Pragmatic 99: StarLink

14 August, 2020


SpaceX and Elon Musk have been accelerating a project to provide high-speed, low latency broadband internet around the world via Satellites and are calling it StarLink. Radek joins John to deep dive into what StarLink is trying to accomplish, why it’s a such a different approach to the problem and why it might actually succeed where others have failed.

Transcript available
Welcome to Pragmatic. Pragmatic is a discussion show contemplating the practical application of technology. By exploring the real-world trade-offs, we look at how great ideas are transformed into products and services that can change our lives. Nothing is as simple as it seems. Pragmatic is supported by you, our listeners. If you'd like to support the show, you can do so via Patreon for early release, high quality ad-free episodes. Visit to learn how you can help. Thank you. I'm your host, John Chydzi, and today I'm joined by Radek Pietruszewski. Hey, I'm so glad to be here again. And after two years, which I can't believe it's actually been two years. Yeah, it has been a couple of years and it's great to have you back. Thanks for coming back on the show, I kind of, I really wanted to bend your ear about something that I've been reading about recently and I've been getting more excited about. And it's some of one of SpaceX's, I guess, I don't know if you could call it a side project or not. Because it seems like Elon Musk has a lot of side projects, but, but yeah, Starlink, would you call it a side project? I would say it started its life as a side project and has been a side project for a couple of years, but I think we're going into the phase that they're really going for it and it's going to be a very expensive project. So definitely not a side project. No, not anymore. It's kind of interesting when I first read about it and I thought, well, this is going to work and that's not going to work and I can't believe they're serious and how many satellites and right. So I've done the whole full circle thing here. And so I thought it might be interesting to sort of like dig into this. And so last time we talked about SpaceX, I mean, a bit of an update since then, actually, it's probably worth just touching on that since the last time we spoke in the last two years. I mean, since then, I'm just trying to remember if the Falcon Heavy had actually made it up and all of the Falcon 9s had landed again afterwards, what else have they? I think they're down to a turnaround time measured in a matter of a few weeks now, 50 days? I'm trying to remember. Yeah, I think two months was just, just beat the record recently. Yeah, that's crazy. I mean, all this stuff is just really driving down the cost of access to space. Oh, the other thing is the, the Dragon capsule, since we last spoke, is now the preferred method for getting astronauts up into the space station. station. Yeah, the Crew Dragon, that's many years in the making, many, you know, a few years over time, but you know, that's a finally and seems like everything worked. It was almost boring. Yeah, well, but that's what you want, eh? Yeah. Nice and nice and boring and serene and no issues. But I mean, I just keep on thinking about how far they've come. It's incredible. And so, you know, having achieved all of this, it's like, well, okay, what else can we unlock? I mean, obviously their main goal, well, SpaceX keep on saying about their main goal is to go to Mars. And that's all well and good and lovely. But then Starlink comes along and they're getting very serious about it. So, all right, I think it's really important if we're gonna talk about what Starlink is to understand what the state of play is now in terms of satellite internet. And in order to understand that, just got to cover off some of the basics. And I know just for the benefit of the listeners, I mean, things like understanding orbits, like ground coverage and why we are where we are and why SpaceX's idea is so different from everything else that's out there pretty much. So starting off about orbits, I guess, and I guess the basic of orbits with satellites is the speed of the satellite in orbit that has to travel needs to be enough to escape the rate at which it's falling, Otherwise it's going to fall back to earth, obviously. Um, but if you're going to go too fast, eventually you're going to fly up into space eventually. So the higher you go, and that's one of the funny little conundrums, the higher you go, the slower you go, depending upon how you want to think about it until you reach that magic geo stationary orbit. Um, and that's the, that's at 35,780. I think it is kilometers or thereabouts. And when you're traveling up there, you're only, in air quotes, only going at 11,000 kilometers an hour, which is still quick, but if you're closer to earth at a lower earth orbit, then you're going faster than that. The interesting thing about GPS satellites, cause everyone knows about, well, I say everyone knows about, that's presumptuous. I think the vast majority of people have heard of GPS. So global positioning system satellites, they hang out around about 20,000 kilometers up and which is not geostationary and they're traveling about just under 14,000 kilometers an hour. So they're going faster because they're closer. And as you get closer, you go faster. So in terms of speed over ground, I suppose, technically. Yeah. Okay. So far so good. Yeah. I think the good way to think about it is in terms of the difference between speed and angular speed. Right. So to get to higher orbit, you must be moving faster. But from the perspective of the Earth, it seems like you're moving slower, because you're so high up that the angular speed from your vantage point is slower and slower until you're at geostationary, and it feels like the satellite is not moving. It's at one point in the sky, and it's actually really fast. Yeah, okay. No, that's a good way of thinking about it. And it leads to one of those funny little conundrums too, that took me a while to get my head around years ago, when I first heard it. If you're trying to actually increase the orbital speed, or if you have thrusters on a satellite and you fire them in the opposite of the direction that you're moving, you'll decrease the altitude. Whereas if you want to increase its altitude, it's the other way around. And you're right, it's probably easy to think about it from Angular. It's an interesting little one to get your head around, but it kind of makes sense once you do. It's just, yeah, it's just a bit odd initially. I highly recommend Kerbal Space Program, which is the best way I know to get an intuitive feel of how orbits work. Because orbits are really unintuitive until you play with it and, you know, thrust at different points of the orbit and then they make total sense. Cool. All right. Excellent. Have you got a link for that one? And I can throw that in the show notes. Ah, yeah, I'll find it for you. That's okay, thank you. All right, a couple of other things just real quick, just to set up as things like inclinations. So the inclinations is the angle of the orbital plane that is traveling. So zero degrees would be perfectly around the equator. Thank you very much. I'll add that to the list show notes. So as you move away from the equator, then the angle will increase. So if you're going directly over the poles, then that would be 90 degrees. So that's the inclination. So, the inclination that the satellite is traveling in its orbit, that's all inclination is. Nothing too earth-shattering. Anyway, I wasn't even trying to be funny, and I'm not even sure that was funny, but never mind. Okay, satellites. One of the problems with satellites is that, how should I put this? It's not so much the satellite's fault, it's the atmosphere's fault. And I'm grateful that we have an atmosphere, don't get me wrong, that's why it keeps us alive. But the problem is that even though you're in low Earth orbit, let's say you're in low Earth orbit, you're going to get some drag due to the atmosphere. It's not a perfect vacuum. It's close, but it's not close enough. And there's not just that, there's also other gravitational objects. I mean, the most common ones being the moon and the sun. I imagine, you know, different bodies in the, you know, around the place will eventually tug that poor satellite. and eventually you're gonna have to adjust its position with thrusters. That's why satellites have got thrusters on them, just to give it a little bit of a boost from time to time. Otherwise you get this thing they call orbital decay. And that's generally not good, generally. I mean, sometimes you wanna get rid of a satellite, I guess you wanna de-orbit burn it and bye-bye. But generally, if you're gonna put a multimillion dollar satellite or multibillion dollar satellite up, it's probably a good idea to keep it up there, generally. I say generally, (laughs) Starlink is a different, interesting take on it. Anyway, all right, so there's that. Interesting fun fact, well, it's a fact, whether it's a fun fact or not, solar flares. So our sun goes through an approximate 11-year solar cycle and you've got lots of sunspots and solar flares and then they sort of peak and then they dip down again. So we've been observing this for hundreds of years. It's not a new thing, but the thing that's interesting is that when the flares are most active, it actually heats up and changes the density the upper atmosphere to the point at which it creates more drag. Funnily enough, during those periods of time, if you have to boost a satellite, you may have to boost it up to four times in a year just to keep it up during high periods of solar activity, whereas you may only have to do it twice, let's say, at solar minima, which is interesting. The other thing is also, I guess, just an aside that when I was in ... I'm still an amateur radio operator but I haven't actually played a radio on an amateur radio frequency for quite a while but obviously you know you get some good skip propagation based on your solar cycle so that could be good for ham radio operators but not so good for satellites but anyway. Right all of this is relevant so stick with me. Alright so ground coverage here's here's an interesting thing and it's probably obvious maybe it's obvious as your satellite is in orbit at a certain height there's only so much of the earth that it can see at any one time. And at some point, if you try and widen your field of view, you'll reach a point where you can't possibly see any further. And essentially, that ground track that it follows as the satellite is in orbit, the closer it is, the less it can see. That's the idea anyway. And I think one of the terms they use for it is swath width, or some people pronounce it swath, but like lathe. That's ultimately one of the issues is the closer you get your satellite close to the surface, then you're gonna have less latency. So as I send a signal up to it and I get a signal back, again, it's gonna have less distance to travel, so it'll take less time. But the flip side of that is you're gonna get less coverage because you can't see as much from that height. You could compensate for that by having a much bigger antenna and that bigger antenna can cover a much bigger area, but the problem with like as much as you can physically see, but then that drives up, but that drives up the cost of the antenna and the amplifier. You need to actually get that signal spread over a large area. So you start eating up your power budget. So it's a bit of a problem anyway. So what I thought might be worth doing is, I guess let's talk a little bit about geostationary satellites for a second, because that's kind of like what we've got now and a little bit about some of the other closer to low earth orbit ones like GlobalStar, Iridium and so on. So geostationary satellites that far out, the latency of those is measured in half to one second. It's pretty tragic, actually, it's terrible. 'Cause they're so far out. It's a long way to go and a long way to come back. - If only speed of light was faster. - Yeah, I know, hurry up light. - We're dealing with such distances that the speed of light is just not fast enough. - Light speed's too slow? That's right. Oh dear, it's terrible. And it's just, it's the fact of physics. There's nothing you can do about it. But the other thing that people don't often think about with geostationary satellites is that you can always tell when you're going up, if you're out in the outback or the middle of nowhere, you'll know the difference between something which is a low earth orbit satellite and a geostationary satellite, because the size of the dish, the satellite dish for the geostationary satellite is huge. It's usually one to one and a half, maybe even two meters in diameter. and the low Earth orbit satellite will be much, much smaller like the size of a dinner plate, perhaps. And, you know, it's just a function of distance. If you wanna get a radio signal that far, that much further, the signal spreads over distance. And 'cause the actual area that the radio signal covers, and I've covered this previous in episodes of "Pragmatic" whereby it's the inverse square law. So for every, if you were one meter out, it's one square meter. If you're two meters out, it's two squared. So it's a quarter of the signal strength over the same area. So you have to have a narrow and narrow beam width in order to actually get a reliable communication with that geostationary satellite. And that sucks because that means that you can't actually have like a mobile or handheld. It means you have to have a large physical area and it means that, you know, well, that's not really portable. And that's, you know, back in the bad old days when satellite phones were first thing, you know, those big like briefcase phones, you'd open up the briefcase and you'd have to line it with the satellite first and then you can make a phone call. It's like, oh my goodness, anyway. So that's kind of what we got. And in terms of data over those, not too many of the geostationary satellites support any decent amount of data rates. You can get, I guess, some of them, but they're very, very expensive and the latency is horrific. So if you're trying to do real time anything, you can just forget it. It's just not, you know, And you know, they get the reporters on and I've got a reporter in a different country and they're coming in live via satellite. And the host of the show on one side says, "What's it like over there in other half of the world?" And then you sit and watch this person with a smile and a nod for about two or three seconds, like they can actually hear them and they haven't heard them yet 'cause the signal hasn't got to them yet. And then you go to wait another few seconds for them to come back to you and say, "Yeah, it's really great over here." Anyway, latency. All right, so that's geostationary. One of the things about geostationary satellites is that because they are so far out and they need more power, they have to be bigger, which makes them bigger and heavier. Getting a satellite to a geostationary orbit is also like a lot more expensive. So I think, yeah, sorry, with the Falcon 9s and the Falcon Heavies, I'm just trying to remember which one of those can get you to geostationary or not, if a Falcon 9 on its own. - Sure it can. Most satellites, most like, you know, TV satellites, a Falcon 9 can get, but there are some that are really heavy and you need something like Falcon Heavy to get them. But actually often the constraint is volume. So a selling feature of Ariane 5 rocket is that it has such a big payload fairing that you can get two geostationary satellites in one launch. - They have to be pretty lightweight for geostationary satellites, but you could just couldn't fit them on a Falcon. - Yeah, it's interesting when you look at some of the complexities of that, and we'll get to the whole Starlink and how that's different in a minute. But yeah, I mean, one of the advantages of the miniaturization and the chip fab sizes and the amount of power that we've got available to us, like processing power in terms of compute power, we are able to get away with smaller and smaller satellites, but when it comes to raw power and large, large battery packs and large solar panels, sometimes you just can't avoid it. And if you need that power in a geostationary satellite, which is the one that's gonna need it the most, you don't have a choice. You just need to make a big heavy satellite and it's very expensive to get up there. Yeah. Generally, if you're going to put up a, just a, regular TV satellite as cheaply as possible. Like you're not going to get it below 200, 300 million. And it only goes up from there if you want anything custom at all. Exactly. So let's talk about probably the two largest ones that I'm aware of. Obviously, there's others, but I'll briefly touch on OneWeb. It kind of started, they launched a few satellites, then they went bankrupt a few months back. So that was that. I don't know what they're going to do with the satellites that got up there. They're certainly not doing a heck of a lot. They've actually been, I don't want to say acquired because they're going through the bankruptcy court or whatever the details are. But the British government and a big telecommunications company from India wants to buy up the assets. and who knows what they want to do with it. Yeah, okay. It makes me think a little bit about what happened to Iridium. But, yeah. So, yeah. So, we'll wait and see what happens with OneWeb. But I don't want to spend too much time talking about it. But let's just say that at the moment it's not doing a heck of a lot. We'll see what happens. It was an interesting idea. Whether or not it would have been successful or not with its current trajectory, we will see. But let's see what happens. But so I just want to talk about GlobalSTAR first and then Iridium and sort of compare and contrast the differences and then what makes them different from Starlink. So let's talk about GlobalSTAR first. So GlobalSTAR has a reliance on lots of ground stations. And the idea is that when you put the satellite up into the into the into orbit, if you've got a mobile device or a spot device to connect to that satellite. You connect to the satellite, the satellite then has to connect down to a ground station, otherwise you have no connection. So the satellites don't actually talk to each other in that network. So to make it work, you need ground stations and you need to have a ground station effectively on the continents that you're operating from, otherwise it's a non-starter. And there's 40 ground stations for Globalstar around the world. Seven, there's seven in North America. And in terms of where they're at, the first generation satellites were launched and they have a, I think it was 1400 kilometers and they had a 60 millisecond latency as a 52 degree inclination. And originally they launched 77, I think it was, originally, I think, and they're down about 66 all up. anyway, second generation uses only 24 satellites. But the reason that I wanted to mention that is that they were very much more focused on the public switch telephone network when they were on data. Although you could do data on them, wasn't that impressive. And one of the big problems with GlobalStar was that, at what is with GlobalStar is that it has a high latency, but it's not just that, and that's because it's further out, but it's also because it relies on that ground station connection. there is no inter-satellite hopscotch, I guess, for the want of a better word, as the signal goes up, it hops between the satellites to get down to the nearest ground station to get back to where it needs to go. And that's where Iridium was different. And Iridium originally was launched in November of 1998. But the problem with that was that they went bankrupt, I think it was, shortly thereafter. And then they got rescued. And all the original satellites were up there for I think almost 20 years. And they've now just only just recently been replaced. So they've done a de-orbit, a burn up of all of their original satellites and they've been fully replaced by Iridium Next. And the thing that I find awesome about that is that all the Iridium Next satellites were launched by, what kind of rocket? Sorry, spoiler, it was a Falcon 9. So I just, I don't know why I find that funny is it's like SpaceX are gonna do Starlink and yet the Iridium, which is a direct competitor was launched on a SpaceX rocket. I don't know why that's funny, but anyway. So they were all launched between 2017 and 2019 on a whole bunch of Falcon 9s. There are 66 satellites in use, nine are in orbit as active backups. So that means that they launched 75 and they got six sitting down on earth as spares. They're designed for 15 year lifespan. Now they're only 780 kilometers up, which is 485 miles in a 100 minute earth orbit. So because they're closer, they have the service they call it's Iridium Certus, C-E-R-T-U-S, I think that's how you pronounce it. Anyway, 40 to 50 millisecond latency, so better. And data rates up to 700 kilobits per second now, and they reckon they're working on refining that and pushing that to 1.2 megabits per second soon. Now, as I said before, one of the other great things about Iridium is that it does do inter-satellite comms. So if you're in a location, you can go up to satellite A. If you can't get to where you need to go, then it'll go across the satellite B and maybe the satellite C before it goes back down to a ground station. And they just added a whole bunch of extra ground stations as well. They don't need anywhere near as many as Globalstar. And yeah, so I think they put the latest one in the Southern Hemisphere. Actually, I think, is it the only one in the Southern Hemisphere? Anyway, it's in Punta Arenas in Chile. I'm assuming it's nice and high up there. And the one before that they finished off was the Chandler ground station in Arizona. There's a few more than that, but they don't need anywhere near as many. But still, we're not talking about a lot of satellites. And they're still technically low earth orbit, but not quite as low low as Starlink's gonna be. Anything else to add on Iridium? - I'm just going to plug in if we're discussing Iridium, the book "Eccentric Orbits, The Iridium Story" by John Bloom is simply amazing. The story behind Iridium's success and then demise and then kind of success again is wow. Like it's a great book. I just wanted to plug this. It's a great story. - Yeah, yeah, for sure. I mean, I think that Iridium is amazing technology, but I also think that the Iridium Next design sort of pushed the envelope forward with more of a focus on internet. However, I think that Starlink's vision, or SpaceX's vision for Starlink is sort of like next level. It's kind of the Wayne Gretzky quote of skating to where the puck's going to be rather than where it is now. So Iridium feels more like where it is now, which is now already like not where it needs to be. Whereas Starlink is where it needs to be, I think so. All right, so let's circle back just quickly before we get into Starlink talk about some more of the upsides downsides trade-offs before we can understand Starlink as to why it is the way it is. The closer a satellite is to the service the lower the latency but the less coverage you get so therefore if you want to have the same amount of coverage as something like Iridium or like a geostationary satellite you have no option but to add more satellites. You can't just keep making the antenna bigger because if you do that then you're gonna need more power. If you need more power then you're going to need a larger satellite and if you need a larger satellite then you're going to need to have, it's going to cost you a lot more money to launch it and it's probably, you know, anyway, it makes sense to have lots of smaller satellites lower to the ground because it also means that it's no different to cell coverage. So inside the cities we've got metro cells, micro cells, pico cells, femto cells and as we get smaller and smaller and smaller, they're lower and lower power but because you have lots of them you can increase the density of connections on a mobile phone. It's no different with satellites, it's exactly the same idea. So you can have a lot more people connected in a square kilometer, for example, if you've got two or three satellites for them to hand off their data connections to, than if you've only got the one. So it's pretty straightforward. So anyway, that also therefore means that you can be more power efficient because you need less power, which means you can get away with smaller antennas on the ground, which is a massive plus as well. So you don't have to have a massive dish that you have to point at a geostationary satellite or iridium. You don't have to have this long sort of like, I don't know, what is it about, about 20, 15, 10, 15, 20 centimeter long, funky looking long antenna sticking out the top of your phone. Which is not, to be fair, it's not the end of the world, but I think people have gotten used to the fact that phones don't have long antennas. And besides, if we have a long thing, long antenna protruding from a phone that tended to break, remember when I worked at Dick Smith Electronics, we sold a fair replacement of the Motorola StarTAC little antennas that you would pull out because they would just regularly break. So anyway, showing my age. Anyway, all right. So anyway, it's fine. Hey, do you remember kids when they had antennas on sticking off the end of your mobile phone? No, no one remembers that, Dad. Okay, fine. Right. Okay, so here's the downsides, because it's not all upsides for having lots of satellites. The first downside is that more satellites are harder to coordinate, the handoffs between them because playing hopscotch between satellites is a pain in the neck and it's computationally intensive, it's quite complicated to do the switching, so that's definitely a downside. The more hops you have between satellites also means that you're going to probably need more ground stations to mitigate your latency, otherwise you're going to have random variable latency through the network, which you may well have anyway to an extent. So yeah, for example, if you've got a connection between point A and point B and you don't know if you're going to get there through one or two satellites or three or four depending on where they are in their relative orbits and how many times you've got to hop between them before you come back down to earth again, it's kind of that could be a bit of a problem. So managing that means you can't get away with as few ground stations, you need to have more of them. So there's that. And then I guess I have to talk about Doppler effect because it kind of drives some of the decisions and some of the stuff that's been talking about. So I've done a lot of talking, but I guess I just, what's your take on just not necessarily to that Starlink yet, but talking about using light or laser for essentially using inter-satellite comms, not using radio, but using light, rather like fiber optics without the fiber. Have you had much of a, how much do you know about that sort of stuff? I'm just curious. Well, you get to have a much higher throughput, but it's pretty complicated to do. Like it's not easy at all. - No. - Especially with as many satellites as Sarlink is gonna have. - Yeah, exactly. I think that the thing that's interesting is that radio has its place and optics has its place. And typically we've only really done optics at any kind of scale reliably using fiber optics. And once it's trapped in a fiber, you have complete control over it. There's no interference, there's no problems. Well, I say there's no problems. I mean, there's always problems, but you know. Whereas if you're going through open space, that's a completely different thing 'cause you've got all sorts of problems with diffraction in the atmosphere and you aren't gonna get a reliable connection. Alignment needs to be extremely precise. So these are problems that are very difficult to overcome at ground level, even going up and down through the atmosphere. Although there've been all sorts of different experiments that have been data even to satellites in deep space using lasers. But there's lots of error correction 'cause there's lots of scattering and it's not something we generally do because it's generally pretty unreliable. But the funny thing is when you're up there in space, a lot of those problems don't exist. So you don't have an atmosphere that's gonna scatter your laser. you don't have as much interference. And if you pick frequencies that are outside of the sun's spectrum, then you're not gonna get interference from the sun either. So it actually sounds like if you can solve the alignment problem, that suddenly this could be really useful. So one of the things that Starlink is going on saying that they're going to use for inter-satellite comms is not radio necessarily, it's going to be optical. So when we, and the other benefits you get out of that is you get away from the Doppler effect problem. So Doppler effect on radio. Any thoughts on that before I dive into so go ridiculously technical, just curious. - About Doppler effect? - Yeah, as it applies to satellite radio for each satellites and radio signals. - Not really. - No, okay. Well, I'm sorry, I'll just bear with me for a few more minutes. I just gotta, when I first got into satellite radio, because I really didn't get into satellite, - When I say satellite radio, I don't mean Sirius XM. I mean, you know, like amateur radio satellites is that what you realize is that when you've got your VCO and you're trying to tune the frequency as the satellite's approaching, you have to actually tweak the frequency up a few kilohertz. And as it's going away from you, you have to tune it back again. Otherwise you'll lose a signal. And so this whole thing of Doppler effect as affects radio signals is a thing. So, I mean, I think most people understand the idea Doppler effect as it applies to a sound. So as an ambulance is coming towards you, the sirens going off, it'll sound like a higher frequency. As it's in line with you, it'll sound like it's at exactly the right frequency. And as it's going away from you, it'll sound like it's at a lower frequency. And it's just a matter of the sound, the source of the signal is moving, and therefore the faster it moves. And if you're relative to you, you will see a corresponding change to the wavelength and hence V equals F lambda. Hence you change the wavelength, change the frequency. And this is a real problem. The other thing that people don't think about is they don't think about broadband. The problem with broadband is that if you've got a really wide channel, let's say it's a 100 megahertz channel and you're operating in 3 gigahertz microwaves, that's a pretty wide channel, 3 gigahertz is pretty low frequency but just some rough numbers, then that difference between the high end and the low end in terms of in terms of that wavelength, you'll be looking at 100 millimeters at one end and you'll be looking at 0.09966777 millimeters at the other end of that. So as your orbital velocity is going, so Starlink's going to be going at 27,000 kilometers an hour at its height. So if you actually do the numbers and all that, I'm not going to go into all the detail, although I've got it in the show notes for people that are interested, is that you're looking at an error of, well, and this is without the Doppler effect, your error from one end of the other of the spectrum is 0.000099 millimeters in wavelength. But if you go to high frequencies like light, so once you've gone from three gigahertz to 300 terahertz, which is essentially frequency for light, it's 100,000 times less of an impact. And so essentially you've taken something which is computationally problematic because you've actually got a varying frequency that's gonna cause a distortion of your data from the low end to the high end of your broadband spectrum you're trying to use, and you've essentially made it negligible. So optics have got a lot going for them because you don't have to worry about that so much. So anyway, all the numbers are in the notes if you really want to know. Right, so the other thing I thought about is, well, what about, 'cause people saying, well, optics is faster than electrical signals. It is and it isn't. The speed of light is fast, sure, but the speed of an electron is effectively the speed of light. I mean, it's 99.9999992% the speed of light, whatever. It's effectively the speed of light, pretty much. And the only reason that it's not the speed of light is because electron has mass. So the problem is the conversion. So you convert from light to an electron or electron back to light again, that conversion speed is really no different than radio. So if you're gonna go for electromagnetic wave to electrons or back again, so if you're in a switch and that switch has got fiber optic transceiver or it's got a radio front end on it, you're not gonna get any significant difference in the conversion between the two of them. So light's still better because you're not affected by Doppler. You can also do wavelength division multiplexing into a single signal to get more data over the same physical space. When you're on earth, the biggest downside is you can't use lights generally because of interference. So you put fiber optic cable in and everyone says, well, that's great, isn't it? But the problem is that fiber optic cable slows down light, which sucks. So yeah, 'cause the refractive index of single mode fiber, rough figure that we use about 1.5, 1.467, if you're really interested, which is about a 30% reduction of the speed of light. And that really kind of sucks. So whilst everyone thinks, oh, fiber optics super fast, it's actually not. It's not as fast as if you were sending a radio signal. It's not as fast as if you were using optics in free space. So when I say high speed in a vacuum, what they mean. It's simply the fact that you don't have to slow it down by forcing it through a fiber. Other benefits of radio, it's not affected by solid walls, mostly, unless they're made out of steel. You know what I mean? As long as you're not in a Faraday cage. Okay, so there's that. Anyway, so all right, that's all the pluses and minuses and so on and so forth. And it's really, optical is really only useful for inter-satellite links or ISLs because obviously, you know, to and from surface, got to go through the atmosphere. All right, so finally it's time to talk about Starlink. Yay. Nice. Okay. It's like, if I did- otherwise, if people don't like, have an idea of the pros and cons, then why Starlink's doing what it's doing won't make sense. That's kind of my problem, you know. It's like, people say, well, that's a lot of satellites, that's really bad and that kind of sucks. Like, yeah, well, actually, there's a reason. Okay, So, first of all, Starlink, originally, the first lot of satellites. So there's a whole bunch of different generations. So first one's 550 kilometer orbit. That's 340 miles. It's a 95 minute Earth orbit. I don't know why people like talking in minute Earth orbits. I guess it's easy to visualize, like the satellite's going to go past every 95 minutes. I kind of like that idea, but it's not very scientific. It's like 95.4 something or whatever. So round it up, round it down, it's close enough. I think that's due to the fact that a large chunk of people interested in satellites are interested in watching satellites. And then you do care what the time of the orbit is if you want to see another pass. It's a very human measurement. It's kind of like... Yeah, exactly. But it's very human because if you think about it, like 95 to 96 minute Earth orbit, you'd think "Oh yeah, that's not that different". It's actually really different. Yeah, it's like quite a lot of kilometers between a 95 and a 96 minute Earth orbit. But anyway, so yeah, 27,300 km/h orbital velocity. And each Starlink satellite is going to weigh about 500 pounds, which is 227 kilograms. Which for a satellite is actually really light, I think. It's fair to say. Yeah, for what it does, for what it contains, For the throughput it can provide, 200 kilograms is nothing. But actually even more impressive than that is the volume. The Starlink satellites are really small. They're like really densely packed, but they are an unusual shape. Like that's just not how satellites look like. They're flat packs and it's really weird. And they must have done a lot of engineering to get that done, but they are so light that the constraining factor once again became not the mass, but volume in terms of how much you can put on a satellite. And with how many satellites you have to, you know, have with Starlink, you want to put as many satellites as possible every launch. So I think if I remember correctly, they are able to pack 60, I think, or something like that into a single Falcon 9. Yeah. - Yeah. - That's amazing. - So that's the size. So the fairing is the size of like a city bus, right? And normally you just have one satellite there, one big satellite for geostationary orbit and they pack 60 satellites. Like that's simply unheard of. There have been launches, space launches in the past that would deploy 100 satellites on one launch, but not 200 kilogram satellites, but mostly CubeSats, like three kilogram or five kilogram, just tiny, tiny, super simple, super cheap, comparatively speaking, of course, satellites. Not like really powerful communication satellites, 60. - Yeah, no. - Wow. - It's amazing. - It's absolutely amazing. And one of the things that's interesting about it is just because they're small doesn't mean that they can't handle large throughput. So this is the other thing, right, obviously. So during a beta test, SpaceX did a test with the US Air Force, I think it was, and it was a C-120, 150, 180, I forget the designation of the plane, whatever it was anyway. one of the Air Force's large planes. And the test, they called it global lightning because they just like giving names like that to their stuff 'cause it's Elon Musk. Anyway, and the download speeds that they got in their beta test were 610 megabits per second, which is far beyond anything that's up there. It's just ridiculously so far beyond what's up there. And recently they're talking about one gigabit per second. Now I can believe one gigabit per second if you've got very few people on the satellite that you're talking to. And that may well be possible as like a raw data rate perhaps, but yeah, we'll see how that goes. But still, you know, it's good to shoot for the moon and you know, we'll see how that one goes. The other thing that Alan's been talking about is they've been touting a latency of 15 milliseconds to 25 milliseconds. And recently Alan tweeted saying 20 milliseconds was really their goal. And if you think about that, that's insane. I did the rough calcs based on the height and what you can expect for switching delays. And at that altitude, in theory, that should be possible if you only have-- like, if there's no hop. If you go up to the satellite and then you go straight back down again, that should be achievable. But it's still kind of a crazy thing, crazy kind of latency for a satellite. I do expect it to be definitely more than 20 milliseconds, at least for many years to come. If you watch SpaceX or like anything Elon Musk does, you'll see that, you know, he promises a lot of things. Half of them never come to exist. And the other half does come to exist much later than the schedule. he says it will, not nearly as good as what he promised and still pretty mind-blowing that it's even possible. So I don't think 20 milliseconds is going to happen soon, but whatever it is, I think it's still going to be just wow, that you can have like proper data, internet access from a 200 kilogram satellite switching from one to the other with whatever latency it is. It will be completely usable. - Yeah, well that's it. I mean, the thing to remember is that even if they do achieve 20 milliseconds, which is still impressive, that's still about twice what you would get out of decent broadband on Earth. The latency for my connection here is like 7 or 8 milliseconds depending on which server I'm trying to hit. And it's like, that's only problem for certain, sorry, if there are certain applications like video conferencing or like some low latency gaming that needs that sort of low latency. Most applications on the internet, like looking at a web page or streaming like radio or music or whatever else, that doesn't need to have 20 milliseconds of latency. Well, honestly, even for video conferencing, it will be negligible. Yeah, true. It will matter if you try to play a first-person shooter or make algorithmic stock trades over styling. Otherwise, it will be just fine. Yeah, I do think that it's good to have goals and it's good to shoot for the moon, as I I said before and I think that's great. But then it's certainly streets ahead of what Iridium Next is capable of. But obviously we're still in the beta testing stage, so we'll wait and see. Jury's out. Now, just in case you thought Elon Musk wasn't making grandiose enough claims, he's then saying that, "Oh, by the way, you know, we're just working on a Starlink thing." Well, the Generation 2 of Starlink, yeah, we're going to do like, we reckon we can crack 8 milliseconds. Now, I ran the numbers on this, and I think that it's one of those things that is a theoretical possibility only. And I think that the reality is gonna be, once you get a lot of people on there, you're never gonna get that as a sustained, consistent result. And I think a lot of that will be things like, as your phased array is tracking the satellite, which we'll talk about in a minute, the satellite's gonna go and pass overhead, and it's gonna go, as it's approaching, it's further away from you from the ground, therefore you're not gonna get eight milliseconds. Once it's directly overhead at its closest point to you, you might experience eight millisecond latency for a short period of time, then it'll be longer than that again. So it all depends on how you wanna slice it. And again, I really don't know how many people are gonna be that upset if you don't get eight milliseconds, and that's even if that's achievable. So anyway, interesting, but I suppose typical Elon. Okay, right. A couple of things about them. They're using Hall thrusters. All right, Hall thrusters. Yeah, that's the sound of crickets. Anyway, it's okay. I mean, but the thing is when I was reading about this, I'm like, don't they normally use xenon for their thrusters and they're using krypton gas? And I'm like, okay. Yep. See, this is one of those SpaceX things. So why is xenon used? It's because it's better. It's just a better gas for a Hull thruster propellant. Krypton, am I getting this right? Krypton? Krypton is not as good, but what it is, is far cheaper. And you know, when you're investing $300 million or half a billion in a TV satellite to put up there for 15, 20, 25 years, you don't care about the price difference between Krypton and Xenon. No, you just pay for Xenon. Yeah, but I don't think you mentioned this number, but when you want to have 1,500 satellites and then more in the future, and you're doing everything you can to be able to do that financially, and you're making them as small, as cheap, as lightweight as possible, you're putting them on like reused, five-time reused rockets, then yeah, suddenly it's like, Oh, actually we don't need the extra performance of Xenon. It's fine. Let's just use Krypton. It's cheaper. Why, why do you want to use Xenon? Yeah. But you're exactly right. It speaks to the design methodology that they're approaching Starlink with, which is completely different to pretty much everyone that I'm aware of. Cause it's like, well, if we're going to make lots and lots of these and they're going to be kind of expendable ish. So I mentioned that Iridium's lifespan satellites is expected to be 15 years and the last lot of Iridium, so it's Iridium Next and the Iridium satellites originally lasted 20 years. I mean, except for one, which I'll talk about at the end. But never mind. The point is that it was a bit of a whoopsie over the Soviet Union. But anyway, never mind that. It's fine. Siberia, I should say. OK, so never mind that. Point is, some of the other cool features of the Starlink satellites, I say cool in air quotes, is that they're they're touting that they'll have optical sensors to scan for local space debris and attempt to automatically avoid it. I'm actually not sure if how much of that is marketing blurb and how much of that is actually possible. I know that they're going to use the, I think it's NASA's tracking database for all the space junk, and they're going to try and programmatically avoid it. That I would believe. But anyway, nevermind that. So as much as we talked about laser links between the satellites for inter-satellite communications, they're not up and running yet but they're planned for the future. So I suspect they're going to have a lot of bugs to iron out with that but if they can make it work that's really, really good if they can. Just what you said that they'll have bugs to iron out. This again speaks to the complete 180-degree change in design and just product engineering, you know, difference between installing and everything else. No one thinks of satellites, a satellite communication network, as something that you gradually iterate over, like a software project. You just don't get to do that. It's too expensive and you don't have a large enough sample to do that. Maybe you have the first satellite or a first small batch of satellites and then you plan for block two in military contracting parallels or version two, as what software developers would say, with a bunch of improvements. They often fall into the second system syndrome, but that's another thing. But here, they're iterating. And every couple of launches, so every couple of months, there's a change to Starlink, and they're trying something new, and they're improving it and they're tweaking it. And sometimes those satellites, they fail and it's fine because it's like three out of 60 on the launch. - Yeah, plenty more where they came from. You're absolutely right. And the way they're approaching this is very much right. We know it's not gonna be, we're not gonna have any satellite links up and running first, but that's okay. We're gonna get the rest of it working and we'll get the bugs ironed out of that. And then we'll just launch some more and then we'll just launch some more. And every time we launch some more, we'll add improvements and refinements and we'll get better and better as opposed to let's engineer the whole thing on the ground and then launch it up there 'cause it's so expensive. And at the very end, I just wanna talk about why Starlink can get away with that. So, all right. The funny thing about Starlink design though is that it's changed multiple times since it was first submitted to the FCC for approval. Now I thought about, oh, I should go through all the different design phases. And I'm like, you know what? It's iterating and changing so much that I don't think it's worth doing that. So let's just talk about as of right now today. And let's just put your hand up and say, I admit that in six months time, it's probably gonna be different again because they keep iterating this thing. So this will be current as the time of publication and that'll be that. And just refer to the latest submission for the Starlink design whenever you're listening to this if you wanna get the current state. But anyway, as of April this year, the latest plan they submitted is the KU and KA band satellites. They are in orbits of about 540 to 570 kilometers. That's 500, sorry, 340 miles to 350 miles and inclinations of 53.2 degrees, 70 degrees and 97.6 degrees. There's gonna be 4,408 Starlink satellites in total. And then they've also said that they're gonna have seven and a half thousand V band satellites in orbit at 345 kilometers, which is 214 miles. So just about the bands, just real quick, that's just radio terminology. So V-band simply means 40 gigahertz to 75 gigahertz. KU band is 12 to 18 gigahertz. KA band is 26.5 to 40 gigahertz. That's just the different radio frequency bands. And they have different uses depending upon what you're trying to achieve. Some have more or less interference from the atmosphere. Like V-band, I think is one of those ones like to use for inter-satellite communication. So don't just think they're going to go straight to laser. They're not going to. But anyway, so and as of June this year, so only a month ago, SpaceX then said in the US, "Hey, we're going to apply for use of the E-band for their second generation constellation of up to 30,000 satellites." So it's like, okay, this is a lot of satellites, like a ridiculous number of satellites. No one's ever done anything quite like this before because their goal is to provide complete global internet coverage and the whole design ethos is all designed around one idea of it's cheap to get it up there and I don't care how long it lasts or not that much because they're saying each satellite is only going to have five to seven years of usable life that's it for a satellite some of those geostationary satellites have been up there for 20 years iridium like I said 15 year design life they're talking about five to seven years. And with that many satellites, it doesn't matter if one or two of them have problems. You've got plenty more up there. In fact, they're even designing them so that they'll burn up nice and neatly. So one of the things about the revision, the first release revisions they've been touting is 100% of, and this is their blurb, "All components of this design will quickly burn in Earth's atmosphere at the end of each satellite's life cycle." So they're designing them to burn up nicely, like completely, because that's the whole idea. Launch them up, they're good for a while and then burn them up, they're done, put the next lot up again. So that's a bit different. I want to impress upon the listeners just how ridiculous the 30,000 number is, or even even the smaller numbers. The total number of satellites, of objects in orbit, in low Earth orbit mostly, is about 5,000. Over the last, what is it, 50, 60 years, we've put up something like 5,000 objects into space. And right now, not even going to the crazy bright future plans of SpaceX. Just right now, before Starlink is even operational, they've already put up 540. That's already like 10% of everything. Those are big numbers. It's crazy. You're absolutely right. Yeah, that is it. They're insane. Absolutely insane numbers. And I just want to quickly wrap up on the whole other pieces of this puzzle and and then we'll circle back and talk about the how they can do it. So apply, in the US they've applied for 32 ground stations and as of a few weeks ago, they've got five of them approved in five states. So you still need lots of ground stations more than the other technologies because they're trying to keep their latencies low. And the user terminal on the ground, they keep talking about it, it'll be about the size of a pizza box. You can get pizza boxes of different sizes, but you know, again, Elon Musk, whatever. And it'll be a phased array antenna, which is, you know, so in other words, you light flat and the software will phase the antennas to essentially virtually direct it. Same kind of technology behind phase antenna arrays for MIMO, not new, been around a while. It's not the most efficient kind of antenna, but it makes up for the fact that it's over specced for its size. Anyway, it doesn't matter. The point is that it'll work fine, but it needs to be mounted anywhere, so long as you can see the sky, kind of obviously. It's not going to work indoors, so that's just not going to work, okay? I don't know how well it's going to work in a window, or on a window. Like, it might, but I don't think that would be very reliable. You'd probably be better on the back deck or on a roof. The more sky you can see unobstructed, the better. Right, we're talking about satellite communication. Even with orbits as low as 550 kilometers, the signal is still really weak and even a window has, you know, creates enough interference with radio waves at those frequencies, at KAKU frequencies that this will significantly degrade the signal. Exactly right. So yeah, definitely roof. Yeah, I just- I hate that whole idea of, oh, yeah, it's the size of a pizza box. And then people think, oh, cool, I have pizza inside in the house and I'm eating my pizza in the house. It'll be like that. Oh, please don't. No, no, no, no. And I think people also get fooled by this idea that your mobile phone has got a GPS in it. And I know it knows where I am when I'm inside my house or my apartment or the office. If anyone's going into the office anymore, COVID-19, never mind. point is that it's not really the GPS. It's actually getting your location from the nearest radio cell tower and it's not very accurate. So you need to actually put your phone right next to a window or be outside to get any decent GPS lock. So that's just a fact. That's just the way it is. So anyway, all right. Yes, satellite comms, keep that antenna outside. Everyone will be happy. All right. So as of June 23rd, 2020, there are 422 satellites in orbit. I think there might've been another load went up, I think you mentioned. In any case, it's initially only available in USA and Canada, and they're doing... You can sign up for beta testing if you want to. So unfortunately, Rodec, you and I are out of luck for that, but Shrug, oh well. Not the end of the world. I thought it might be interesting to talk about cost. And that is like for the end user. So there've been a whole bunch of people that have had a stab at it. I found one website, there's a link in the show notes, and it does a comparison of the two other most predominant satellite internet in North America and Canada providers. So one's Viasat and that ranges from $30 a month for 12 megabits down to 150 a month for 100 megabits down. Another one's Hughes Network Systems or HughesNet, they range from $60 to $150 a month and you get about 25 megabits down. Now both of those two combined have got two and a half million satellite internet customers. So So there's no question there's a market. There's absolutely a market. And whatever Starlink has, it has to be better than that in terms of both pricing and speed. Otherwise you want to track the customers across. And it's not like VSL or Hughes are going to be able to do anything on the scale that Starlink are. So Starlink pricing hasn't really been released yet, but Elon sort of let slip. He sort of hinted like, "Oh, maybe $80 a month kind of thing, maybe a hundred to $300 modem terminal cost up front maybe." But all that's typical Elon. So I don't know what that translates to in reality. And we're not going to know that until probably next year, I would think. Right. Okay. So why didn't anyone do this before? I mean, it's obvious, right? Jesus, it's so simple. Let's just launch 30,000 satellites. Why not? (laughs) Obviously, the cost of this per satellite launch was so high because, you know, it costs so much to launch a satellite. You would launch as few as you could get away with. It was better to launch 50 expensive satellites than 500 cheap satellites, people also weren't interested in latency or bandwidth for that matter back then. So, you could actually get away with putting satellites into higher orbits, with longer lifespans and all the economics made sense simply because it was so expensive to launch them. But it's not just that, I think also a lot of the technology that we've got improvements in digital signal processing is a big one. We can do a lot of DSP onboard the satellite now and the smaller chip fab sizes, you get much better power efficiency out of them as well, which is also big, which means you can put up a lot more compute up into orbit than you could previously. But the game changer really is the cost per pound to launch. And SpaceX have just slashed that with the reusable rockets on the Falcon 9s. There's no question that without that, this would not be possible. And SpaceX is right now, this year, they are their own biggest customers with reused Falcon 9 rockets. And in fact, in the past year, Falcon 9 has crossed many achievements in terms of their reusability, and it's all been styling launches. So last year in November, for the first time, a Falcon 9 rocket was used four times. So the first fourth time, that was the Starlink-1 launch. This March, they had the first fifth launch. So the first time a Falcon 9 was used five times. And it was the same rocket launching Starlink-5. And in a month, there will be probably the first time that the Falcon 9 rocket will have launched for the sixth time, the same rocket. So they haven't gone beyond three, like one original launch and two reuses for commercial customers. And they just keep reusing the same booster over and over again as cheaply and quickly as possible, like their own hardware that other customers are too afraid to use. But for them, it's not that big of a risk because they're mass manufacturing those satellites. - Exactly. Yeah, the economy of scale is, it can't deny the impact that it's had. And I guess that's also the crossover. So it's not just the cost to launch has dropped so low, but it's also the fact that the technologies allow them to mass produce micro satellites, if you can call it, 227 kilogram satellite, a micro satellite, but anyway, into a constellation that can actually work together. It's still not going to be cheap, though. And I think I was reading somewhere about one of the estimates was like 10 billion dollars, like cost to get all the satellites up there in the end, something like that. Right. So you asked why hasn't anyone done this before? But the thing is, the commercial success of Starlink and their slightly less ambitious and sometimes bankrupt competitors is not yet obvious. Notice how during the first Iridium, when the first Iridium constellation was launched, they promised heaven, they promised everything. They thought they're going to have so many customers and they had many copycats, But they didn't. No one wanted it. And it was, Iridium was sold off, you know, through bankruptcy for like pennies on the dollar. And right now there's another, this gold rush with Starlink and OneWeb, now bankrupt, and a couple of others. But it's not yet guaranteed that it's going to work. Even with Starlink's economies of scale and SpaceX being able to use their own rockets at cost and without the markup that they're charging even with reused rockets to their customers, it's going to be still so expensive with 1500 plus satellites that this might simply fail, that there won't be enough customers who both want it and can afford to pay for it. Absolutely right. And I think that it is going to be borderline on the economics. But I wouldn't bet against Elon Musk at this point. I think he's pulled off enough things, enough miracles, things that people said would never work out. I think he's not a good man to bet against. No. And certainly he has the patience and the deep pockets now for a long-term view on this. So, I kind of think that where the others have failed is that they didn't have multiple companies backing them and multiple other side businesses. I do believe that if anyone can make it work, he can. But maybe in the short term, it'll be confined to a smaller number of satellites focused in North America, and then gradually it'll roll out around the world, which is ultimately the goal, but that just might take longer than he thinks. But I'm hopeful. I'm optimistic. I think that there's a fair chance of success, and I think that they've got the staying power that some of the others that have filed did not have. - It's worth noting that the insane numbers on FCC applications have a lot to do with the rules of FCC and spectrum allocation and whatnot. And you want to shoot for the stars because if you only ask for 500 satellites to launch and it turns out you are successful, then you have a problem, right? So I don't expect that this will ever, even if successful, reach 30,000 satellites. Well, I don't want to say ever, but you know what I mean. But this is their wildest expectation of it. Maybe if everything goes right and all the stars are aligned, this maybe hypothetically could be possible. But this system will be usable with "just" 700 something satellites, which is still 10 times bigger than the next biggest constellation that exists. But if they can launch 60 at a time, then that's not a big problem. They're almost there. And then they can always give up. If it turns out to be a total flop, there was a couple billion, they'll survive, they'll be fine. - So it's funny you should say shooting for the stars there because I kind of, I think we have to just start wrap up shortly, but before we do wrap up is I wanna talk about the pushbacks because there's a lot of people around the world that are concerned about this and it's not all positive. Like, so with stars, I mean like astronomers. So, one of the problems is that astronomers have pointed out that with so many satellites in there, the number of satellites in orbit will outnumber the number of visible stars from ground to the naked eye. And some of them have shown some long exposures that have caught some of the Starlink satellites on camera. Like, I saw that and I'm like, well, okay, that's very, that's media clickbait, you know? Because I mean, I've taken photos in my backyard with my camera and I've caught satellite trails, you know? And it's like, and none of them are Starlink, I know that. So, it's not a new problem, but it's making an existing problem potentially worse. And it's funny when Elon's reaction to that was, well, you can just do photo stacking to delete them out of your photos. And that's technically true, you can. And that's exactly what people do already. But the density will increase, it'll just make it more problematic. But I do wonder though, if we're whether we're starting to cross that moment in time, there were a few moments in time in the past, right, with different technologies that have impacted astronomers. And I mean, the first one is when aeroplanes flew in large numbers at night around the world. So that was a thing. And right now you look up into the sky at any one time, well, maybe at the moment with lockdown and international travel, not so much, but rewind 12 months and just, you know, you'd see blinking lights flying in the sky pretty much any time of the day or night in most, you know, most semi metro areas around the world. And there was an also, so there was a time before that. And then light pollution from cities, you know, I mean, that's become so bad that like dark sky locations are kind of a thing now, You kind of have to go to a dark sky location just to escape the light pollution from the city. And so, you know, we're going to add 30,000 satellites with Starlink. It's like, OK, well, when satellites then are so numerous without image stacking, you're never going to get rid of them. It's like, I don't know, I guess to take the extremist position, you could say, well, with light pollution, what's the answer for the astronomers? As we say, well, because I don't want you to have to drive three hours to a dark sky location if it's even that close. After nine o'clock we're just going to turn all the lights off in the city. Can we do that for our astronomers? It's like, it's the, what's the trade-off? You know, at some point you have to say, well, the benefits outweigh the impact to astronomy, for example. But then again, I was also doing the notes for this yesterday, just finishing up on them, and I started thinking about the start of WALL-E, the movie. WALL-E, anyway. In the first 25 seconds, you're in space and you start to sort of like zoom in on the earth and fly into the atmosphere and you see the entire atmosphere of the earth just covered by millions of satellites. It looks like millions, it's lots. And I'm like, it's a pretty powerful kind of image that kind of stuck in my mind. So, I don't know, I could go one way or another. But then again, I also thought about WALL-E in the introduction and I'm like, "Well, if this is like 50 years ahead of when everyone abandoned the planet, most of those would have been dragged into the atmosphere and burnt up already. So the big satellites would have anyway." So I don't even know if that's actually a realistic view of the future anyway. It probably isn't. Yeah. I don't know. I'm kind of a defense about this. I read about their concerns and tried to read into it, but honestly, my personal take, and I don't want to come off as, you know, an Elon Musk fanboy or something. But I think that if this works, and that's not obvious, that this might completely fail or semi-fail, but if this really works, the benefits will be so immense that it'll be worth it. I don't think we'll get to 30,000, so I don't think, you know, I don't think it's worth extrapolating to those numbers, But it will be, I am absolutely sure that it will be a pain, an additional significant pain to astronomers. It will be yet another thing in their way and it will suck. But the benefits of having like total internet access everywhere. And yeah, it will be kind of expensive at first, but the technology will have been proven, will have been made cheaper and cheaper over time. That's really powerful. And why is SpaceX doing Starlink in the first place? Well, I think it is because Elon Musk is kind of crazy and just is set on this one thing, one goal to go to Mars, which takes a lot of money. And Starlink will either lose a bunch of money or will be super, super, super lucrative. And it's this stepping stone on SpaceX's vision. Like, you need a bunch of capital to be able to move forward with the Starship program and be able to make a big, big freaking rocket. And then when you have that, then hey, suddenly putting up into orbit a massive telescope, you know, way bigger than whatever is possible now. And without the extra complications of like folding it up, like with James Webb telescope, which is MS, but that's another story. Like that will become possible. So yeah, I just, I'm sorry astronomers, but I think it's yet another of those sad things that you'll have to deal with. - Yeah. I do, I mean, okay, as an engineer, I sort of fall on the benefits outweigh the drawbacks. And that's the astronomer side of things. But there's a couple other things to consider real quickly though, before we wrap up. and that is that there have been a lot of people saying this is gonna create a bunch of space junk. My response to that is that they're designed to burn up on re-entry. And if you don't boost them when they're that low, 550 kilometers is quite low, they're gonna get sucked back into the atmosphere and they're gonna burn up. So that problem solves itself in five to seven years, 10 years, they'd all be gone anyway if you did nothing about it. So the whole Starlink idea is that this is a constant thing. like every so many years you'll launch a couple of extra satellites to replace the old ones. And it'll be, you know, the ability to iterate over those times over that period then is much better than other technologies. So I don't think that space junk is necessarily the problem but the other one is collisions. And it's kind of crazy to think with all that space up in space, that collisions would even be possible but it's actually has happened. And I mentioned about a while ago now in this conversation But on the 10th of February in 2009, Iridium, the original Iridium satellite number 33 collided with Cosmos 2251, which was a defunct Russian satellite. And that was over Siberia. And both the satellites broke apart and they've now got a debris field containing two and a half thousand pieces. And so all those debris now, they're all tracked by the US Space Surveillance Network. And it's like, you know, it can happen. And when that does happen, those smaller pieces of debris will take a lot longer to eventually be drawn back down to earth again and disintegrate in the atmosphere. It'll take a lot longer. It'll eventually happen but it's going to take a while. So, by having so many satellites up there, the chances of a collision will increase and people can't say that it's never happened because it has. So, it's, I think that that's also an inevitability and it's something that we We just need to be a lot more coordinated, I think, globally with when we're adjusting satellite orbits and when we're doing launches. And it just needs to be better administered, I think. And if you can do that and people are more responsible about it, then that should be something you can manage. And again, benefits outweigh the risks. And if you've got 30,000 satellites, whether they get to 30,000 or not, if you've got a lot up there, losing one to a collision every so many years is probably not the end of the world. the concern being raised by people is the space junk that might create in the process. That's actually a huge benefit of putting those satellites in a really low orbit. 550 kilometers, that's pretty low. Because 550 and 800 kilometers, that might not sound like a huge difference, but it is a huge difference in terms of how long the satellite will last in orbit. So there's not much out there on 550km orbit because of this, because not much else wants to stay so short a time in orbit. Yeah, no one else is crazy enough to put them that low. Yeah, exactly. So the biggest danger to Starlink is Starlink, because of how many there will be there. And you know, it will be pretty much, you know, everything else on 550 or so orbit will be a rounding error compared to the number of Starlink satellites. And if you have an operational satellite, you are able to track it and position it with such accuracy that it's fine. It's when it breaks, then it starts to become a problem and you just have to track it really carefully and maneuver around it. And actually, I think most satellite maneuvering is pretty much done manually, like someone observes it and sends commands to the satellite. But they already have so many that they're forced to create a proper automated system that just reacts to these situations. Right? So, that's kind of the benefit, that they have so many and they're pretty much the only users of this orbit, that they'll make it work. And if not, well, it will decay and degrade in a reasonable number of years. And then Elon Musk can say, "Well, we tried and it's all burnt up now, so, oh well, off to Mars." in any case. All right, well, if you'd like to talk more about this, um, you can reach me on the Fediverse at [email protected], uh, on Twitter at John Chigi or one word or on the network at engineered underscore net. Uh, if you're enjoying pragmatic and want to support the show, you can, uh, via Patreon at or one word, uh, with a thank you to all of our patrons and a special thank you to our silver producers, Mitch Bilger, John Whitlow, Kevin Kosch, Oliver Steele, and Shane O'Neill, and an extra special thank you to our Gold Producer known only as R. Patron rewards include a named thank you on the website, a named thank you at the end of episodes, access to raw detailed show notes as well as ad-free, higher quality releases of every episode. So if you'd like to contribute something, anything at all, there's lots of great rewards and beyond that it's all really really appreciated. Of course there's lots of other ways to help like favouriting the episode in your podcast player app or sharing the episode or the show with your friends or via social. Some podcast players let you share audio clips of episodes so if you have a favorite segment, feel free to share that too. All the things that I just listed are ways that you can help other people to discover the show and can make a big difference. Now if you'd like to get in touch with Radek, what's the best way for them to get in touch with you, mate? That's going to be either through Twitter, Radeksp, R-A-D-E-X-P, or through email, and you'll find my email at You're working on a product that's called Nozbe. Did you just want to take a couple of seconds just to quickly talk about that? - Oh yeah, sure. Okay. (laughs) For the past couple of years, we've been working on this completely new product. We call it Nozbe Teams and it's essentially a to-do app for teams. It's a product to help remote teams mostly communicate, collaborate, get their stuff done. And after four years, we finally just today as we're recording this, kind of officially officially launched it. We're on product hunt, yada yada. And I'm really proud of this product. So if you're a team and you want to get away from Slack and email and meetings, then Nozbe Teams might be for you. - Awesome. All right, cool. Well, thank you very much for that. And yeah, by all means, check it out and awesome. All right, lovely. Well, once more, a special thank you to all of our patrons. A big thank you to everyone for listening as always. And thanks for coming back on the show, Radek. It's been great. Thanks so much for inviting me again. - Anytime. (upbeat music) [Music] [MUSIC] (upbeat music) [Music] (upbeat music) (upbeat music) (upbeat music) (upbeat music) [MUSIC PLAYING]
Duration 1 hour, 17 minutes and 44 seconds Direct Download

Show Notes

SpaceX Achievements:

Radio Bands:

Competing Satellite Technologies:

Links of potential interest:


Radek’s Recommendations:

Nozbe Teams:

Episode Gold Producer: 'r'.
Episode Silver Producers: Mitch Biegler, John Whitlow, Kevin Koch, Oliver Steele and Shane O'Neill.
Premium supporters have access to high-quality, early released episodes with a full back-catalogues of previous episodes


Radek Pietruszewski

Radek Pietruszewski

Radek is a software developer and is behind both Tadam App and Nozbe, podcasts regularly on The Podcast and is an avid follower of all things Space X.

John Chidgey

John Chidgey

John is an Electrical, Instrumentation and Control Systems Engineer, software developer, podcaster, vocal actor and runs TechDistortion and the Engineered Network. John is a Chartered Professional Engineer in both Electrical Engineering and Information, Telecommunications and Electronics Engineering (ITEE) and a semi-regular conference speaker.

John has produced and appeared on many podcasts including Pragmatic and Causality and is available for hire for Vocal Acting or advertising. He has experience and interest in HMI Design, Alarm Management, Cyber-security and Root Cause Analysis.

Described as the David Attenborough of disasters, and a Dreamy Narrator with Great Pipes by the Podfather Adam Curry.

You can find him on the Fediverse and on Twitter.