The Satellite Solution That Won’t Scale - Episode 666 of the Community Broadband Bits Podcast

Christopher Mitchell (00:12)
Welcome to another episode of the Community Broadband Bits podcast. I'm Christopher Mitchell. I'm at the Institute for Local Self-Reliance in my abode in St. Paul, Minnesota, talking with one of my favorite guests of all time, Sascha Meinrath, the Palmer Chair for Telecommunications at Penn State University. Welcome back to the show, Sascha.

Sascha Meinrath | X-Lab (00:28)
Thanks

It's wonderful to be here. I'm happy to be your favorite guest of all time, though I don't buy it at all.

Christopher Mitchell (00:40)
As of right now, you're my favorite guest forever.

Sascha Meinrath | X-Lab (00:43)
Very good.

Christopher Mitchell (00:44)
I can't tell you that you were my favorite guest an hour ago or that in an hour you'll still be.

Sascha Meinrath | X-Lab (00:47)
That's right.

I'm your favorite guest for all the time of this podcast.

Christopher Mitchell (00:53)
That's right.

And if you listen to it in 10 days, it will still be that hour. So it's really a magical thing like that. We're going to talk a little bit about what we're hearing on Bead lately. We're going to talk about measuring things because Ookla released some really innovative new tools that no one had ever thought of before. Certainly not one of my guests. We also are going to talk a little bit about

Sascha Meinrath | X-Lab (00:57)
That's right.

Hmm.

Christopher Mitchell (01:13)
Satellite service you've been measuring that and there's some more news on that front But we're also gonna talk a little bit about China and in doing that in their satellites I just saw a story that involved frickin lasers and so we're gonna talk about that on Connect This! later this week But as many times as I can say frickin lasers as I can I'm gonna try to work that in so should be pretty packed First of all beads so I just came back from four weeks on the road and I came back to some

Sascha Meinrath | X-Lab (01:33)
Let's do that.

Christopher Mitchell (01:41)
some rumors and a week in which supposedly this is the week, this is the week that the stuff is gonna happen. What's going on Sascha? You're tracking it more closely than I've been lately.

Sascha Meinrath | X-Lab (01:51)
Yeah, and tracking is of course complicated by the fact that we have left kind of

regular time here. We have a government agency that seems pretty blatantly is not following the Administrative Procedures Act, which is a long way of saying there's a secret other negotiation that's happening outside of public purview and accountability, whereby this government agency is telling states, number one, the money that you are guaranteed by law we're not going to give you, and number two, even the lesser amount that we had

said you could ask for, also not going to give you. So the final amount of funding that will be granted to states in violation of both statute of the IIJA, but also of administrative procedures is still unknowable, certainly is going to be further diminished and will result in more Americans getting worse service options, which is to say,

we're gonna have to overbuild all this, which is to say, we're then gonna have to pay a second time to bridge the digital divide. This will be money squandered in large part due to this myopic short term and pretty foolhardy attempt at saving money in the here and now, but costing us a lot more later.

Christopher Mitchell (03:12)
Great, but let's

be clear. This is not money that's intended to be saved. This is money that is intended to be transferred, right? I mean, I mean, it's possible. It is possible that they may ⁓ try to, that they may not just switch it over. They might give it back to the treasury. But first of all, Congress would have to act to do that legally. But if we just pause for a second, I do want to make sure people know.

Sascha Meinrath | X-Lab (03:19)
Well, that... Yes. ⁓

Correct.

Christopher Mitchell (03:32)
Sascha and I are not like institutional people who will tell you the Administrative Procedures Act is like the way to do everything, right? I have my doubts that that is the best way to organize things. However, the Administrative Procedures Act was created in part...

to deal with a situation where you would have a President that would come in and abuse his position by giving spoils to those who supported them for people to make important government decisions absent any kind of public process or public input. And I think we could reform that process, but this is so far in the other direction of just saying we do what we want, law be damned. It is worth noting. So it's worth, I'm glad you started off there.

Sascha Meinrath | X-Lab (04:06)
Yes.

Yeah, well, the APA is meant to prevent nepotism, waste fraud abuse, what we ordinarily just call corruption. And is...

clearly being violated and let's put just names behind this. It's being violated in order to repurpose federal grant funding from the purview of states and to grant that directly and explicitly to satellite providers and in particular to SpaceX.

Christopher Mitchell (04:44)
And we are aware, NTIA was exempted from certain things such as FOIA, but you're saying that they are required, that they are not adhering to other laws that they are still intended to follow.

Sascha Meinrath | X-Lab (04:55)
Correct. So once NTIA puts out, a notice of funding availability, they have to abide by it. So they can, within the purview of the agency, they can determine what that notice looks like. They can't then just have a secret other process of divvying up funding. No, it's like what you have publicly released.

you have to abide by, and they are not even doing that, that there is this secret negotiation happening behind the scenes predicated upon number one, bad data, but also secret data in terms of the maximum amount that they will grant to serve, in essence, to build out infrastructure. So it's bad data, it's bad process, it's illegal processes, and right now it's unfolding. Like this will be happening this next week.

he pointed out.

Christopher Mitchell (05:44)
There is no intention. This project at this point, the BEAD program, which was supposed to be about an investment program for rural America that would also significantly improve Internet access across all of the rural areas that it was targeted against. This program has been hijacked and turned into a program of rural uninvestment.

de-investment. ⁓ don't know, can't think of anything else to call it, but it's literally pulling back billions of dollars that was intended to be creating jobs in rural America that would also build infrastructure that could be used for decades to lead to a more productive, more informed and better economy. You know, and that's all just being pulled back. And it's not like they're saying, hey, we've got freaking lasers now from Project Taara, which is something I'm learning more about Taara from Google, you and I

Sascha Meinrath | X-Lab (06:05)
Mm-hmm.

Christopher Mitchell (06:33)
both know someone who's been involved in that from the beginning. But we're gonna be talking about this week. That's one of the frickin' lasers that came up. They're not like, found a new technology to save money. No, they're just saying, we're going to save money. And it does not matter what the costs are among the families, the millions of families who will have to bear higher costs and much slower speeds of Internet access and much less reliable access and possibly usage caps and any number of other things. There's no weighing of the...

Sascha Meinrath | X-Lab (06:35)
Yeah.

Hahaha

Christopher Mitchell (07:02)
pros and cons. There is only this idea that there is money there that we can grab and we are going to grab that money and we will justify it in any way. And in particular, Arielle Roth and Secretary Lutnick are the ones who are running this to say we're going to grab the money. You know, I don't want to be very clear because I don't want to be like, kind of like, oh, like some people are doing things like they are trying to grab the money is what is happening.

Sascha Meinrath | X-Lab (07:16)
with it.

Correct. And, you know, it doesn't take a genius to realize that if the Infrastructure Investment and Jobs Act, as its name would imply, is about building infrastructure and creating jobs, I have a hard time figuring out how satellite does that in a local community, number one. But number two,

Christopher Mitchell (07:37)
you

No, was, I

think, Justice Scalia was famous for saying, the only thing that Congress ever does is hide elephants in mouse holes. So this is not surprising. This is...

Sascha Meinrath | X-Lab (07:50)
That's

right. So, but the other component of this in terms of like this quest to save money, nobody's actually done the analysis. Like, so an idiot would look at this and be like, hey, I didn't spend any money here. I saved five bucks. And when somebody points out like, but you're spending $10 over here because you took five money, $5 from here, like that's a net negative.

And I think when you actually do the econometric analysis, which NTIA has refused to do, you would find that this is a false economy. They're not actually saving money. For every dollar they save from this program, we're going to have to outlay more because in essence, we don't have core infrastructure and all of these attendant costs will then...

Christopher Mitchell (08:35)
The Veterans Affairs

folks are going to have to pay more because the veterans who are living in deeply rural areas will not be able to access telehealth. Not only that, the thing is that it drives me crazy is that we can talk just about the money, but we also know the money understates the impact. Because what happens is those veterans in the absence of health care have much more impoverished lives. They have sacrificed for the country and we're supposed to like, you know, come through for them. And instead we're saying, you know what?

Sascha Meinrath | X-Lab (08:43)
Correct.

Christopher Mitchell (09:02)
We think you should have your rich uncle get you the satellite service and we'll pretend that it's all good. if you can't do that and your mental health declines, well, I guess you're a sucker because we're figuring out how to call that money back so Secretary Lutnick can have a slush fund or so the Treasury can do something else that Congress hasn't actually voted on.

Sascha Meinrath | X-Lab (09:21)
Correct. And all of this is to say it's foolhardy and illegal and a false economy and corruption.

And it's all unfolding now in November 2025. And it's not hard to see this, right? They're issuing these giant national press releases, publicly announcing that they're doing this stuff. It's a very bleak situation, but it's very damaging one. The communities that are now forced to live even longer on the wrong side of this digital divide, this is existential threat to

their near-term futures. that, yeah, the oddity of this targeting especially deep red Trumpian territory is not lost on me. That also doesn't kind of alleviate the real pain points that these decisions are causing across rural America.

Christopher Mitchell (10:13)
Yeah, and we're not gonna belabor this, but I'll just say that for people who aren't familiar, you and I aren't just anti-Trump hacks. We are hacks that call it like we see it and spend a lot of time criticizing the Biden administration for not getting this stuff right. But I have to say that like, I mean, when I look back at the things that they got wrong, was like they were a little bit wrong or a middle amount wrong. And this level of wrongness is truly superlative. You know, I have...

Sascha Meinrath | X-Lab (10:20)
Yeah.

As we did, yeah.

I mean, yes, but

I will also point out like IIJA, it was money allocated in 2021. And the only reason why we're having this problem in 2025 is that the Biden administration never got the money out the door. That's maybe the one truthful thing that Ted Cruz Trump said about the failures of this program.

Christopher Mitchell (11:02)
You

and I have argued about this in the past as to like how reasonable it is that they should have got the money out earlier. it sure would have been nice if they had, although I do kind of wonder like if we would just see, I mean, this is the argument there that you and I have had, which, you know, people could go back. We're not going to rehash it now. But it also is sort of predicated on the idea that like the Trump administration would respect contracts if they had already been signed. And I'm not saying that I would bet my mortgage on that.

Sascha Meinrath | X-Lab (11:30)
Correct, well, and as

NDIA is suing over right now, and as we had to sue over multiple times for the Open Technology Fund, like, no, they don't respect contracts. And I'm like, what's more like, you know, kind of capitalistic than like respecting the contract? They don't even do that.

Christopher Mitchell (11:35)
the National Digital Inclusion Alliance, NDIA Right. ⁓

Well, this is, there are these times where people forget.

that this term capitalist drives me crazy because it does presume that contracts are honored, that there's a police force and other things that the state has to do. And people sometimes act like it can happen in a vacuum and sometimes they have to be reminded why we have laws against corruption and more importantly, why we usually have norms that try to protect against corruption. And when we see superlatively corrupt people, we try not to put them in positions of significant power, even if we don't like the alternative choices.

So I wanted to talk to you though about measuring Internet access because I don't know it's 2025 and I can't believe that we still don't have good measures of Internet access. Like when I...

When I write anything and I'm trying to figure out how to get in the mind of a person that's likely to be reading it, whether that is like a mayor, a city council person, a person supporting city council, or even just a sort of well-informed person that might be interested in this stuff, I'm always trying to think of how to describe Internet access because like,

It could be okay. And then, you know, like in my neighborhood on Comcast, when I had Comcast, it was pretty good. And then like, you know, 10 blocks over, it was not good because the neighborhood literally matters for like the wiring. There's so many points of failure and everything. And we don't even have like a language. No one takes this seriously to actually measure it. And I know that you spent a lot of time doing that. Ookla just came out with something.

Sascha Meinrath | X-Lab (13:13)
only 20

years

Christopher Mitchell (13:15)
Yes. You

don't have as many gray hairs as I do, but you've been at it for longer than me. But Ookla just came out with an announcement last week. And it's one of the things that I thought was interesting, because I was like, I'm really glad Ookla is doing this thing that Sascha pioneered. And we should have had it available for many years, even if Ookla is just going to try to monopolize it. Not monopolize it, but they're going to try and take advantage of it then make it proprietary and never share it with anyone. ⁓

Sascha Meinrath | X-Lab (13:28)
Hahaha

Correct. And

I'm glad that they're doing it too, even though they basically stole their methodologies from folks that came before them. Like, it's better to have something than nothing. And amidst the kingdom of the blind, they're the one-eyed man.

Christopher Mitchell (13:52)
So we talked about it, I talked about it with Doug last week on this show. And the thing that I find the most interesting is the continuous pulse is what they're calling it, which is this idea that you have a little device on your network that is basically every 15 seconds being like, am I online? Am I online? Am I online?

and then it tracks when it's not online. And ideally, it would go even further, which I think you've been pursuing, which is to say, if I'm not online, how far can I get to being online? how is this like, am I disconnected from the router or is this something that's multiple hops away in the ISP's network? Where can I get, you know?

Sascha Meinrath | X-Lab (14:26)
Yes.

Yeah. And these are tried and true.

network diagnostic tools, which is to say, when we created Measurement Lab, whenever that was, in 2009, so 15 years ago, we were ourselves taking advantage of the tools that our precursors had built. And these sorts of diagnostics are wrapped up in what we call the network diagnostic tool, which became the measurement lab speed tests. This is what now 15 years later, Ookla is emulating. And I'm thankful

they are, it's also important to know that because these tools were originally built by and for the research and and the network administrative community, open source freely available variants of the exact same function are widely available tested to IETF specs and like you know the whole shebang. Which is to say, Ookla has a much louder and more widespread PR like

capabilities in the research and scientific community. There is of course nothing new to what they're doing, but it is good that they're making people aware of, there are ways to test where the failure is, who's responsible, as there have been for decades in this space.

It's also really interesting to me that, you know, when you read the headline of OOKLA, and I think the one you were referring to is one where it's saying like, you know, wireless speeds are improving as is Starlink, blah, blah, blah. And then you read through the thing and then like referencing, oh, they did this thing for Starlink, blah, blah. And then you read into that and you find out right now, 17 % of Starlink customers are actually getting broadband.

Christopher Mitchell (16:08)
was so let's be clear first of all you're doing something most people don't do right so like for every like person that sees the headline I think less than 1 % of people click through it and then for everyone who reads that story less than 1 % of them significantly less than 1 % goes on to check it out so you're in a rarefied community ⁓ but

Sascha Meinrath | X-Lab (16:17)
Yes.

Yeah.

Christopher Mitchell (16:27)
But one of the things I wanted to note with that is as before we go into what you're the point you're raising is worth measuring. Ookla goal is to measure what is the maximum capacity of your Internet access. And so it tested over a period of what like 20 or 30 seconds, right? And so like the speed of your connection is determined by the maximum instantaneous throughput at any one time, which is another way of saying that like, this is not the speed that you're experiencing. This is the speed that you almost never

Sascha Meinrath | X-Lab (16:42)
Yeah.

Christopher Mitchell (16:56)
experience.

Sascha Meinrath | X-Lab (16:57)
yes, that's right, but I would just correct. It's not actually Internet access It's the maximal idealized speed between you and your ISP Which is to say it It's basically measuring like what's the first step on your way to the Internet?

Christopher Mitchell (17:06)
Well, that depends on where the test server is. Right.

Sascha Meinrath | X-Lab (17:14)
and then nothing further. And what we've seen consistently again and again and again is congestion points. The places where you actually experience slowdowns to the Internet are those gateways from your local Internet service provider to the remaining 99.999 % of the actual Internet. And Ookla doesn't measure that.

Christopher Mitchell (17:38)
It doesn't in two different ways. Like let's just dig into this and nerd out for a second, right? So first of all, you could pick another speed test server, right? You could choose that. Now, as Doug has mentioned, Doug Dawson in our show, I think last week, but also has mentioned many times, they know this. And so they set their routing tables and all of their technology to make sure that you're gonna have the best possible experience when you are doing that test. So they can game that.

Sascha Meinrath | X-Lab (17:43)
Sure, let's do that.

Yes.

guess.

Christopher Mitchell (18:07)
There's a different experience. if you are on an ISP like Mitchell's ISP, some small, barely holding it together, poorly managed network that I'm to some extent, it would make sense if I'm then having my connection to the Internet run through level three. Should I really get penalized if level three is not great at that throughput? Well, possibly not. I'd want to know what is

ISP's responsibility and what is not but also the question is then if I'm deliberately refusing to size my backhaul to the most of the content on the Internet if I'm sizing that too small in order to like make a larger profit well then yes I should get blamed and so like people people want to like kind of say ⁓ like I heard Sascha and Chris talking about this thing like and this like I'm not really measuring this I want to give people a sense of the nuance that like there are different reasons why you'd want to do that and that

different speed tests may actually be measuring different things. If we're going to talk about Netflix in a minute, Netflix's speed test wants to measure from the content delivery network to you, which is probably already inside the ISP or inside an Internet exchange that is connected via fiber controlled by the ISP to the network.

And so there's a lot of complication here, but fundamentally we're getting the question that I feel like we should be asking is the one that you and M-Lab try to answer, which is not, what is some technical like maximum throughput and more, what is your experience like? Right?

Sascha Meinrath | X-Lab (19:42)
Yes.

Well, often what we see, so let's imagine that you have a house fire, right? And I give you a fire hose to put out the house fire and you turn it on and it's just not working and you trace it back to the fire hydrant and turns out the last 50 feet to the fire hydrant is being fed through a cocktail straw, right? At that point, you're like, I get it. Like the cocktail straw is limiting my throughput to the fire hose. The fact that we don't test that.

is a problem. The fact that the advertised speed will be whatever the capacity of the fire hose is, the fact that many of the Ookla tests are measuring the girth of your fire hose and everyone's ignoring the fact that that's being fed by a cocktail straw. To me, like, that's a fundamental problem because what I experience on the end is the throughput of the cocktail straw.

Christopher Mitchell (20:31)
Right, right. The weakest part of the chain is the weakest link, not its best links.

Sascha Meinrath | X-Lab (20:35)
Correct.

Correct. And so I'm not saying that the fire hose doesn't matter. That is under ideal circumstances, the maximum throughput that I could push to put out this fire. But what actually matters in real life is what is the weakest link? What is the greatest congestion point? What is the limiter on the throughput?

And so you really do need a diagnostic tool that's measuring all the different facets. Now the irony here is the same ISPs that are like, no, we don't want to measure the cocktail straw, are also the ones that are constantly saying, but the real limiter is your wireless access point. They are aware that there are these different limitations to different parts of the network.

they only want you to blame your home network. And you have to imagine, I've had many ISPs and I've ended up in many discussions with my ISPs where I'm like, I'm not getting my throughput. And then they're arguing to me over like where the slowdown is. I'm like, it's in your core network. It's right here. I can give you the link. They're like, no, it's your home network. I'm like, it is not. But long story short, they're trying to blame the consumer. They're trying to convince us all that somehow we

screwed up and that the problem is on our side of the equation and sometimes, sometimes it is, especially if you're using say antiquated OG 802.11 equipment, then yeah maybe your wireless access point but for most of us most of the time the limiter is actually this congestion point inside or at the congestion point is at the peering point of your Internet service provider but they'll never admit that. They're gonna keep

I'm you, no, no, it's your home network that's the problem, even when it is not.

Christopher Mitchell (22:17)
Let's move on to talk about satellites, where the problem is almost certainly not your home network. The problem is that you live in possibly a very beautiful area and you're lucky to have electricity.

Sascha Meinrath | X-Lab (22:20)
Yep.

Sure.

That's right. Satellites works great

if you're not on earth but other than that.

Christopher Mitchell (22:35)
You're

living in an area that is quite remote and you have electricity. Thank yourself that Arielle Roth and Senator Lutnick weren't around in 1936 to be working on electricity.

Sascha Meinrath | X-Lab (22:45)
That's correct. Yeah.

Christopher Mitchell (22:46)
So you had done a study.

in which you had used what publicly available information there is, which is limited and constrained deliberately by SpaceX in the Starlink system to figure out what are the constraints. Where do we find that too much, one of the things you found, where do we find there are too many local people trying to use Starlink and Starlink is not able to meet that? And I'll just note, I think you'll agree, under their current spectrum constraints,

number will change over time with changing technology if they're able to get their hands on more spectrum. There's a lot of things that can change. But with the current technology that they're using, you found that there was a significant limitation and they responded that you were wrong and provided numbers that suggested that you were mostly right is my understanding quietly. then

Sascha Meinrath | X-Lab (23:36)
They

provided numbers that showed that we were actually being really generous. They wanted us to use Shannon's Law. They're like, you've never heard of Shannon's Law. And so we ran Shannon's Law through the equation. was 33 times worse for them when we did, which we knew, obviously. But we wanted to give them the benefit of every doubt. It was less a...

a research study and more of like, let us give you the equations from physics of like reality. And in essence, looking at how much throughput could actually be achieved under ideal circumstances, given the spectrum that's available, given the way that they have actually built their satellites.

and externalizing a lot of real world limiters like that would further deprecate that throughput. And the importance of

Christopher Mitchell (24:22)
No, let me

just for people who might be like struggling to sort of think how you could do this. We, this is not new, right? There are on the order of like 300 or 400,000 macro cells across the United States. And in our cities, we on the regular are overloading them. If you have a parade, if you go to a sporting event, a lot of times you will find that it cannot meet your need. This is the exact.

Sascha Meinrath | X-Lab (24:28)
No.

Yeah, if you've ever been

on a laptop in a crowded cafe, right? Like, yes, we've all experienced network congestion.

Christopher Mitchell (24:53)
Yeah, and then this is just different because it's IN SPACE!

Sascha Meinrath | X-Lab (24:57)
That's right.

And as such, and because the technologies are new to most people, there's very limited knowledge of its actual capabilities. And into that space, there's been a lot of mythology around what it's capable of. And we're like, look, let us provide you a better understanding of what's happening here. And to do that, we will use

Well, as it turns out, Starlink's own filings with the FCC. So we plowed through this and extracted from that information about what the actual capabilities of satellite connectivity are, and then showed people with math, in essence, like here is the theoretical maximal capabilities that these satellites can push.

And if I have X amount of capacity and I have Y number of users and the maximum we can get is like X over Y, it's not a very complicated analysis, just nobody had done it. So in us doing that, we wanted to spur better understanding, conversation, but also a warning.

that in fact with limited capacity what this means is you really only can provide satellite connectivity to a certain number of residents and still achieve broadband speeds.

Christopher Mitchell (26:10)
And we talked about that with regard to real world tests when there were natural disasters and all of the first responders show up with their Starlink kits and find that they overwhelmed the signal. So this has been empirically tested as well. So we've talked about that before, Sascha, what's the new news where you've had some other folks who are recently adding to that conversation?

Sascha Meinrath | X-Lab (26:30)
Yeah, so we come out with this analysis showing like, this network is already at its limit and going to fall down into the swamp very quickly with the addition of a large, a substantial new user base.

And over the last two weeks, in essence, there's been now a number of follow-up studies, both by independent research laboratories. There's a group out of Virginia Tech that's done a bunch of analyses that back up our claim. But perhaps more importantly, Netflix at the Internet Engineering Task Force meetup in Montreal, I guess it would be now like 10 days ago, came out with analysis of like their entire national data set.

And what they said is actually it's already screwed up. We're already able to document on nationwide.

that people are unable to get broadband today without the addition of all these new BEAD households on the existing infrastructure. We're already seeing the deprecation of speeds such that they right now are not getting broadband service, which again, that aligns with what Ookla is also saying, which is only 17 % of Starlink users today or two quarters ago are getting broadband speeds. So you have our analysis being like,

here's

the physics and the equations for that. You've got real-world tests from Ookla but also independently from Netflix. You've got backing that up, additional independent research labs, all pointing to the same problem, which is that this is not going to work as a broadband solution for rural America. And into that, you have NTIA being like, we don't care.

And to me, this is real problem. It is a bill of goods being sold to rural America. But it is also the case that IIJA, the federal legislation, requires NTIA to provide broadband service with these public funds. It is illegal for them not to provide that. And what all of these things are aligning is saying you cannot provide broadband service with this medium.

This begs the question of waste, fraud, and abuse.

Christopher Mitchell (28:36)
Right, when you say they have to provide broadband, it would not be fair if the federal government, this is not a new issue, the federal government has given out money before to many ISPs and said you have a performance requirement. Now, what I was gonna say is that it would not be fair to say to Starlink or to another ISP, ⁓

If you ever provide a connection that is slower than broadband, then you're gonna, we're gonna call the money back, right? So they had to develop a methodology. So the Federal Communications Commission for some time has been using basically the 80 % rule, right? Which is that like, you do speed tests regularly and 80 % of the speeds have to exceed 80 % of the target, right? Is that roughly it?

Sascha Meinrath | X-Lab (29:16)
Yeah, yes, and Starlink doesn't and has been shown consistently.

Christopher Mitchell (29:19)
Right. I don't think they'll do it now.

They're saying, they're saying, trust us, bro. We're going to get there. Right.

Sascha Meinrath | X-Lab (29:25)
We're going to get there.

That's right. Which I'm like, give all give give Meinrath Enterprises the billions of dollars under the claim that we're going to get there. And like, you know, at some point, I mean, Starlink isn't the worst, right? I still can't fathom giving money to Kuiper. Right.

Christopher Mitchell (29:41)
Yeah, Project Leo

now, Amazon Leo is a new name, but Project Kuiper which has a few satellites.

Sascha Meinrath | X-Lab (29:45)
it's been, yeah, because nobody knew how to pronounce

this thing, right? But long story short.

Christopher Mitchell (29:48)
Come on,

man, space nerd did, you knew how to pronounce it. But yeah, a lot of people didn't.

Sascha Meinrath | X-Lab (29:52)
That's right. But long story

short, like they're giving money for an infrastructure that simply doesn't exist. that to me is like, wait, so we're going to magically believe that this thing is going to appear in space and keeping in mind they're already years behind their own launch schedule. Correct. And all of that's before you really understand that.

Christopher Mitchell (30:10)
and they don't have any spectrum to speak of.

Sascha Meinrath | X-Lab (30:16)
the LEO space, the Low Earth Orbit Satellite Space is rapidly getting congestion itself with other providers, which means that all of those equations that we laid out of like, if Starlink had...

singular access to all of these bands. This is the maximum they can get. As those bands get congested with other satellite providers, obviously each one of them gets a slice, but nobody will get it all, which will further degrade overall capacity. None of that is taken into account by NTIA. And I would argue you have to willfully disregard not just the physics, reality,

of this, but willfully disregard independent analyses spanning multiple years across the entire country and say, you know what, all of that is wrong.

Christopher Mitchell (31:09)
will say this, I was more concerned when I thought we were talking about many millions of households. But at this point, we're talking about they've kicked so many people who are eligible for.

investment off of the program that we were talking about like what one to two million households potentially being on Starlink maybe a little bit more like a lot of them are choosing not to go on it and so i i'm practically speaking i don't think we're ever going to be able to test the proposition that they can't do it

Sascha Meinrath | X-Lab (31:23)
Yeah.

Christopher Mitchell (31:39)
but they are gonna get billions of dollars anyway. And so I agree with your point, which is that NTIA is being grossly irresponsible with trying to run this experiment without taking into account the preferences of local elected leaders or the local people themselves. ⁓ But I don't think we're ever actually gonna even know if you're right, because most people that are eligible are not gonna be signing up for it. They've already chosen not to use it.

Sascha Meinrath | X-Lab (31:57)
Correct.

Correct. Well, and that's due to a lot of reasons. The problems of getting connectivity, getting reliable connectivity being a big problem with satellites. As long as you don't have trees, more rains, hills, mountains, or say weather, it's going to be great for you. If any of those apply,

Christopher Mitchell (32:15)
the affordability is a big issue.

Sascha Meinrath | X-Lab (32:26)
I'm not saying it's going to be terrible. And I want to be very clear here. Satellite has a niche role to play that is wonderful and should not be thrown out with the rest of this. It's just, yeah, it's just not a functional equivalent for better solutions that end up also costing users less. So satellite is a less reliable, less robust system that costs more. And NTIA is like, that's the answer.

Christopher Mitchell (32:35)
Yes, yeah. I was trying to go there actually.

Sascha Meinrath | X-Lab (32:53)
and is being applied so widely and without regard to the physical limits of the medium and the physical limits of the architecting of that medium.

then it's gonna fail. It's gonna be a bad solution that's gonna cost us, the taxpayers, more over time. none of this achieves its stated purpose to begin with, even if I bought into it and disregarded physics and bought into the whole premise of this is the cost-effective, it's not, even that.

Christopher Mitchell (33:23)
Now you mentioned the bands, the paths across the orbits are gonna be congested with more entities in there. Project Kuiper, I think has a much smaller constellation, but might there be another constellation, Sascha, that is being built that may have more ⁓ of the satellites in it?

Sascha Meinrath | X-Lab (33:37)
Hahaha

Yeah.

Mr. Mitchell asks leadingly, yes, indeed. ⁓ So, you know, when you think about the global scale, right, US is the second largest economy on the planet and number one is China.

Christopher Mitchell (33:54)
No, I think you're wrong. think US is still the number one.

Sascha Meinrath | X-Lab (33:57)
No, no.

Or are you speaking as the Trump administration there? I can't quite do it.

Christopher Mitchell (34:00)
No, no, no,

it was expected that China was going to surpass us, but I don't think they did. I'm going to look it up quick while you talk.

Sascha Meinrath | X-Lab (34:05)
Okay, so long story short, one and two, one way or the other, kissing cousins economically speaking and...

This will surprise no one that's been paying attention to all the other functional equivalents that China has spun up in recent years, but satellite is no different. So there is a system called GuWang, which China is deploying rapidly that will also be in play in this confined space of satellite uplink and downlink bands sharing.

that space because I want to believe that people can play together well in the sandbox of space communications that will over time and I don't mean long periods of time I mean like over the next three to five years greatly congest the available space for Starlink, Leo, etc. Which is to say this is going to be a shared medium and nobody's paying attention to those

capacity limits which will rapidly diminish what you can do with Starlink

Christopher Mitchell (35:05)
I'm gonna challenge you on that in an entertaining way in a second. The appropriate for this show, it depends on how you measure it, China versus the US. So, gross domestic product, the United States still is superior. China's economy hasn't grown in recent years the way it was expected to. But in other measures such as purchasing power parity, then China's is larger. So it depends on how you measure it. They're both quite large. Regarding...

Sascha Meinrath | X-Lab (35:12)
Hahaha!

It's good that we are both correct

and both wrong.

Christopher Mitchell (35:31)
The regarding congestion up there, I mean, I feel like it is worth noting that there are more cars in New York City than there are satellites over space. There's a lot of space up there and most of the cars in New York City, you know, don't hit each other. But I'll also say that most of them are not moving at 17,000 miles an hour. So, you know, there's some differences.

Sascha Meinrath | X-Lab (35:51)
Yeah, well,

that's right. Well, and everyone should look up Kessler syndrome because like to me, like that is the terrifying cascading failure of satellites and it satellites that then hit more satellites.

I think there's a Star Wars reference from there too, but like long story short, Kessler syndrome is becoming more and more probable as you get more and more devices up there. And if you look at the projections, for example, how often SpaceX satellites had to be moved to avoid, near misses, it, it exceeds the projections by, think almost an order of magnitude, which is to say things, even though it's not

very congested, things are getting more more congested. And it's a little bit terrifying to think about what might happen if you get too many devices, not enough accountability, and a collision in these low Earth orbit shells.

Christopher Mitchell (36:49)
In policy terms, we sometimes talk about this as ⁓ low probability, high impact, which is a ⁓ really dull way of saying, you know what, we think this is probably not going to happen, but if it happens.

Sascha Meinrath | X-Lab (36:54)
Yes.

Christopher Mitchell (37:00)
There's no going back. You like there's just, you sort of enjoy those last few years of us having the GPS satellites up there because they're probably above the big mixture of stuff that we can't get through. But there's no putting anything new up there. And so whenever those ones that are up there come down, come down. This is.

catastrophic and I feel like we are just waltzing in there. In fact, last week a piece of tiny space debris, I believe, is believed to have hit the Chinese, one of the Chinese, I don't if it's their space station, but they were unable to return some of their astronauts to Earth because there was space debris that was believed to have been set up a fresh craft that they knew was not compromised. And so like this is very, very serious and I would say that I would feel a lot better about it if the federal communication

Sascha Meinrath | X-Lab (37:21)
Yes.

correct. ⁓

Yeah. Well, again, this points to like you really do want kind of a minimal number of things whizzing around at 17,500 miles per hour.

Christopher Mitchell (38:20)
the last 500 that getcha.

Sascha Meinrath | X-Lab (38:22)
That's right. And you could greatly lessen the sheer scale number of devices if you had common carriage in space, for example, which is to say if things interoperated and you allowed people to sort of borrow capacity from each other, you wouldn't have all these redundant systems. And these redundant systems that may or may not be centrally, they won't be centrally coordinated, all running independently, just greatly increase the probability of a catastrophic event.

In retrospect, we will look back and be like, yeah, we shouldn't have done it this way. The problem is that that is a one-way transformation, which is to say, if we do end up with something like Kessler syndrome wiping out low earth orbit for a few years, decades or beyond, like the damage is already done. And it's a...

Christopher Mitchell (39:08)
Yeah, the damage is

indescribable. Like the amount of tech that depends on these things not being destroyed is, you literally could not make a list because it is so long. I mean, the role of GPS in things, and I don't know if GPS, like how long the system can go without being ⁓ replaced periodically. But you know, people have no idea how much space technology controls everything that we do.

Sascha Meinrath | X-Lab (39:33)
That's right. Yeah, think ⁓ Neal Stephenson wrote a Seveneves talking about when the moon explodes and basically everything becomes untenable because of just the sheer magnitude of the number of things falling out of the sky. But it's like maybe not that catastrophic, obviously, but the...

the inability to get things into space safely because we have basically this giant kind of shell of debris in low Earth orbit. That's a very real problem that affects everything above it. So even if our GPS satellites are above the fray.

Christopher Mitchell (40:06)
they

are.

Sascha Meinrath | X-Lab (40:06)
they are dead men walking, which is to say we do have to replace those every once in a while and like we will not be able to do that safely were something like Kessler syndrome to be realized. It's very bad.

Christopher Mitchell (40:10)
Yeah.

Well, not just that. mean, there's so many other things to talk about that we're going to skip over, such as the fact that these, we don't know the environmental costs of doing the launches that we are doing today, let alone increasing that by a factor of 10 or 20 times. And I'm curious to see where that shakes out. But.

When you said about that, it's funny, you and I talked this morning about this Chinese satellite program to do the low Earth orbit. At the same time, I saw something and I wanted to just talk briefly about it because people might be reading this stuff. this was a piece that was, I think, fairly poorly written. It was in digitalgalaxy.com. Chinese satellite crushes Starlink with two watt laser fired from

Sascha Meinrath | X-Lab (40:57)
Okay.

Christopher Mitchell (41:02)
36,000 kilometers in space. So the Chinese did this really cool demonstration. I mean, so like already like the lasers in space are off the hook, right? Like Sascha, any respectable science fiction book has communications done with lasers from space to space. They're hard to, they're hard to like detect. They're hard to intercept. Right? Like you're shaking your head a little bit. I don't know if you disagree. Like I'm none of this magic sci-fi stuff.

Sascha Meinrath | X-Lab (41:04)
Space lasers.

Yes.

No, no, it's just the space.

They are harder than say like using like an omnidirectional antenna. Yes. But even like a one degree cone multiplied by like a million miles is a big, like it's pretty big on the other end of that. like, yes, it's harder to detect than the others, the alternatives. But, you know, it's not like a point beam. We think of the laser pointer as sort of minuscule, but if you actually were to map out that cone,

Christopher Mitchell (41:40)
Yes.

Sascha Meinrath | X-Lab (41:52)
It's a big cone as a 2000.

Christopher Mitchell (41:54)
Right. Well,

nonetheless, SpaceX has done amazing things with its laser, laser satellites and what's coming next is remarkable. And so I just want to point this out because they were like,

Sascha Meinrath | X-Lab (41:57)
Yes!

Christopher Mitchell (42:04)
It was like, they're basically saying like, ⁓ like, like this Chinese satellite company is going to crush Starlink. And it just, I this like everything that's old is new again, because like this idea of like, this is great. We don't have to put up so many satellites. And I'm like, you're 36,000 kilometers away in geostationary orbit. It takes 120 milliseconds at the speed of light to cover that distance. You got to do it four times to have a there again and back again, you know, connection. So you're at

Sascha Meinrath | X-Lab (42:30)
Yeah.

Christopher Mitchell (42:33)
point at a ping that is ⁓ approaching dial-up again. When people see some of this stuff about the new technology, I would just say take it with a grain of salt. It is amazing to do a ⁓ gigabit per second sustained transaction across 36,000 kilometers, but it is not going to replace SpaceX and Starlink. It's not what we're doing.

Sascha Meinrath | X-Lab (42:37)
Hahaha

Yeah. You know,

as I tell people just to drive this home, the number of satellite based e-sports champions remain steady at zero. And the reason is like latency and latency at that level is a byproduct of physics and that whole, you see like you can't go faster than light. And there's no, there's no routing around that.

Christopher Mitchell (43:21)
Yeah, well...

Sascha Meinrath | X-Lab (43:26)
physical limitation of our reality. And so when people pretend like that doesn't exist, and especially like it is majestic, like some of the technologies that SpaceX has developed as a technology is like, that is a beautiful piece of technology. I appreciate that. And pretending like, say, physical limits of reality don't exist. I'm like, that's a bad way.

to try to sell a beautiful technology because the end result is like you're reducing this beautiful technology to vaporware. It's simply not possible to deliver what is being promised. And then you end up with the inevitable backlash where people like this technology doesn't work. I'm like, whoa, it works beautifully for this thing. Right. You want to send text messages, having a satellite up like

Fine, like the extra tenths of a second delay don't matter in that media. But you want to do like real time stuff. You want to have your intelligent transportation system where all your cars are avoiding collision an extra tenth of a second late. That's not good.

Christopher Mitchell (44:30)
Okay, let's talk

about this because this is the last thing we'll talk about. I was going to skip it, but I just can't. I don't think any autonomous vehicles...

that actually work will be dependent on low latency networks. I just think that these devices have to be using onboard technology. For the same reason, gonna just like, just go back for a second. Like I was listening and reminded on the Vergecast last week, just the stupidity and the sheer like audacity of like the wireless companies, AT&T, Verizon, T-Mobile, Sprint at the time, to be claiming that 5G was gonna change everything when they knew it was gonna change absolutely nothing except for the number of

handsets that they were selling. Like just ridiculous. The idea that like, we're gonna have autonomous vehicles. No, we're not. No one's gonna insure an autonomous vehicle that crashes when it gets a hiccup on the network. That's not something that happens. And insurance matters in these industries, right? Like insurance runs the world in some ways. So I just, continue to believe low latency networks will be great, but mission critical stuff will always have to be done on board or it's not gonna happen.

Sascha Meinrath | X-Lab (45:34)
Yeah, and that's largely true, but you know, as an American driver on American roads, what I experienced time and time again is I come up to a construction site and they've rejiggered like the roadways in some sort of Escher-esque configuration where I'm just trying to figure out like where I'm supposed to go. And as a human, I'm like, okay, I think it's here. There will inevitably be these moments where a car's like, I need to phone.

like somebody else to figure out what's happening here and that you want this fail over.

Christopher Mitchell (46:04)
That's what Waymo does.

Waymo does that, right? They make the decisions on board and in like truly edge cases, there's a telepresence ⁓ solution, right?

Sascha Meinrath | X-Lab (46:14)
Yeah, or just a, like, I want to query what other cars around me have done in this unknowable situation, right? And that requires networking, which is to say this won't happen often, but it will happen. And those tenths of a second do matter in those cases. And you really do want the lowest latency possible for these kinds of mission critical communications. And...

Christopher Mitchell (46:37)
And that doesn't

involve going to the moon and back.

Sascha Meinrath | X-Lab (46:39)
And it doesn't, yes. Or waiting in some buffer bloated queue for like an extra few tenths of a second. Yeah, I all of this. And again, these use cases are very real.

they're very important to solve for. When you sell vaporware, you don't address the fundamental technical problems that play in these infrastructures. And you unfortunately have decision makers that don't understand the technology, and that leads to more crashes. This leads to death. This leads to very real problems. Which is of course why, like, if there's a...

a take home message from our conversation today. It's like believe in science and have experts in the mix when we're making these critical decisions. No technologist would look at the numbers and the capacity limits in the architecture of SpaceX and make the political decision that NTIA is making to the detriment of rural Americans all across the country.

Christopher Mitchell (47:41)
That's a great way to end. Super depressing, but accurate. Sascha, once again, thank you for a fun conversation.

Sascha Meinrath | X-Lab (47:47)
My pleasure.