In the most remote parts of the U.S., reliable internet is still hard to
come by. Forty two million Americans still don't have access to broadband. This is a 9.5-acre homestead in Idaho. And when we first got here, we had zero cell service. We tried to put boosters at the top. We tried to put boosters at the bottom. We tried to put boosters anywhere we could. We had no cell service. So we really needed some good, reliable internet. And then Starlink came along. Mike Lorden and his fiancee, Liz Racer, are one of more than 10,000
customers using Starlink, SpaceX's ambitious project to build an interconnected network of thousands of satellites to provide high-speed
broadband internet across the globe. They've been using Starlink's service for about six months now and are
documenting their experience of living in a homestead on YouTube. There are some times when the Starlink does drop out. The dropouts aren't really significant. It seems to be just like short little blips, you know, a couple seconds
here and there. The only time that you really, really notice it is if you're like live
streaming or if you're using the Wi-Fi calling option. Other than that, you know, if you're just surfing the web or something, it
seems to be very fast. So far for what it is, it's a real good connection, you know, considering
our circumstances. Starlink is still in beta and currently serves select customers in the
northern U.S., Canada, the U.K., Germany and New Zealand. Experts estimate that there are around 70 million households worldwide
that are good candidates for satellite-based consumer broadband and have the capacity to pay. Amid a global pandemic that's kept employees from going to offices and
children from schools, the need for universal broadband has become undeniable. If the service expands to its intended global customer base, Starlink
could be key to SpaceX's success and Elon Musk's vision for a colony on Mars. I think it's fair to say that the majority of SpaceX's valuation today is
tied to the Starlink business model. Generally speaking, the global launch industry is about a $5 billion/year
industry. SpaceX, in previous discussions has talked about a $30 billion/year
opportunity in the Starlink business, looking about three to five years out. The basis of Starlink's internet service involves three components: a
satellite dish, ground stations and the satellites themselves. The service is meant for customers like Lorden and Racer, who
live in a sparsely populated area that's not being served by traditional internet companies. Given the high cost of laying cable or fiber, which can be as much as
$20,000/kilometer, terrestrial service providers tend to focus on urban and suburban areas where there's high density. It simply does not make economic sense to reach out to consumers in
low-density areas. The only option we really considered was we had a local internet company
that wanted to come in. They would have had to erect a tower on the top of our mountain. They maybe could have given us five megs a second and it would have been
about the same monthly price as the Starlink service. We have other friends that are in similar situations as us, and they have,
you know, those other satellite internet providers and a lot of them are data capped. Most of the rates are real high. Starlink customers pay $499 for the hardware needed to connect to the
network and an additional $99/month for the service. Currently, there are no data usage caps or contracts. So far, SpaceX has launched over 1,000 satellites into orbit, but the
company plans to deploy 4,425 satellites by 202 4. By the time that SpaceX is done building out its global Starlink
satellite system, known as a constellation, the company will have launched about 12,000 satellites. And although not yet approved by the FCC, SpaceX has requested permission
to launch an additional 30,000 satellites, which would bring their total to 42,000. Unlike traditional internet satellites, which are as big as a school bus
and orbit at around 36,000 kilometers above Earth's surface, Starlink's satellites are much smaller, and closer, at about 550
kilometers above Earth's surface. But the satellites' closer placement means they can see less of the Earth
at any given point in time, which is why SpaceX needs so many. On the flip side, SpaceX claims that this closer orbit allows the system
to have a lower signal delay, known as latency. Latency is important for things like video streaming and gaming. Lorden says speeds vary according to factors like weather. But on average, he's been getting speeds of about 75 megabits per second
for downloads and around 12 megabits per second for uploads. In a filing to the FCC, SpaceX said that Starlink's service is, quote,
meeting and exceeding speeds of 100 megabits per second for downloads and 20 megabits per second for uploads. Musk has said that he expects the service will double in speed by the end
of the year. To do this, the company is developing a technology known as intersatellite
links. The basic premise is that you have an optical sensor, which is effectively
a laser, that points from one satellite and connects to the optical sensor on another satellite and
creates this, you know, unseen path from one to the other. That's this stable connection, almost like a wire. And if you you have a stable connection, you can drastically increase the
speed because you're now moving signal from one satellite to the other at the speed of light rather than
the speed of whatever is on the ground. Another benefit is reducing the amount of infrastructure needed. Given Starlink's current architecture, we estimate that they will need more
than 100 ground stations in the U.S.. If, however, they move to the laser communications on the satellites, they
can greatly reduce the number of ground stations needed, as well as the complexity of the overall system on a global basis. On the consumer side, Lorden says that working with the technology was
simple. SpaceX does not currently provide an installation service, so users are on
their own for the most part. They ship and deliver a big ol' box to your door and inside that box, it's
really not that many items. It would be the Starlink dish itself, the router, the power supply and the
little tripod mount to set the dish on. And it's super straightforward. Starlink actually has an app that you can download on the phone and it
walks you through everything. Literally, just plug this thing into the wall and put the dish where the
app tells you to put the dish. Starlink's dish comes equipped with a built-in heating element, which keeps
it free of snow and ice. It's also motorized, allowing the dish to automatically orient itself to
stay aligned with the satellites overhead. But making its equipment user friendly has not come cheap for SpaceX. The biggest challenge that SpaceX is facing with Starlink is also one of
the big unknowns about the actual equipment cost itself. People aren't sure exactly how much either SpaceX is paying
for, or paying to build its antennas. There's been a lot of speculation that it's a thousand dollars a unit, two
thousand dollars a unit. If they're, you know, selling these kits for $499/kit and it costs them
$2,000 to make, they're running huge losses to begin with. And they have to get a lot of people on board to make up for
that in the long term by charging for service. Spacex has been tremendously successful in their launch business,
dominating the industry in just 10 years. However, the launch industry, global launch industry is not a large enough
market to subsidize Elon Musk's dream of building a colony on Mars. The key to Musk's vision for colonizing Mars is Starship. The massive rocket is meant to be fully reusable and capable of launching
up to 110 tons of cargo at once. Though SpaceX has not revealed how much it spent on the Starship program,
in the past, Musk has estimated that it would cost the company about $5 billion to complete. That's where Starlink comes in. The decision to move into Starlink gives the company a market adjacency
that's significantly larger in size and has the potential to generate the types of revenues and profits that would
be needed for that Mars colony. Earlier this year, a SpaceX job posting revealed that the company is
planning to build a factory in Austin, Texas, to manufacture its Starlink kits. But although Starlink is initially targeting the consumer market,
experts say there's a lot of room for the service to expand. The intended markets for Starlink measure in the tens of billions of
dollars, ranging from consumers, to enterprises and mobility applications, including vessels at sea and aircraft in flight. Starlink's initial focus on consumers is a byproduct of the way that the
satellites are launched and the coverage that they provide. Currently, Starlink is not able to provide an enterprise-grade
service, so the company is leading with consumers that are a little bit more tolerant. In the past, Starlink's been used by emergency responders in Washington,
where the satellites are manufactured, to set up an internet connection in areas devastated by wildfires. The U.S. Air Force and the Army are also both testing Starlink. There is a huge hunger for investment in satellite internet. The sector could be worth $412 billion by 2040. And it's not just VCs that are investing. Satellite internet is ripe for government subsidies. In Canada, the government of Quebec has invested millions in Telsat. In China, satellite maker, Commsat, received a massive government
investment as part of the country's new infrastructure drive. The U.S. too is betting big. Last year, the FCC awarded Starlink nearly $1 billion in subsidies to
bring internet to rural areas. And in late March, President Biden said that his administration would
spend $100 billion to expand broadband access to Americans as part of its $2 trillion infrastructure plan. SpaceX is also in talks with the U.K., where Starlink could earn funding
as part of the country's $6.9 billion internet infrastructure program. But one place where Starlink may not be welcome is in Russia,
where the government is reportedly considering enacting fines for individuals who sign up for the service. Russia is working on its own satellite internet constellation, which it
hopes to begin launching in 2024. Starlink's potential is huge, provided that the project can overcome some
majorhurdles. Satellite broadband as a whole is an industry and a sector that is just
littered with warning signs and the corpses of former companies that have tried to go
out and do what Starlink is doing already. It's a little discussed secret that all of the successful satellite
operators today were subsidized either through government subsidies or through the bankruptcy court. In the case of companies like Iridium and Orbcomm that went bankrupt in
the 1990s, only to recover and come back for a second act. One of the challenges of building out a new satellite network is all of
the capital expenditure needs to go up front before the very first customer can be signed up. And that creates a really tough financial model in terms
of generating the revenue to both pay for the existing constellation as well as the follow on satellites that will be needed to
continue the service. SpaceX's leadership, about two or three years ago, estimated that they
thought it would cost upwards of $10 billion to get Starlink running in an operational capacity. That's a probably a pretty fair estimate still today. Starlink is still fairly early on in development, so it's not perfect. SpaceX makes this clear on its website, saying that there may be periods
when customers experience no connectivity, but service will improve as the company launches more satellites. However, having thousands of new satellites orbiting the Earth comes with
its own set of problems. Initial launch of SpaceX's Starlink satellites in May of 2019 took
astronomers by surprise. They didn't realize how bright the satellites would be, especially as
they're raising up to final orbit and how many there would be. So if you looked up, you would see this long line of like what looked like
slow moving shooting stars all in a perfect line going across the sky. So people called them Starlink trains. Now, that brought up a pretty significant issue, which is that because
they were so bright, they are so clustered together and there was increasingly more and more and more of them covering different parts of
the sky, astronomers started seeing them pop up and really ruining different imagery and causing all sorts of distortion and effects and
things like that. This image taken from a telescope in Chile in November 2019 illustrates the
problem. The telescope, meant to see images of distant stars and galaxies, instead
captured the light trails of 19 Starlink's satellites. While some image processing tools can be used to remove the trails, Walker
says it's not 100 percent effective. But SpaceX has not ignored the problem. The American Astronomical Society, the AAS, and NOIRLab and others
contacted SpaceX, and ever since then they've been so generous with their time. And what was just a passing interest in our concerns became about half a
dozen people that they have on staff that are dedicated to finding out mitigation solutions. SpaceX originally tried to paint them like this dark material. The problem was they were still too bright, generally, and they also got
pretty hot. They instead of came up with what are known as sun visors, and it helps
keep the reflectivity of the solar panels from creating a lot of light and a lot of brightness. The other thing they did is they changed the orientation of the satellites
themselves so that they were more on like this knife's edge. So instead of the whole panel catching the sunlight and reflecting down to
earth, it was just only a piece of it. The huge concentration of satellites also worries radio astronomers who say
that the interference from radio frequencies of internet satellites could hinder the ability of their instruments to look for organic and water
molecules in space. Then there's the problem of congestion. Since Russia first launched Sputnik in 1957, over 10,600 objects have been
sent into outer space. If SpaceX were to launch all the satellites that it's requested, the
company would by itself be responsible for almost a fourfold increase in the number of spacecraft launched by all of humanity. And with the lifetime of Starlink's satellites only being around five
years, experts are also worried about space debris. An idea known as the 'Kessler Syndrome' summarizes just how detrimental
having free floating junk in space can be. The Kessler Syndrome can be described as two satellites that collide, and
they create more debris that will collide with other satellites. And this creates orbits that are just unusable. For its part, SpaceX has said that it plans to deorbit satellites that are
nearing the end of their life by pushing them back into Earth's atmosphere where they burn up during reentry. All of SpaceX's satellites have propulsion systems on board, which
theoretically are more than sufficient to deorbit a satellite in a very reasonable period of time. However, if the satellite communication is lost, which has happened in
certain cases, you lose the ability to tell the satellite to deorbit itself. In other cases, the propulsion system itself may have failed. And when you're launching over 4,000 satellites, all it takes is a very
small failure rate to create a lot of space debris. Jonathan McDowell, an astronomer at the Harvard Smithsonian Center for
Astrophysics, has been tracking Starlink's failure rates and says they've improved. For now, dealing with space debris is still largely left up to
individual companies. There is no enforceable law that can cause a company to deorbit a satellite
that has failed. Generally speaking, the FCC, along with other governmental agencies, have
rules out, guidelines, to deorbit satellites within 25 years. But there's no enforcement mechanism that could cause a company to spend
money to actively deorbit a satellite. At least today. But experts like Walker are pushing for more national and international
oversight of satellite makers. The United Nations Office for Outer Space Affairs is already looking at
ways to mitigate the effects of light and radio pollution coming from satellites. In the U.S., the Biden administration recently announced that
it would continue the National Space Council, which will assist the president in setting national space policies. Plans to spin off Starlink into a separate company have been swirling since
last year. And earlier this year, Musk confirmed the idea. Elon Musk again making headlines this morning. This time with SpaceX, saying its broadband satellite business will go
public, eventually. Musk tweeting, quote, "once we can predict cash flow reasonably well,
Starlink will IPO," but warning that the business will need to pass through a, quote, "deep chasm of negative cash flow" over the next year or so to
actually make that financially viable. Spinning off the part of SpaceX that has the most earning potential may
seem counterintuitive, but experts say there could be some benefits. Right now, Starlink is a huge cash use for the SpaceX business. And if it were spun out independently and charging back, paying SpaceX for
those launches, it would obviously boost the revenues of the remaining SpaceX business. If SpaceX is still the majority owner of Starlink after it spins off, then
the company can still see the revenue stream, still see the benefits of that service. But it's then de-risking the overall SpaceX company, if you will, by
having it operate as a separate business so that if Starlink does fail, or does go bankrupt or something bad happens to
Starlink, then it's not damaging the core space business as much as it would as if it was still in the entire private fold
of the company. SpaceX is currently the leader in low earth orbit satellite internet, but
competition is heating up. Amazon has said that it will invest over $10 billion in its satellite
internet network known as Project Kuiper. UK satellite maker, OneWeb, recently launched another 36 satellites into
orbit, bringing its total number to 146. The company says it expects to begin limited service by the end of the
year, though unlike SpaceX, its service is geared towards enterprise customers. Finally, Canadian satellite company, Telesat, has said that it
will begin commercial services for its satellite internet in the second half of 2023. If Starlink is successful, it may ultimately do more than just fund Musk's
vision for a colony on Mars. SpaceX put in the Starlink terms and agreements that there's bylaws in
regards to how their service will be treated for people on Mars. And so SpaceX is already looking down that path and seeing, OK, well, we
can have Starlink satellites around Earth in orbit, but then we can also put them in orbit around Mars and then just connect the
two and have this expansive, not just a global satellite system, but a multi-planetary satellite system. But for now, earthlings are glad that Starlink exists, even if it's not yet
perfect. When they first reached out to us, they did label their beta test as the
'better than nothing beta test.' And for people like me, it's one hundred percent better than nothing.
Associated article : What early users of SpaceX’s Starlink satellite internet think about the service, speed and more
Reporter's tweet about the article
Elon replying: This is accurate. Service uptime, bandwidth & latency are improving rapidly. Probably out of beta this summer.
Tim's tweet: Will users always be locked into one location or in the future if a user has the standard Dishy McFlatface (not a new portable one), could you say put it on an RV or tiny home? Or maybe take one you have in Iowa and put it in a studio in Texas
Elon replying: Yeah, should be fully mobile later this year, so you can move it anywhere or use it on an RV or truck in motion. We need a few more satellite launches to achieve compete coverage & some key software upgrades.
We’ve gone from consistent 50/10 to 70-90/10-15. Still drops out 2-3 times per hour for about 20 seconds each time, but the progress is unmistakable. Great for TV streaming and 99% of all internet work. Can’t VPN with it because of no fixed IPs available/can’t use dynamic DNS, or Zoom calls because of the dropouts. Really close now though to cable/wire line internet.
Saving for later. Need more battery. !remind me tomorrow
This can we solve IPV6 model very soon. Already rolled out in Germany🇩🇪.
I ordered mine about 1 1/2 months ago. When do you think I’ll get it
Can't wait till skylink has an IPO.