April 19, 2024

The mountain of shit theory

Uriel Fanelli's blog in English

Kein Pfusch

Take “ciao ciao” to the telco.

You do

I started to get interested in the "starlink" project from the very first moment the numbers and underlying technologies were presented. It is a constellation of new generation satellites (extremely small and light satellites, but equally capable), in the order of thousands of satellites (not tens as it has been up to now) capable of covering the whole planet.

Over time, that is, satellite technology has improved a lot. Once a telecommunication satellite (for example, the classic IRIDIUM constellation) were beasts several meters long, several tons heavy, which therefore had enormous times, costs and risks for putting them into orbit. Thus the classical constellations often cover only part of the sky. But never everything. Furthermore, due to the enormous distance between them, those at low altitudes often needed stations of repeaters on the ground, spread all over the world, further aggravating the costs.

The new satellites, weighing a few dozen kilos and having dimensions in the order of half a meter, are much cheaper and less risky to put into orbit. So it's possible to think of mesh networks, without ground repeater stations, made up of thousands of small satellites.

The "starlink" project is a project based on this concept: covering the whole world with a network (satellite? Wifi? 5G? 6G? Nobody knows) able to connect every device.

To do this you need thousands of satellites. For this SpaceX had already requested the authorizations to put 13,000 satellites in the sky, and now it has raised the number asking for another 30,000. A sign that has demands, and needs to devise a network with greater capacity.

In short, the industry likes it.

You do
https://www.technologyreview.com/s/614566/spacex-just-filed-a-request-to-run-30-000-more-starlink-satellites-in-orbit/

The thing that puzzles me is "the reasons are unclear", because to me the reasons seem almost obvious.

Those who have worked in the connected automobile world know very well that the ideas were many and all beautiful. But when the houses went to accomplish this thing, they clashed with reality:

  • a jungle of rates.
  • bad coverings, especially outside the city.
  • different standards with different penetrations depending on the country (3G, 4G, 5G …)
  • the need to cover the roaming problem.

trying to overcome the problem they resorted to so-called "global providers", that is to companies the cul work is explicitly to keep the contractual complexity necessary to offer connection at a "flat" or at least "predictable" cost on the whole globe. In practice, brokers of telephone contracts.

This eased the contractual complexity, but did not solve technical problems.

The problem is common to the entire IoT automotive world, to global commercial logistics and to the world of aeronautics and tourism.

Added to this is the frustration of OTTs, such as Apple, Samsung, and others. They must from time to time negotiate the release of their phone with telcos, in the sense that they need local SIMs (the e-sim will not solve the problem because it appears while roaming on the VLR) and must come to terms with the traffic rates, which change from country to country.

Moreover, even Facebook, Google & co have a problem: by listening to their traffic, the telcos manage to collect almost all the data that they also collect. In practice, they do not have a data monopoly because by combining BNG telemetry, DNS data, CDR records, HLR and MSC records, they have practically the same level of espionage as the user who has them, while Facebook , Google & co need the monopoly. Furthermore, the use of their services is impacted by local telephone rates.

But now Starlink is coming.

Imagine that Apple makes a deal with Musk, and gets a phone that does not need a contract. Buy an Apple, and you're on the net. You don't have to go to TIM or others. You are on the net because you have a telephone. The same is true for industries like the car industry: connect all your cars to this network, and you have connected cars around the world. You no longer have to ask yourself where they are before understanding what contract / connection they will have.

The same for Facebook, Google & co. In this case it is a matter of making an agreement with Starlink, and from that moment Android connects everywhere, with anyone, and advertising flows, and no government can do anything about it. And the same with Facebook: once anyone can use Facebook just because it has the Facebook client, there is very little that a local government can do (it should jam signal all over the territory, which is unthinkable) .

Obviously, every technology has its limits, but this is even more worrying:

  • These are outdoor technologies. Indoor, that is inside the buildings, would have problems. It would therefore be necessary to sell satellite routers / repeaters to WIFI.
  • Defining a global tariff would be complicated, since each country has its own cost of living, to which telcos normally adjust prices.

The problem is that passing a product with non-perfect qualities compared to the present is only possible in a way that is already known to the public.

Offer it for free.

It means that if you buy a mobile phone that has a contract with Starlink, you will pay nothing but the phone. Then Apple will take care of the costs, with all the ecosystem that can sell you without limits, from music to anything.

It means that in the Android license the possibility to connect to Starlink will probably be sold, so that Google has any possibility to send you advertisements, collect data, position, etc.

Obviously such a plan cannot be publicized too much, because to announce to the world that you are working to cancel the entire telco industry, or a good part of it, produces an earthquake in any stock exchange. Furthermore, telcos are normally strong lobbies, which would be screaming by governments asking to block this thing. They would have the usual motivations on their side:

  • Lawful Interception: governments would lose the ability to intercept traffic.
  • Integration with emergency services: emergency calls of all kinds, police, fire brigade and so on.
  • Inability to influence in the event of a blackout.

But these are relatively surmountable problems. It is certainly possible to bring the link to work with emergency numbers, and a government agreement on legal interceptions is feasible (provided you are friends of the USA).

If instead Starlink wanted to get serious, it would have to do nothing but make agreements with Apple, Samsung & co in silence, and suddenly mobile phones would appear on the market that "have no tariff". And at that point governments should explain to their citizens that they can't have the new Apple phone. Also because, how could they stop the smuggling, since they would work anyway?

From the point of view of the US government, this is very convenient: it means being able to remove the network from a nation, with all the disasters of the case, if there was a war.

So I see a similar project supported by:

  • Big telephony producers, who hope to no longer depend on telcos.
  • Big OTTs, who no longer want to charge their traffic.
  • Automotive / logistics industry, which wants global mobile devices released by telcos.
  • US government, which wants global power.

Since everyone sits on a pile of money, there are probably no budget problems. You can consider it infinite, for convenience.

To do such a thing, as with any IT project, you need three things: budgets, experts and technology. Right now there is the budget, there are experts and there is technology. There are no particular industrial obstacles.

Are there things that foreign governments can do? The same thing that was done with GPS: since the US could silence it at will in a given area, (by encrypting the payolad so that only the military would use it), GLONASS (Russia), Galileo (EU), BeiDou (China ) etc. The only possible reaction would be to create a certain number of networks similar to Starlink, and impose its use on a given territory, or allow multi-network devices.

But one thing is certain: if I ran a telco and saw someone put 13,000 low-altitude satellites into orbit, and planned 30,000 of them, I would have some problems.

You do

links

Leave a Reply

Your email address will not be published. Required fields are marked *