back to article Amazon hasn't launched one internet satellite yet, but it's now planning a fleet of 7,774

Amazon wants to launch another 4,538 satellites to provide wireless broadband internet under Project Kuiper, according to a fresh filing to America's communications watchdog. The mega-corp was previously approved to send 3,236 birds into low Earth orbit by 2029. Now, it wants to expand that number to 7,774. “Kuiper Systems …

  1. Bartholomew Bronze badge

    The future's so bright i gotta wear shades

    It is nearing the the stage where we will only be able to see the brightest stars - If they keep on launching massive low earth orbit constellations.

    1. Stork Silver badge

      Re: The future's so bright i gotta wear shades

      I find it a bit sad, as someone who appreciates seeing the Milky Way

      1. Snar

        Re: The future's so bright i gotta wear shades

        Yup - We lost the HF spectrum to non EMC compliant switched mode power supplies, xDSL and Fuckwit Powerline junk. I fought against it and Ofcom took it up the arse like good little puppies.

        Nothing is protected from not putting barriers up to innovation / business. Just junk whatever so someone else can make a fucking fortune.

      2. Geoff Campbell

        Re: The future's so bright i gotta wear shades

        You will still be able to see the Milky Way perfectly well, everywhere you can now see it. LEO satellites are visible to the naked eye under certain circumstances, but they don't really intrude on the night sky to visual observers.

        It's long-exposure telescope photos that they screw up.


        1. imanidiot Silver badge

          Re: The future's so bright i gotta wear shades

          With that many satellites in orbit even their low apparent magnitude is going to wash out the sky and cause a "glow" that makes the milky way and stars much harder to see. It's definitely NOT just long exposure telescope photo's that are affected.

          1. Geoff Campbell

            Re: The future's so bright i gotta wear shades

            You won't see all the satellites, only the ones low enough on the horizon to catch the sunlight. You're panicking over nothing.


            1. imanidiot Silver badge

              Re: The future's so bright i gotta wear shades

              That's what I'm saying. You DON'T see the individual satellites. But together they reflect enough light to still be a nuisance.

              1. Geoff Campbell

                Re: The future's so bright i gotta wear shades

                Reflected from where?


                1. badflorist

                  Re: The future's so bright i gotta wear shades

                  "Reflected from where?"

                  As usual, anywhere light is reflected from.

                  Not sure on the number needed, but with a dense enough mesh they'd simply bounce any light from Earth, Moon, Sun, stars, themselves etc. to cause noticeable haze. For example, without changing the intensity of a given light source, a simple screen door can be used for diffusion with varying mesh sizes. Another effect can be seen in schools of fish and even fireflies which don't light simultaneously but still cause some value of haze. However, I'm not sure on how many satellites that would take but, apparently one day we're going to find out :-/

                  Also, see this link posted 1 day after this article:

                  1. Geoff Campbell

                    Re: The future's so bright i gotta wear shades

                    I'm not buying it, I have to say. Even millions of satellites on the surface of a sphere over 13,000km in diameter is not going to get even close to the sorts of densities you are talking about, and at the moment we are talking about tens of thousands, not millions. But, OK, we'll see.

                    The bottom line is simply that LEO is such a useful place, that it is going to become rather busy, one way or another, so we'd better start planning for that. Me, I'd rather like to see the Milky Way from an orbital platform, and for that to happen, Low- and Mid-Earth Orbit is going to have to become a routine place to visit, just like an extension of our planet.


                    1. badflorist

                      Re: The future's so bright i gotta wear shades

                      Low Earth orbit is a useful place. While this is tinfoil hat'ish, from a "defense" point of view, being able to fling one of these things down on top of some place is a very nice non-nuclear option :-/. Yeh that seems far stretched but, Ronald Regan's Star Wars project had provisions for exactly that (I think that was the NPB's default upon kill switch). Like I mentioned, it's tinfoil hat'ish, but it's not exactly beyond the stretch of reality.

                      Theoretically, you could put a food dispensing network in Low Earth orbit to feed just about anyone... you can put all sorts of shit up there.

  2. Neil Barnes Silver badge

    Just for curious: I assume there must be international treaties that permit overflying of a country's territory by satellites: is this the case? And if so, is it still relevant to bulk swarms like this?

    1. Chris G Silver badge

      The only thing I know of is INMARSAT who are involved with marine sat's.

      Judging by the increasing numbers of constellations and the size of the constellations, it would seem to be time for a genuinely international body that can oversee the use of near space to the overall benefit of humanity and the planet.

      Of course it would need some legal teeth that can't just be ignored the way the UN and the International Court are when it doesn't provide the results the some require.

    2. Sean o' bhaile na gleann

      " treaties that permit overflying of a country's territory by satellites..."

      I'm fairly certain that isn't the case (and, if memory serves, this idea cropped up in a previous thread on these august pages).

      As I remember things, the Russians effectively shat on that idea right at the very start by sending Sputnik 1 up without initially clearing such 'overflying' in the first place.

    3. Gary Stewart

      Orbit mechanics doesn't allow you to pick and choose which countries a satellite passes over.

    4. Mike Richards

      In 1955, President Eisenhower proposed an "Open Skies" policy where the US and USSR would allow spy flights over one another's countries. The Soviet Union rejected it outright because it was all too aware that its announcements of endless bombers aimed at America wasn't actually the case.

      Eisenhower then announced that the US would build and launch a satellite for the International Geophysical Year. He had received a proposal from the Science Advisory Committee which said that non-military satellites would establish 'freedom of space' where satellites could pass over other countries' territory without consequences.

      The Soviet Union, eager for a publicity coup announced its own satellite (which would eventually become Sputnik 3), but this ran behind schedule; so in order to score a first they lobbed Sputnik 1 into orbit and effectively established 'freedom of space'. They couldn't then complain when the US started launching its own satellites.

      1. Anonymous Coward
        Anonymous Coward

        It's interesting, but Open Skies is about planes, not satellites, with a lot of conditions on those planes that would and could not apply to satellites.

        And also, it's now the US who've been the first to formally withdraw from the treaty, which is unfortunate.

    5. Steve Todd

      International law holds that anything in orbit is extraterritorial. It does however state that radio transmissions out as far as the moon are regulated by the ITU, and the member nations under it. So SpaceX can fly satellites anywhere it wants, but to broadcast to a station on the ground it needs permission from the host country to use the specific radio band.

  3. Anonymous Coward
    Anonymous Coward

    Nightmare scenario...You just know it will be a constellation of satellites...

    You just know it will be a constellation of satellites, that writes the words "Amazon" in the night sky, with their logo beneath it, formed from 'strategic' reflectors on board the satellites as they pass over, and worse case, in ticker tape form.

    The sky's the limit, as they say.

    1. Mike 137 Silver badge

      Re: Nightmare scenario...You just know it will be a constellation of satellites...

      "that writes the words "Amazon" in the night sky, with their logo beneath it"

      See 'Buy Jupiter' [Isaac Azimov 1958]

      1. Sean o' bhaile na gleann

        Re: Nightmare scenario...You just know it will be a constellation of satellites...

        See also 'Watch This Space' [Arthur C Clarke 1956]

        1. adam 40 Silver badge

          Re: Nightmare scenario...You just know it will be a constellation of satellites...

          I wonder how many laser pointers would we have to point at the new moon to show up a visible dot of light?

          1. imanidiot Silver badge

            Re: Nightmare scenario...You just know it will be a constellation of satellites...

            All you could ask for and more

            Because ofcourse Randall Munroe looked into that already ;P

    2. NeilPost Silver badge

      Re: Nightmare scenario...You just know it will be a constellation of satellites...

      Wasn’t there some advert wheeze made by simultaneously (as visible on earth) pushing a whole bunch of stars Supernova in Hitch-hikers guide or Red Dwarf?? My memory remembers something but not where.

      “Prime Day Deals” !

      “Clarkson’s Farm Season 2” now available !

      1. ArrZarr

        Re: Nightmare scenario...You just know it will be a constellation of satellites...

        That scale would be much more on the Hitch-Hiker's scale than the Red Dwarf scale.

        Probably wouldn't be Earth focused though. Maybe Ursa Minor Beta.

  4. TeeCee Gold badge

    II have heard...

    ...that if you pay a sub, once every year one will de-orbit, fly over your fence and knock over the dustbin in your back garden.

  5. Martin an gof Silver badge


    They are talking about "tens of millions of subscribers" and they are talking about 7,774 satellites. That doesn't seem a very good ratio to me - even if you assume 99 million subscribers, that works out to fewer than 13,000 subscribers per satellite. How much do these things cost to launch, how long do they last, and what are the running costs on the ground?

    Other constellations with more satellites will be in a worse position unless they are aiming for "hundreds of millions" of subscribers...

    Oh, and it isn't clear from the article if the 7,774 is a total over both generation 1 and generation 2 constellations, or the number they require for gen.2, that is in addition to the 4,000 (or whatever it was) that have already been approved for gen.1.


    1. John Brown (no body) Silver badge

      Re: Economix

      It's a marketing figure, as used by all ISPs and mobile phone companies. ie it's the potential number of people who *could* access it because they are within range of the service. Whether they can afford it or want it or have gone with a different service, is another another matter. Or, far that matter, whether it would even be worth using with that many customers all signed up.

      1. Phil O'Sophical Silver badge

        Re: Economix

        "1bn homes passed"?

  6. itsborken

    Perhaps it is more prudent to get the two prototypes into orbit and observe their capabilities before talking about scale.

    1. John Brown (no body) Silver badge

      Marketing will always try to sell stuff that isn't ready yet. Worse, they will imply or even state abilities and functions the engineers haven't yet, or will ever, implement.

      1. The commentard formerly known as Mister_C

        so my car might not ever be able to drive itself?

        1. Fruit and Nutcase Silver badge

          Well, they've reached the point in the development whereby it is able to crash by itself. Just need a few more agile sprints to develop the full drive by itself (without crashing into things) mode.

          1. Anonymous Coward
            Anonymous Coward

            Self Crashing Cars

            My 15 year old car is (I believe) quite capable of crashing itself. It's my task to keep it between the white lines and prevent it crashing.........

  7. elsergiovolador Silver badge


    If other businesses could afford tax avoidance structures like Amazon, maybe we could see more satellites in the making.

    1. IGotOut Silver badge

      Re: Advantage

      This is a tax avoidance scheme.

      Amazon Inc.

      Registered office. Space.

  8. spold Silver badge


    I'm going to invest in an AI-equipped tennis racquet - because anything launched spacewards will need one to swat the Internet-of-Tin cans out of the way.

  9. EveryTime

    I can't imagine what Bezos will announce after Musk announces his composting toilet business.

    1. elsergiovolador Silver badge
      1. Geoff Campbell

        "Dined with him" in the loosest possible sense of the phrase

        They were all at an event organised by a third party. Hold the presses!


  10. Version 1.0 Silver badge

    Climate change?

    What will 7,774 launches cost us all? And you can bet that every 5-15 years the satellites will need to be replaced ... what's the climate cost of building them and launching them? If Amazon starts doing this then it's a safe bet that Facebook and Google will launch at least 7,000 or more too.

    1. elsergiovolador Silver badge

      Re: Climate change?

      Don't worry, you'll be forced to eat less meat, take a bike to work and maybe only use internet to buy stuff on Amazon.

      We all be forced to make sacrifice so the billionaires can have their fun.

    2. Geoff Campbell

      Re: Climate change?

      Both Blue Origin and SpaceX are working towards renewable fuel. Blue Origin uses hydrogen, SpaceX are using methane (that will be) created from atmospheric CO2 and water in the Starship engines, which I'm guessing will find their way into the Falcon boosters before long.

      Both have reusable boosters, so that's not a huge concern either.

      But even if that weren't the case, rocket launches have a surprising low GHG footprint. From memory, a Falcon 9 launch is the same footprint as a single one-way flight from London to New York.


      1. imanidiot Silver badge

        Re: Climate change?

        "which I'm guessing will find their way into the Falcon boosters before long."

        EXTREMELY unlikely. That would involve a ground up redesign for Falcon 9, which doesn't make economic sense. Nor can the F9 really use the (much more powerful) Raptor engines as they can't be throttled down as much in general, and using fewer more powerful engines means you can't just shut a few down and use only a single engine for landing. Using Raptor on F9 would mean giving up re-usability.

        Theoretically it MIGHT be possible to redesign the Merlin to run on liquid methane, but that would still be a massive effort that they don't really need. I'd find it more likely they'd invest in making synthetic liquid fuel from methane that the Merlin 1D can run on without modification (which is proven to be possible, just currently not economically viable).

        1. Geoff Campbell

          Re: Climate change?

          Fair enough.


        2. ArrZarr

          Re: Climate change?

          Why would using Raptors on F9s sacrifice re-usability?

          Also, is it not the case that the "expensive" bit of any liquid fuelled rocket is the engine and the "cheap" bit is the body? For smaller payloads, the BFR, Starship, X Æ A-12, whatever the hell he's calling it this week is exceedingly large so building an F9 equivalent with Raptors would mean you get to shut down Merlin production while maintaining a medium-lift capacity entry in the market.

          1. imanidiot Silver badge

            Re: Climate change?

            Because Raptor can't throttle down far enough. A single raptor at minimum throttle (40%) still puts out nearly the same power (750kN or so) as a Merlin at full thrust (850kN roughly)

            The landing burn is done on a single Merlin at nearly the bottom of the throttle range (500kN or so) and even then it's cutting it close in terms of lighting the engine and performing the divert maneuver to actually reach the barge or pad (since the rocket aims just to the side during the free-flight phase of the return so as not to hit it at "oops the rocket didn't work" velocities). That "hover slam" maneuver is designed to reach as close to zero fuel, at zero feet altitude and zero velocity as possible. For a rocket like F9 in that weigh/performance class the weight ratios can't really change, so with a raptor engine with much more thrust the time between lighting the engine and the rocket coming to a dead stop (or rising up again) are too close together to light the engine, divert course to end up on the barge/pad and come to a stop at 0 altitude without some seriously risky aggressive maneuvering angles. Plus lighting an engine at that higher thrust level also puts more forces and stresses on the air frame, which the structure might not like very much.

            The whole point of F9 is that the engines are no longer the super expensive bit either, since they don't get thrown away after every launch. Sure you could maybe stop the assembly line of Merlin engines, but that line is basically on "tick-over" as it doesn't need to supply new engines for every launch. The expensive bit for engine production has been done already anyway. The production line exists, the engineering is done. Other than a continued cost of personnel and upkeep keeping it going is not super expensive anymore compared to the umpteen million spent on building the line in the first place.

            Redesigning F9 to take Raptors is probably not going to save more in Merlin production line costs than it costs in engineering time investment alone to redesign F9 from the ground up to take Raptor engines. That is the paradox of engineering. Sometimes it's cheaper NOT to do something, even though every fiber in your being as an engineer SCREAMS that something can be done cheaper, better, faster and more reliable. If you're only going to be building maximum 30 of something in the next 10 years, and you might save 50.000 per unit (so roughly 1.5 million in total) but it costs you 2 million in engineering and tooling costs, does it make sense? Only if you expect to keep building the thing for the next 30 or more years.

            1. ArrZarr

              Re: Climate change?

              All makes sense, thanks for the explanation :)

  11. Grunchy Bronze badge

    I guess the only way satellite internet latency can be superior to land-based fibre optic is by completely eliminating the fibre optic: both parties communicating have to subscribe to the same satellite internet service.

    1. Geoff Campbell

      Not quite

      That is, of course, the ideal solution, and I have no doubt that SpaceX and others will be looking to sell such links to corporates with deep pockets.

      However, going up to a satellite from the UK, around the planet by frickin' lasers, then down to a ground station in New York for onward transmission to a local business will also shave some milliseconds off the latency.


  12. Valeyard

    and in a few years

    and in a few years we can't see the stars because the only light show available is the equivalent of tens of thousands of old freeserve routers flying through space

  13. imanidiot Silver badge

    RIP Astronomy

    Can't let science get in the way of making dollars now can we?

  14. Potemkine! Silver badge

    Will these satellites de-orbit when they will be out of order, or it is already planned that there will be 7,000+ pieces of junk more into orbit?

    1. Geoff Campbell

      Yes, they will de-orbit when they are due for replacement.


      1. imanidiot Silver badge

        The ones in a low enough orbit (below 500 or so Km) will de-orbit by themselves soon enough (within a few years) by atmospheric drag. The ones above that will take much longer to deorbit if this is not done actively. While all the sats will likely have the equipment to do so, and the plan is to do so, with 7000+ sats up there even a 1% failure rate due to unforeseen hardware problems (or micro-meteoroid/space debris impacts) means 70+ sats left up there out of control. Just from this constellation alone.

  15. 89724102172714182892114I7551670349743096734346773478647892349863592355648544996312855148587659264921

    All of these companies should be forced to launch a vast array of gigantic space telescopes, before launching any constellation of satellites even remotely likely to obscure our view of oncoming extinction-level-event asteroids.

  16. 89724102172714182892114I7551670349743096734346773478647892349863592355648544996312855148587659264921

    Maybe if they launch enough, it'll look like a huge mirror - appropriate for billionaire narcissists.

  17. 89724102172714182892114I7551670349743096734346773478647892349863592355648544996312855148587659264921

    Why has humanity decied that there's "Nothing to see here"? It's bonkers!

  18. HereAndGone

    Who will provide Launch Services?

    There are a very limited number of Launch Services, who will launch the Amazon satellites?


    Hmmm, NASA doesn't actually have much in the way of launch capability, they sub all their heavy lifting to Space-X.

    Blue Origin?

    So far they can't orbit anything.


    I can just imagine that negotiation.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like

Biting the hand that feeds IT © 1998–2022