recently told the FCC that it will "initiate commercial service this year across the United States."
But exactly what kind of service will Starlink launch? I had hoped to ask the company that question directly, but neither the Starlink nor the SpaceX websites have contact information for members of the press, and my attempts to get in touch with a Starlink spokesperson did not pan out.
That said, Starlink and its executives over the past few months have offered some details on how the service might work. First, customers will need to get a Starlink antenna and point it at the roughly 500 LEO satellites that Starlink has so far launched into space (the company eventually hopes to operate tens of thousands of satellites). Starlink hasn't provided many details about that antenna, such as how much it will cost, other than its diameter (roughly 19 inches).
For his part, Musk recently tweeted that the antenna has "motors to self-orient for optimal view angle. No expert installer required. Just plug in & give it a clear view of the sky. Can be in garden, on roof, table, pretty much anywhere, so long as it has a wide view of the sky."
But analyst Tim Farrar with TMF Associates, who has more than 20 years of experience in the satellite industry, said the installation of a satellite antenna may not be as simple as Musk makes it sound. He said the antenna likely will need to be attached to a roof or other stable structure, work that some users might balk at. Farrar noted that Dish Network employs hundreds of technicians for just this reason.
Further, Farrar said the motors needed to orient Starlink's antenna would likely raise the cost of the gadget, and could quickly wear out if the antenna constantly needs to reorient itself in order to track Starlink's LEO satellites as they whiz by overhead. Farrar explained that Starlink's satellites orbit the Earth every 80 minutes or so, which means each one will only be over the US for a few minutes at the most. He said it's not clear how much moving around Starlink's antenna will need to do, and that it will in part depend on how many satellites Starlink launches. The more satellites Starlink launches, the lower the chance that users' antennas will need to reorient themselves.
"Reliability of these antennas is going to be a major, major concern," Farrar noted dryly.
But once customers get their antenna, what kind of Internet service can they expect?
As noted by Farrar, a Starlink sales engineer told regulators in Nebraska last year that the service would provide rural customers in the state with speeds of 100 Mbit/s on the downlink and 40 Mbit/s on the uplink, but noted those speeds would "depend on how dense the user-base is within a region." The executive also said that latency "will be very low, ~30 ms or so."
That's certainly not bad. After all, I currently subscribe to a 100Mbit/s service from my local ISP and find it perfectly serviceable for streaming TV, playing online video games and occasionally doing work too, I guess.
However, Farrar explained that LEO Internet services – like most other telecom networks – are a shared resource. Meaning, the more users you put onto the network, and the more data they consume, the slower the entire system gets. Starlink, like any other telecom operator, will need to carefully balance how many subscribers it can handle with how much money it can get from each of those users.
"The question is, how many people can get 100 Mbit/s," Farrar noted.
This basic fact is why a number of telecom operators still impose monthly usage caps – those caps are ostensibly designed to ensure fair sharing of a finite resource.
What we don't know
Starlink officials have not yet said what their service might cost, and whether usage will be capped.
Starlink's Gwynne Shotwell hinted in 2019 that the company's pricing would be competitive: "Is anybody paying less than 80 bucks a month for crappy service? Nope. That's why we're gonna be successful."
Another major unknown is whether Starlink will be eligible for government funding for its offering – a key issue given concerns over Starlink's financial situation. After some lobbying on the topic, the FCC said that it would consider Starlink's service for its upcoming Rural Digital Opportunity Fund, albeit begrudgingly. The deadline for companies to apply for RDOF money was last week, but the FCC has not yet identified which companies might have applied for funding nor whether it will allow Starlink to participate in the program.
Finally, Farrar explained that another major unknown for Starlink is whether it will be able to offer service around the world. He said most countries follow the ITU's guidelines for spectrum usage, and according to the ITU's regulations OneWeb – recently back from bankruptcy – has first dibs on the spectrum Starlink wants to use.
"That's a big issue that needs to be addressed over time," Farrar said, explaining that some countries might eschew ITU suggestions in favor of their own, homegrown LEO companies, like Canada may do with Telesat.
As for whether existing ISPs should be worried about Starlink, Farrar said that LEO services like those from Starlink are often a "last resort solution" in rural areas, and typically can't compete with services in urban areas.
"Terrestrial ISPs probably shouldn't be very worried about this," he said.