Tech Brew Ride Home - (Bonus) Internet From Space! With The Internet Society

Episode Date: November 24, 2022

Check out the extensive paper on LEO satellites and Internet from space at InternetSociety.org/techmeme Learn more about your ad choices. Visit megaphone.fm/adchoices

On April 4th, 2023, around 2 in the morning, a man was found stabbed multiple times on a sidewalk in downtown San Francisco. Hey, who did this to you? What happened next turned the story into a political firestorm. Reports have identified the victim as Bob Lee, the founder of Cash App. From Bloomberg Podcasts, this is Foundering, the Killing of Bob Lee, beginning April 16. everybody, Brian here. I long ago promised you that if we ever did a sponsored episode, I'd let you know that we were. Well, this is a sponsored episode. As you've heard, the Internet Society has been advertising with us all week to get the word out about their paper on issues surrounding low-Earth orbit

satellites and the whole internet from space. They wanted to know if they could have a longer conversation about this topic, and I said yes, because they're the Internet Society. I've long respected their work, and also because the whole space tech industry is something I find fascinating. So enjoy this discussion, but know that the Internet Society paid to have this discussion. So be sure to check out the paper we're talking about at internetsociety.org slash tech meme. Welcome to another bonus episode of the TechMeme ride home. I guess a special Thanksgiving holiday sort of bonus episode for us. Today, I couldn't resist doing this. The Internet Society reached out.

You might have heard their ads on the show this week. They have a new paper out that the ads hopefully have told you about all this week, about Internet from Space. And we have from the Internet Society today, Dan York, to talk about that paper, to talk about, you know, satellites, space tech, all that good stuff. Dan, first of all, welcome. Great. Thanks for having me on. I'm a long-time listener, so I've enjoyed your show over these years. Well, thank you so much.

I think what we should do is first start off just real briefly with the Internet Society, because as I said, when you folks reached out, I couldn't resist because I've respected the work that you've done all these years. So just briefly tell us about the Internet Society and what your goals and projects are. Sure. So we were founded in 1992 by a group of early internet pioneers. And we're a global nonprofit. We're working really to ensure the internet remains a force for good for everybody. And it's a bigger, stronger internet that's trustworthy, open, globally connected, all of these things like that.

We have about close to 100,000 individual members. It's free. Anybody can join. We have about 130 chapters. And we have close to 100 organization members that are all part of our organization. We do a lot with the open standards world, supporting a lot of the work of the Internet Engineering Task Force and other groups like that. And so that's what we're doing. This year we're really focused on how do we help connect the unconnected and grow the Internet,

this global network of networks to include the about third of the world that's still not on. And then we're also focused on a lot of work around ensuring that when people get on the Internet, and for those of us who already are, that it's as secure and trustworthy as possible. So we have a lot of projects in those different kind of areas. Well, given that that's your main remit, as you just described it, you know, getting everybody online in a fair and equitable way, it makes sense that y'all are taking a look at the Internet from space, space. I should have some sort of a reverb thing going there. So. Space lasers. Let's get into that. So, you know, I'm, as I like to say, as people like to say, I'm old enough to remember that the idea. idea of getting data from space. So even cellular networks from space and stuff like that,

or the internet from space has kind of a long history. And not one that's bathed in glory. I'm thinking of things like Arridium and stuff like that. So however, as we've discussed on the show, space tech is pretty big right now. And one of the first obvious use cases for it is to provide people with. connectivity. So can you, let's start by just asking why now? What has happened with the industry

and the tech that has allowed this to be something that seems to be viable now? Sure. I mean, like you said, we've had a satellite-based internet access for decades, but pretty much mostly using the geostationary or geosynchronous, what we often call geosatellites that are based out about 36,000 kilometers away from the earth. And that's been what we've had. We've had those from a variety of vendors for a long time, and it works. It gives people connectivity in remote regions. But the challenge is just it takes a long time for packets to get all the way out to 36,000 kilometers and back.

And you wind up with a latency of around oftentimes 600 plus milliseconds, which, you know, we're recording this on a Zoom call, etc. You couldn't do that over a geosynchronous connection. You just can't. So, you know, those kind of things, whether it's, you know, real-time communications or or virtual worlds or gaming or just any kind of, you know, any kind of communication doesn't work well over something with that much of a lag. So now, you know, we're seeing people look at these systems in the low Earth orbit or Leo.

And that's generally around, you know, 150 to 2,000 kilometers. But most of these satellites are kind of around 400, 500 kilometers away from the Earth. And they rotate quickly. They go around the Earth very fast. And so you don't have, we're, we're, we're, With a geosatellite network, you might have maybe three to cover the entire Earth at the equator, and you have that kind of connection. With Leo systems, you have, you need a whole, what they call a constellation, which might be hundreds or thousands or tens of thousands. Some of these satellites,

there's even one that's out there for 300,000 satellites that could go around the Earth in some way. So you have these large constellations of networks, which give you low latency, high-speed connections. You know, some of the speed tests will generally say that you can get somewhere around like 40 milliseconds of a latency, which is perfectly fine for a call like this. We can do those kind of things like that. It's not as fast as like a fiber connection, but heck, you can get it wherever you are and where you don't have fiber and that kind of thing. What's to your question about why now, we've had an interesting intersection of a number of different things. One of them is we've now got mass produced satellites in the past, you know, for building those geosings. satellites or the things. They might be the size of a bus. They took, you know, months to years to build,

their custom built, everything else. Now you've got companies like SpaceX and Amazon who are looking to stamp these things out. You know, it's just, it's mass production of satellites to get out there. And they're smaller, right? Like that's the other thing. And they're smaller, like a car, you know, or something. As opposed to a bus or an RV or something like that, yeah. Exactly. They're small like that and they do it. The other pieces, we've gotten, you know, mass produced antennas, what they call a terminal, a user terminal, but it's an antenna. That, you know, with a geo satellite, you would go and you point the antenna toward it,

and that was it. Well, when your things move, you can't just go out there and move the antenna around, but we've got these electronically steerable antennas that automatically track along that, including while things are in motion, I mean, in tremendous advances in that. And then the other aspect is you've got,

you know, remarkable stuff happening in the space business for launch systems. You know, you look at SpaceX, with the ability to go launch these things, you know, come back, reuse the rockets. I mean, this is, the economies of space launching are just amazingly different than they were even 20 years ago. Back in the 90s, there was a push to get Leo systems up. But it didn't, it didn't work, you know, because it didn't have all this stuff happening at this time. So with this really interesting time when people were able to do this and provide these kind

of systems, it's amazing. So I'm a dumb person. So I'm a dumb person. So I, I didn't know one aspect of this. So you have the constellations. Like you said, you know, folks trying to set up these systems are launching sometimes thousands of satellites. So it's sort of like a swarm as opposed to, you know, even a net or something like that. I knew that. I knew that you need a satellite, you know, going back to like Bloom County and people making

fun of getting HBO, stealing HBO from your backyard satellite. I understand that. What I never factored in is that there's also like terminals. Like you have to have ground stations that connect you to, at least right now, because we're going to get into how that might change, but at least right now, is it just that the signal is just still bouncing back and forth? So you need the ground stations as well to connect to the actual internet, which is still a terra firma kind of thing. Yeah, right, right. So you've got it, you've got this space-based, you know, constellation of satellites. So you've got a network there. And that network has to connect to the rest of the internet and all of the other millions of networks that make up the internet. And that connection comes through a ground station. So, and sometimes they call it a gateway, they have different terms for it. But, you know, it's essentially it's the large antennas and things that you would see on the ground that take the signal. Because the signal has to go from your user terminal, from your antenna that connected to your, you know, Wi-Fi. network or whatever, it's going to go from that to the satellite, bounce off the satellite,

and then go down into the ground station, which then connects it out to the rest of the internet. So for this to work, it has to go and the satellites need to be in range of a ground station, subject to the other thing we'll talk about, but they generally need to go and do that. Now, the Leo satellites are moving so quickly that the antennas are actually able to track multiple satellites so that the signal you send to go load the TechBeen Ride Home webpage or something, that may go to the ground antenna and then the response may actually come back off a different satellite. The antennas track all this stuff and make it all work. But they do, traditionally, they've had to be in range of a ground station. And that's the piece that's there.

Well, another thing that I learned from your paper, which I was remiss at not saying at the beginning that you can get it at internet society.org slash tech meme. there's exciting tech coming. We're still basically doing radio waves now, but we're getting to the point where we can do lasers and things like that, which would allow actual CDNs. It would allow more of what I'm talking about, like the operations to be in space.

Yeah, explain that to me. Sure. And actually, Starlink, SpaceX and the Starlink service of SpaceX, We need to their system. They already support this with some of their, the satellites have been flying up since 2021 or so. They just had an announcement the other week where they had some people in Antarctica who were using a Starlink antenna to connect to the internet.

But if you think about it, there's not really any ground stations that connect to Antarctica to the rest of the internet from there. But what was happening was they're connecting up to a Starlink satellite, which is then using what's called a, well, it's a laser link, but it's called an inter-satellite link sometimes that was connecting to another Starlink satellite and conceivably to another and to another. We don't know the exact way it was mapped there, but it went to what found a station that could connect down to a ground station. So it went up to satellite, satellite, satellite, satellite, satellite down to a ground station. Suddenly, this opens up great possibilities. There's some people in one of the Reddit forums around this who are talking about how they've gotten, you know, connectivity from way up in northern Canada and Alaska where they're able to go and get this kind of connection, again, without being near a ground station, but they're being able to get that it's traveling across there.

Now, we don't, I don't know, and I don't know it Starlink has necessarily talked too much about the capacity of all those, but it is possible now. And so we're able to reach places that you just could not get before, which is amazing. I want to come back to the capacity question. But so right now, in November 2020, the main advantage slash use case slash best practice for this technology is, as you say, connecting people in places where, you know, it's not cost effective to string cables and et cetera, et cetera. Right now is that what it is, like, connecting the course. corners of the world that are unconnected? Is that the primary goal right now? Well, I think it's a combination of use cases. They're pretty interesting. That is certainly one of the primary places, right? Is people who don't have other options, they can go and get it.

One of the interesting things is Starlink has enabled it people to use it on recreational vehicles or actually in motion. And so boats, planes, but also just individuals driving around. You know, it was kind of funny. In the early stages of the release, I caught a thing, a stream, a live stream on Twitch or something of these guys who had modified a Starlink Contenna to run on their rally car in some kind of race down in Baja, California. And they were driving around live streaming the whole time and everything else. Now, that was just, they were hacking the machine to go and do it.

But today, that is actually a supported use case. And people are being able to go and use this in a variety of different mobile ways. that they couldn't do before. There's also a lot in disaster response in kind of ways looking at that. I volunteer a bit with the IT DRC, which is the IT Disaster Resources Center, and they are the folks who deploy to disaster areas like Hurricane Ian in Florida. They brought down their trailers with their Wi-Fi and mobile phone setups and stuff, and they use Starlink to go and make connections back and be able to bring connectivity in there,

not only for first responders, but for the community as well. So tremendous capability there in building resilience. So, again, I assume the Internet Society is interested in poking at these ideas because ultimately the goal is to get every human being on the planet that wants to be connected to the Internet. But what about the affordability? Because, you know, I would love to test out SpaceX's Internet. whose name escapes me right now, but like it is $600 up front just for a tinker tool.

And then it's like $100 a month or more or something like that. So like what are we thinking in terms of if this is going to bring, you know, we're not only thinking about, you know, Western rich people that can RV around the country. We're also thinking about people in remote third world areas that will eventually get it. So the affordability question, Do we know when that's going to come down? Is it possible to come down? So this is one of the big questions we have, yes, because Starlink, you're right. If I want to get it here in Vermont where I live, it's $600 and then it's, you know,

around $100 or so. Now, Starlink started out, SpaceX started out kind of doing that pricing globally. But then they have adjusted it. They've changed it to be more reflecting local pricing in other markets, but it's still pretty high. And there are certainly costs around creating that hole in. antenna and creating that whole kit and stuff around that line that are somewhat fixed. So I don't know. And this is one of our big questions.

Will this be something that can be used everywhere? I think what will be interesting to see is as we get more competition out there. Right now, Starlink, SpaceX is the one. One web has about two-thirds that are satellites up right now. Amazon is going to start launching next year and Telesat from Canada. Some others are going to be coming on. and there's like 16 constellations out there that are going to be launching over this next while. And it'll be very interesting to see what business models emerge, what can happen with competition,

can people bring down the pricing on these different parts so it is actually accessible? What kind of government programs will happen in various different regions? What kind of subsidies or what pieces are out there? This is certainly one of our big questions. I'd love it to see if it would be something that could be affordable and help those folks in those remote regions. We'll have to see. You know, because it's important. It's not just connecting people in some areas.

It's also helping people who currently have solutions that don't give them a lot of connectivity. You know, they might be in an area that, but all they've got is a geos, and it's not really great. Well, another thing that I never thought of that your paper turned me on to was we're sort of talking about business models here. Like, you know, will the price come down with competition, et cetera, et cetera. One of the things that we don't know, I think, according to the paper, is we don't know the capacity of these systems yet. Like, how many users do we think SpaceX's system has right now? It's like 800,000 or something like that. Yeah, that's what they're saying is around that.

So do we know, can you support a billion users on a? I don't know. I was at a presentation. I did a presentation for the Armenian Internet Governance Forum last week, and they asked me that question. And they're like, can we put everything on there? Can we put all of our IoT devices? Can we put every kind of thing onto the network? And the answer is, I don't know.

We don't know. You know, with only 800,000 users across the world on that kind of system, we don't know where it goes. We see different questions raised in some of the online forums. Some people who have had fast connections are now seeing less and other stuff. And SpaceX introduced a bit of a tiered pricing. that you would get, you know, it's not, you would get a, you know, prioritized traffic up to a certain mountain and beyond that, it was kind of best effort and stuff. But the hard part we don't know is

how much of that is just growing pains right now, because they've only got only got 3,500 satellites up there, which is more than we've ever had, okay, just to be clear, the history of the world, it's, you know, up there right now. But they're on the path to get to about 4,500 in that system, but then they have another shell of another 15,000. So at what point, what will be the capacity of this whole system? One of our hopes for the vendors that we were certainly looking for is, can they help us understand that? Can they make their information available to researchers and people who are doing this kind of work? That's a kind of a key piece. We'd love to be able to understand and do testing and stuff like that like we can do for mobile networks or broadband.

Right. Well, and we're talking about capacity in terms of service. end users and things like that and bandwidth and stuff. But another thing to think about that, again, I hadn't thought about before the paper, was that, you know, optimal orbits are a limited resource. And right now, there's, if it's first come for a serve, if you can get a satellite up there, boom, you've got that orbit, right? And so we don't know necessarily, I must, there's no, there's no, traffic cops saying

stay in your lane, pick this lane, there's best practices, there's suggestions from like the UN and stuff, but like right, so like the fear would be people get there first, clog up all of the best lanes or orbits and distances and things like that. And then oh crap, since we didn't plan it out, like now what?

Like if things get too crowded or something. Yeah, yeah. I mean, there's a whole, I mean, we don't actually have, we'd have to have a, whole other show just to talk about how these things get allocated. But the net of it is you talked, like a company talks to a regulator, gets an allocation for spectrum and for orbits. The regulators talk among each other, the international telecommunication union has a role in here. But there isn't, there, it's, it is a lot of first come for serve. You know, now they have the ITU have what some

guidelines on here, which basically says, you know, once you get the allocation, you have to launch, you know, about, I think 10% of your constellation orbit within the first two years, and you have to have 50% up in five years and 100% in seven years. So they put some limits on there so that you don't get people just saying, I'm going to take this, you know, altitude. Because, you know, there is all this different. We talk about constellations in terms of shells, like a shell around the Earth and at different altitudes and orbital inclinations and all that kind of stuff that's there. And that's the part that, you know, It is limited. There are concerns around all of that.

And so we don't want people just to sit on that and not use it because it's a shared resource. Well, one of the things that would help ameliorate that problem would be if all of these various satellite systems had some form of interoperability. So that, sure, maybe you would gate it in terms of if I'm paying for this service versus that service, I'm using this satellite versus that satellite. But at the same time, if all of the satellites could talk to each other, share their networks or whatever, that would make things better. So where are we on that in terms of interoperability? Well, I think it's still too new, right? Because, again, we only really have SpaceX with their Starlink network up there and in one web and some pieces of more come on. But, you know, it is interesting.

If you look at like in the United States where I live, when we deployed mobile networks, right? You had Verizon, you had AT&T, you'd Sprint, Nextel, whatever. You had your own separate phone, which only worked with that infrastructure. We basically built like three different infrastructures, four different instructors across the United States for mobile phones. Whereas in Europe and most of the rest of the world, Asia and everywhere else, they built one infrastructure. They agreed on using GSM.

They did this. And so you had people building on infrastructure. You had that, but you had shared, you had roaming fees, you had lots of stuff on that line, but you had a shared common thing. This is a question. Which way will we go? Will we go where we have completely separate infrastructures for, for, you know, for Starlink, for OneWeb, for Amazon, for, you know, all of these will they completely build separate infrastructures? Or will we have more of a shared infrastructure that is possible to share? And I don't know the answer. I'm not even sure they know the answers. But that's why you want to talk about this right now, because we still have time to make those things happen before everybody else, before Project Kuiper gets all of its satellites. we can make that decision now?

Yeah, exactly. That's the thing is that's why we wrote this document was partly to say, these are the questions that we as a society need to be talking about and asking as we're launching all of these satellites and getting all these systems up there. Because if we don't, we'll go at the whims of whatever entities are doing it. I think it's more important that all of us are involved. We'll go ahead and end in a second with sort of your recommendations that you came up with in the paper. But one more question that I had never thought about, which is the security of this, in the sense that geopolitical stories about, like, you know, different countries spying on each other.

There are certain Chinese vendors that are not allowed to create or we're not allowed to use certain Chinese vendors for our cellular networks now and things like that. One of the things that I hadn't thought of before the paper, again, is the fact that, like, let's say that satellite internet becomes, like, the main backbone of a communication system. Like, that's a huge national security concern where how would we know that some other country or some adversary can't shoot something up there that, you know, with a magnet that sticks on there and then can listen in? And like, I'm thinking of like that pipeline that mysteriously blew up in the Baltic Sea or something recently. Like, is this right now something that I hadn't considered, except for the cables that go overseas, in your country, you do have control over the wires in your country, the towers in your country, the pipes in your country. But if it's up there and someone does something to it, then what are you going to do? Well, right. And it's not like you can't, you can prevent the satellite from going over your, you know, with thousands of satellites, they're going to be going over your country. Unless we add lasers to all of our satellites, offensive laser. Well, right. You know, like so Starlink is going to be going, I mean, satellites go over China, right? But China will never allow Starlink to operate. Well, I would suspect, who knows, I would suspect they won't allow Starlink to operate there. Likewise, China has several different constellations they're going to plan to use. I highly doubt the United States. will allow people to subscribe to those services from within there.

So you're going to have all these satellites orbiting the Earth in different forms, but parts of the world will not be able to access or use those different systems, at least legally. But that is a, I mean, it's a fascinating question. I think there's a lot going on. There's obviously countries too that have anti-satellite systems. There's all those different things that does introduce a whole new level of security concerns for our internet access levels, yes, which we haven't had before.

All right. So at the end of your paper, you all have a bunch of recommendations in terms of not just, it's not, this isn't a policy paper where it's like this is what we should do. It's more that you guys are raising the questions that you'd like to see the debate sort of coalesce around right now. So just hit me with some of that stuff like what we should be talking about lobbying for thinking about right now as we're building this out. Sure. I think the main point is this space-based internet, these Leo satellite systems, you know, really have some tremendous opportunity. Ideally, we want them to be able to work with the rest of the network of networks that are also, you know, with, you mentioned cables, with wireless, with all of those things. We want them to work on par with those. We want them to be similar, which means things like supporting the latest open Internet standards, allowing competition that allows, you know, multiple providers to be in there. there, you know, using, you know, peering with other networks, supporting the latest routing, all those kinds of things. You could pass a law that says, common carrier style, like if you're putting a satellite up there that's going to beam back down in my country, you also have to have some sort of interoperability with a competitor that could do likewise or something like that.

That could be something out of that. But yeah, I mean, we really want to ensure that the systems are being thought about in that way, that they're secure, right, right, but they're confidential, all those things. Yeah, I interrupt you. So, yeah, go on. Yeah. No, no, no. I think that those are good points.

And then also, you know, things like advocating for the fair use of spectrum, allowing competitors or allowing new competition, new entrance into the market. And also that piece I mentioned around the capacity, helping being transparent, letting people understand what is really happening, how resilient, what's the uptime, you know, how, what are these kind of things that we can really learn about? Well, that's a key. That's a point that, again, the paper made me think. of that I never thought of before in the sense that we all sort of understand internet connectivity

is a utility now, except for the fact that in the 90s, internet service providing as an industry was not necessarily built out as a utility in the same way that electricity and energy and things like that were. So would we have an opportunity again to learn lessons and be like, we're not going to nationalize things necessarily or whatever, but we could build in sort of the safeguards and the best practices that would allow it to function more as a utility that would include the public trust and the commons and things like that. Because we understand while it is a capitalist system, it's still a utility that is vital to, any country in the 21st century?

Yeah, I mean, the pandemic has certainly shown us that it's absolutely, you know, internet access is absolutely critical for every aspect of our life. It's woven into the fabric of what we do. So certainly any of the things that we can take a look at and the business models, the government regulation models, the different aspects to help in this space. You know, even in this space is changing so fast. Even as we wrote this paper, you know, and we're getting it out the door, the European Union announced an initiative to go and launch their own constellation, which could

have a different business model being more government kind of funded in some way, which is different from what we have right now in some of these constellations. So I think it's an exciting time. There's a lot going on. There's so many things happening. And the industry is so young and changing so fast. You know, we will be the Internet Society. We will be revising this paper next year to to see what's changed in the year that's there. And already, we know some revisions will need to make. It's an exciting time, but it's a time that it's an opportunity we have to shape these conversations, like you said, and really help decide what does Internet from space look

like? I mean, access from space. This is the last one because I skip this in my notes. But let's mention before we go privacy in the sense that for all the ways that we're surveilled all the time now because of all. our devices and things like that. Again, the paper made me think, in theory, I still have a fail-safe physical mechanism if I want to shut it all down.

You know, I can unplug the cable to my house. I can turn off my device. I can do all these things. But if what the internet from space comes true, it would be sort of the promise of, oh, it's just in the ether. It's everywhere. So what have you guys thought about that in terms of trying to future-proof the idea of, if the Internet is everywhere, it's like in the air, how can we safeguard privacy and security and things like that? Well, I mean, that's really the key question, is how do we do that when we're making those things?

Because, you know, we've just talked about here. We're talking about things like SpaceX Starlink, One Web, Amazon Project Carpenter, that are focused on providing kind of large level broadband internet connectivity. But there is a whole host of other providers looking to launch into Leo space that are, for instance, providing connectivity directly to cell phones or providing messaging or providing other pieces like that. So there's a telecommunication aspect of that. And I think ultimately that is where some of these different folks want to go, get you to the point where you can just have your smartphone and be able to have ubiquitous internet access wherever you go.

Or the chip in your clothing is probably five years away, yeah. Well, right. And so you do get in those questions of, like, do we want to trade off that convenience of being able to know where your shirt is or something like that to the potential, you know, privacy ramifications of having your shirt tracked wherever it winds up being? So those are those societal questions we have to answer. And if we don't ask those questions, you know, we'll wind up with whatever some company decides is the right thing to do.

Well, Dan York, thanks for bringing this all to our attention. Thank you for this paper, which again, you can find at the internetsociety.org slash tech meme. I encourage everyone. Instead of talking to your annoying relatives this holiday, go ahead and download this paper and do some thinking about what this means, especially if you're in the corner of of the industry that are going to make this happen. Dan, is there anything else that you want to plug about the Internet Society, or is the paper

good enough for now? Well, the paper's good enough. I like you, I'd say, please go share this, read it, share it, critique it. Tell us why we're wrong. Right. I mean, let us know, you know, this is the time. We've got to probe at this and ask these questions. So, you know, please look at that, you know, tell me why we're wrong, whatever.

But, you know, share it with people. help get the word out that these are the questions we need to have. I'd also encourage people. You can go to internet society.org as well and just go there, join as a member, see what we're doing. You know, if your organization is interested in being involved, you can join as an organization member. There's a lot of different activities and things that we have doing really just all in service of, looking to build a bigger, stronger internet for everyone. So thanks for having us on, Brian. Yeah, and that's why I love the Internet Society because it's not, okay, these are the rules from on high, It's more, we want people to think about this and have debate and raising these questions and things like that.

Right. We're not the protocol police. Yeah. I love you, folks. Thank you, Dan. Thank you, Brian.