I started this research wishing to dive deeper into Starlink and to understand further the implications of an internet almost exclusively run by one company. I uncovered so much additional context along the way that I thought to bring more attention to the subject.
It is not obvious to most people (it wasn't to me) that internet satellites are a supremely better way to scale the internet for the next 50 years. To prepare us for the future of increasingly needy bandwidth. Much like Bitcoin miners are allowing energy infrastructure companies to over-build their capacity, I acknowledge that we have reached the point that it is necessary to increase our future potential bandwidth. As well as upgrade the internet in ways that fibre and our fixed cable infrastructure cannot match.
Why focus on one company?
SpaceX, and only SpaceX, is in a position to rapidly build out an enormous literal internet mesh around the globe, because only SpaceX had the vision to spend a decade struggling to break the government-military monopoly on space launch. That does not mean that we should need not be wary of this fact though. Quite the opposite.
SpaceX and specifically Starlink should be attracting more scrutiny that any other company or product in the world right now. More so than Tesla, OpenAI, Nvidia, Apple, Google, Tencent, Huawei or any other over-valued tech giant. And I do not say this because of Space exploration, which is completely overplayed at this point. Elon is great at managing the narrative and this is a perfect example. SpaceX should be under our scrutiny because with their Starlink product, they will in all likelihood find themselves in a monopolistic position the likes of which we have not witnessed before. If successful, they may put the likes of Kodak, Intel, Google & others before them to great shame.
The Starlink team have been ruthless and at times disingenuous to get to where they are today. But they do have an incredible product offering. The company has a formidable future ahead, with the competition struggling for the same levels of traction, and that is without even considering the future likely merger between X (payments, social, AI), SpaceX (communications) and Tesla (automation & energy). It is therefore time to dive a little deeper into assessing what a reimagined future internet may look like.
Note: Some of this work has been paraphrased or lifted from research that I reference throughout. The best research piece I found was this blog post from 2019 (, which goes even deeper into areas like demographics & technical deployments. Think of this write-up of mine as a consolidation of my research, with a few of my own 2 sats thrown in for good measure.
Contents
- Why should we care about Starlink?
- Why is Starlink even needed?
- More Users
- Thirst For Bandwidth
- Infrastructure Decay
- More Innovation
- Huge Potential
- Profitability
- Government Pivot
- Why should we be sceptical of Starlink?
- Privacy
- Centralisation
- What are the drawbacks of their network?
- What are the advantages of their network?
- Can you benefit without a kit?
- Describe the Starlink network (like I'm 205)
- Overview
- End-To-End Encryption
- Secure Networks
- V2 Mini Satellites
- Ground Stations
- Shareable Bandwidth
- Laser-Linking Tech
- Related Innovations
- Conclusions
1. Why should we care about Starlink?
- The growth of all satellites in space is going parabolic / exponential / bonkers! (2)
- Starlink have around half of ALL SATELLITES in orbit (3)
- As of writing on 18 Oct, there are 5,287 Starlink satellites in orbit, of which 4,925 are operational. (4)
- OneWeb, the next biggest internet competitor has just 634 (576 operational) (5)
- Starlink project is projected to expand far further as they have been requesting an additional ability to deploy 30k to reach a total of 42k satellites. (6)
2. Why is Starlink even needed?
More Users
Global demand for internet and the wealth it brings will continue to grow. We are currently at 64% adoption of the internet (5.18bn) (7).
The Compound Annual Growth Rate (CAGR) of internet adoption from 1995 to 2022 was 21.5% per year. From 2010 to 2022 it's been hovering around 8.3%. However with some of the largest countries in the world still yet to fully come online (e.g. China, India, Indonesia, Pakistan, Brazil, Nigeria to name a few) worldwide penetration alone is by no means complete.
Thirst For Bandwidth
Then just think of our own internet usage in recent years. Here is the timeline of internet expansion and our ever-increasing requirement for more bandwidth.
-
1995: The majority of people connected to the internet via dial-up connections. The speed of such connections was typically 56 kbps or lower. I miss those sounds!
-
2000s: The introduction and widespread adoption of broadband connections (DSL, Cable, Fiber) dramatically increased average internet speeds. By the mid-2000s, it wasn't uncommon to find connections offering several Mbps.
-
2010s: Fiber-optic networks and advanced cable technologies increased available bandwidth significantly. Gigabit internet connections became available in some areas. 4G also became the standard for mobile connections, offering tens of Mbps.
-
2020s: Many developed countries were deploying or had deployed 5G networks, which theoretically can offer speeds of up to 10 Gbps. In-home, fiber connections of up to 1 Gbps are becoming more common by the day.
Infrastructure Decay
In spite of the above timeline seeming lengthy, the internet is still relatively young and growing rapidly. We’re still changing the ways we use it. Just like our energy needs grow each year, so do our bandwidth needs.
The underlying infrastructure of the internet, microwave cell towers and optical fiber, are extremely expensive to build & upgrade. It’s easy to take for granted just how miraculous the internet has been, particularly how we have been able to upgrade our connection speeds over the course of the last 30 years. (. Just look at this map of the underwater internet cables:
Side Note: For a history on the build-out of the internet and all its cables across the world, you must watch this entertaining hilarious take from the Map Men (which I have posted separately here on SN):
More Innovation
It's easy to take for granted the tools of the internet that we have today. Stop for a second and think about it. Think also how lucky we would be if the most transformative application on the internet (or at least Bitcoin) has not even been invented yet. What if Bitcoin today is what the lightbulb was, when electricity came to our houses for the first time?
We're starting to see drastically new business models empowered by A.I and of course our fair share of over-hyped zombie companies. But we have not seen true intelligence combine with "truth" itself, empowered by the most secure network in the world. We have not witnessed autonomous employee-less organisations yet... nor a completely circular value-based economy without the friction comes with 'fiat'. Therefore we need the internet infrastructure that can empower this, no matter where you are in the world.
Liking this content? Follow me on Nostr...
On that topic, we also know that billions of people are still at present being left behind by the information revolution for no better reason than an accident of birth, or their country being a long way from a major undersea cable. In large parts of the world, internet is still provided by geostationary satellite at prohibitive cost. Although this has been changing in recent years, with many connecting and coding using just their mobile phones. It still remains obvious that more could be done.
Huge Potential
There are only few trillion dollar industries in existence: energy, high speed transport, communications, chemicals, IT, healthcare, agriculture, government, defense. Despite some Bitcoiners believing these are just around the corner; space mining and space-solar power are not going to be viable businesses anytime soon.
Elon has a significant involvement in the energy industry with Tesla, but only communications provides a reliable, deep market for both satellites and launch. (. This is particularly important given that Starlink plans to refresh the Starlink megac-onstellation every five years with newer technology. (8)
Profitability
SpaceX & spare exploration is only a sustainable business if it can keep putting satellites in our own orbit and providing communications to the world.
Once SpaceX have their 42k cheap & reliable satellites in our own orbit, they will be able to sell “server space”, dedicated bandwidth, hosting and/or partitioned networks for future organisations & entities. And they will be able to do so at significant margins, given their incredible cost advantage at present.
To help comprehend the cost efficiencies of adopting the Starlink network ( :
- A 'middle of the road' estimate for Starlink (by someone more intelligent than I) was that it would be 8 times cheaper, all in, than even JUST installing the undersea cable alone.
- Some have estimated that the cost to built and deploy each satellite could be as little as $100k per satellite. That is more than 1,000 times cheaper than a conventional communications satellite launch.
So yes, very very cost effective and therefore a very profitable business. Who needs space exploration or mars landings when you have that kind of business model?
Government Pivot
It is clear that only governments’ desires for assured military access to space, has kept Starlink around today. Without those subsidies Starlink may not exist, or they at least would not be so far along with their roll-out plan. But this relationship works both ways given that the U.S. needed to find a way to unplug from expensive obsolete rockets, built by big defense contractors. Both the United States government depends upon SpaceX and SpaceX depends on the Government. For now at least, it is important to acknowledge this.
3. Why should we be sceptical of Starlink?
Well, I just shared one reason. They are very much in bed with Big Government. Initially when researching Starlink, I had a viewpoint that Starlink was going to be like what Big Pharma has been to Health. I had severe concerns both about privacy and centralisation. So let's drill deeper into those topics as well as others.
A. Privacy Concerns
Firstly, is Starlink using strong encryption for their network? Yes. They actually are setting new standards.
I dive into this in more details in a later section (7), but essentially they are using “end-to-end encryption encoded at firmware level” so that the network has apparently no knowledge of any packet-level data. That said, Starlink themselves are capable of identifying and profiling their users based upon the serial numbers of the hardware, their payment information and other metadata.
In some ways, internet metadata is more of a privacy concern than the packet contents themselves (8). The metadata encodes the graph, social and otherwise, of anyone on Earth along with everyone else they have communicated with. A unique, time-evolving fingerprint that often reveals more about a person than the packet contents.
Exposing even metadata, and even to SpaceX, is a bad idea. However we MUST remember that in this new era that some call the Information Age, Data is a precious resource. For Starlink to build-out the next internet, without the ability for them or the United States Government to have any access to those riches, would be quite the gift for all us ants on Earth. So why would they do it?
Geopolitical Implications
If for one second we put Starlink's intentions to one side, the fact that traffic broadcast across the network is completely private will mean that other nations (especially those not on the U.S. Christmas card list) are at a near total loss of control over what their citizens do over the Starlink network.
For that reason alone, we can be sure China will NOT be issuing a license or recommending Starlink to their population unless the Starlink network routes all China-centric traffic through a "Great Firewall of China" ground station.
That said, if SpaceX were to have access to the data on the network, it would be legally compelled to filter content at the discretion of the United States government, undermining freedom, net neutrality, and greatly increasing legal risk. Has Elon learnt his lessons from Twitter (or X)? Probably not, but it is worth romanticising about the possibility.
Integrated Businesses
It's also worth acknowledging for a second too that Elon talks a good game, he is very charismatic. But like many others in his position, Elon is not incentivised to change censorship on X. He is very good at controlling the narrative. Elon clearly wishes to introduce payments as his new toy on X, that much is clear. And when 'fiat' payments are involved, identity is needed. Elon is likely to use the 'risk of AI' as a means to identify his customers, in order to bring widespread payments to X. Payments will therefore surveilled, so why would he be incentivised to prevent U.S. Government from surveilling its people? Particularly when as we mention the only reason they and Tesla exist is because of heavy heavy subsidisation, of over $1 BILLION in fact.
Government Snooping
As we know all too well, Intelligence agencies can and will use our data to spy on EVERY citizen, given any opportunity. Just ask our friend, Ed.
In fact, you may not be even aware but back in 2017 United States revoked a set of privacy protections that restricted telecommunications companies from collecting web browser search history, location tracking, and other data from its customers. Why would they revoke this?
Even though it is well known that they had not been respecting this law up until that point, they still felt it necessary to do this. By rescinding those protections, it now allows data from internet service providers to sell personal information to third parties. The same rules apply for satellite communications providers. So why would Starlink, having received state subsidies, choose not to sell this data either? They legally & commercially have the blessing as well as the incentive.
This is in complete contrast with what happened in 1945, when the United States courts PREVENTED powerful communications platforms from becoming monopolistic, by agreeing that countering such moves would not constitute infringement of free speech or of the press. (9)
In that specific case, the U.S. government intervened to ensure telephone giant AT&T did not also control other large communications verticals, as Musk now does with both Twitter (sorry X) and SpaceX (Starlink).
Back then, it wasn't in the United States' interest to let this happen, but one could argue that today Starlink can be the darling for the United States government. Most significantly because Starlink allows the United States to digitally expand their reach into regions and markets where local governments cannot and will not compete. It extends the life of the welfare state at the expense of many others.
As mentioned above, let us not forget that the United States government has heavily subsidized Starlink's "land grab" in orbit and will likely wish for their snoop-shenanigans to continue, given the value of our data. Whilst everyone is focused on potentially re-drawing the boundaries on earth, the United States is extending their reach across the sky.
The Counter-Argument
Bear with me whilst I flip and flop to play both sides of the argue, but you could argue that SpaceX would prefer to spend money on engineers than lawyers.
Building a private network for everyone, that lacks access to metadata, is a great idea in the Information Age. Such a network by default preserves privacy, ensures security for e-commerce, and discourages hackers from intercepting or compromising their network (0.
Being private of course removes a huge potential revenue stream. It's certainly plausible Elon's other products & services will have more than enough personal information on us, without needing to identify us on the base layer of the revamped internet. Better to be cynical and consciously optimistic than doom and gloom on SpaceX's ambitions.
Ambiguity Speaks
Until it is proven that Starlink cannot and will not pair our internet usage with our identities, the safe and logical position to take given recent history, given the legal prescendents, given state subsidies and given the overriding incentives on offer, would suggest that they would wish to collect this information on their customers. Even if the network itself is encrypted on route to and from our hardware or between each of the satellites. This data is just far too valuable and the status quo has set.
B. Centralization
The other significant concern is around how much influence one company would have on our internet. What remains to be seen is if the Starlink team will meet any headwinds in the future, to achieving monopoly status and becoming what Kodak were for the camera, what Intel were for the chip or what Google still just about remain today for search.
Amazon already has a robust data collection business in its retail sector, which the company could pair with telecom data to become a micro-targeting data empire (1). However with Amazon's Project Kuiper network hardly getting off the ground and with OneWeb possessing less than than 13% of Starlink's network, Starlink are literally miles ahead in the sky above the competition. And Amazon are having to resort to writing to the FCC to attempt to slow Starlink down (2).
Starlink could feasibly become a monumental monopoly in just a few short years. So much so that they are even willing to deploy their competitors own satellites into space, having inked a deal with Telesat for launches in 2026 and beyond. Talk about competitive confidence. (3)
Loss Leader
According to its own disclosures, Starlink loses more than 50 percent on every kit it produces for customers to keep the price at $499 per package (and $99 per month for service) (4). This has likely come down in recent years but has undoubtedly worked in Starlink’s favour. Starlink's main competitors (5) will likely continue to have stunted growth at their expense if this continues.
Rapid Growth
The company has added customers at a rapid rate, more than quadrupling a base of 90,000 in 2020 to 400,000 earlier in 2022 ( to now Starlink having just surpassed 2,000,000 customers (6)
Staggering Subsidies
Musk himself has made good use of government subsidies, having received more than $223 billion in government subsidies for Tesla and more recently $886 million for SpaceX. (7)
Questionable Behaviour
One last point to mention with regards to SpaceX's monopolistic behaviour and desperation to grow quickly and take market share; the FCC was forced to cancel a billion-dollar grant awarded to Starlink (8) fairly recently after investigative reporting showed Starlink had fudged numbers on its application.(9) Questions can quite rightly be asked if as we know Starlink has repeatedly tried to exploit agency negligence. Therefore, can they be trusted with maintaining the decentralised infrastructure of our internet? And our data? That's a tough one too.
4. What are the other drawbacks of the Starlink network?
There are other well-documented downsides to having Starlink besides the privacy & centralizing consequences discussed above:
- Light & reflections interfering with space exploration
- Signals interfering with other satellites, communications & astronomy
- Debris in orbit, potentially causing collisions & obstruction
- Short expected life cycle of these satellites
Here for instance is a video on how the amount of "space junk" has drastically increased over the last few decades: . SpaceX has however published a document seeking to address some of the stigma with regards to their satellites (20
5. What are the key advantages of the Starlink network?
The drawbacks ARE likely surpassed by the benefits that faster, wider-reaching internet can bring - given the obvious lack of need for infrastructure. Some have been mentioned already but these include the fact that:
- Starlink satellites are small and can fit in the palm of your hand.
- They are MUCH cheaper, as highlighted above in the 'Profitability' section.
- This network can bring billions more into our connected world.
- Future capacity - we are already using existing fibre optics near enough at their limits.
- Starlink allows us to build-out a surplus for the future and re-deploy or re-distribute bandwidth globally, instantly. Rather than leaving internet capacity stranded/unused in remote locations. The world can be far more inter-connected.
6. Can you benefit from Starlink without their kit?
Somewhat. You will need a Starlink kit and subscription to take full advantage of the network today, particularly at the impressive speeds being achieved. That said, Starlink has signed deals with a number of mobile telcos in recent years, enabling them to fill gaps & blind spots in their service. It was rumoured that many countries (mentioned below) would have their cell networks using Starlink internet by end of 2023 potentially (21).
And that was confirmed just last week, with Starlink "Direct To Cell" (22), now bringing texts by satellite in 2024 and voice calls by 2025. The following have signed agreements with Starlink, although they are all likely now in direct competition from Starlink from here on out:
🇺🇸🇬🇧 - T-Mobile (23)
🇪🇸🇬🇧🇨🇴🇧🇷🇵🇪🇨🇱 Telefonica (inc. Movistar, o2) (24)
🇦🇺 Optus (25), Telstra (26)
🇯🇵 - KDD (27)
🇨🇭 Salt (28)
🇨🇦 Rogers (29)
7. Describe the Starlink network (like I'm 205)
Overview
IP is the fundamental protocol of the Internet, with TCP and UDP protocols riding on that.
Elon has previously stated that Starlink will utilize an IP-less, "simpler than IPv6", “end-to-end encryption encoded at firmware level”, peer-to-peer protocol, which insinuates encapsulation of our IP packets at the network ingress Gateway (our user terminal) and de-encapsulation at the Ground Station peering point (whereupon our packets are released out onto the wider Internet). And vice versa.
'End to End' Encryption
Implementing full encryption across the Starlink network would ensure that the network cannot inspect the content of data packets or identify the protocols they contain. Protocols such as HTTP, FTP, Email (e.g., POP3/SMTP/IMAP), BitTorrent, and others.
In simpler terms, the data packets would be scrambled with encryption, except possibly when they enter or exit the network, for example when traveling between Gateways and Ground Stations or between Satellites and Mobile Devices. This is what they refer to when they mention 'End to End' encryption.
Considering the potential threat of third parties intercepting radio or laser signals on the ground or in space, it's crucial that encrypted signals are practically indistinguishable from random noise.
While the Tor Network was initially designed to provide privacy features, it isn\t perfect for a scalable internet in terms of performance, reliability, functionality, nor adoption. However, if Starlink was to be well implemented, it has the potential to offer privacy during data transit similar to Tor. Additionally, Starlink must deliver maximum bandwidth, minimal latency, and minimize packet header overhead (31). Whilst encryption does introduce some processing overhead, the benefits of data security and privacy provided by encryption would outweigh these minor performance considerations, particularly as encryption algorithms and hardware acceleration have further reduced the impact on network performance.
Secure Networks
It must be said however, that the U.S. Air Force in August 2022 signed a deal with Starlink, making it the exclusive Internet provider in many countries where it operates. It therefore clearly is capable of providing super secure communications. (8). It followed-up that announcement with Starshield in December 2022, which is essentially a type of hosted & separated Starlink instance, specifically for Governments (9).
You can read more about Starshield on their public link, which is basically offering everything we would want from our internet, but strictly to customers desiring 'Defence' and 'National Security' (20 - (https://www.spacex.com/starshield/).
Starlink already offers unparalleled end-to-end user data encryption. Starshield uses additional high-assurance cryptographic capability to host classified payloads and process data securely, meeting the most demanding government requirements.
V2 Satellites
-
The latest satellites deployed to orbit are these named 'V2 Mini satellites'.
-
Each V2 Mini has have 4 times the capacity for serving users compared to its earlier counterparts, known as V1.5.
-
Each V2 Mini weighs about 1,760 pounds (800 kilograms) nearly 3x heavier than the older Starlink satellites.
-
The V2 Mini satellites unveil 2 solar array wings to a total span of over 100 feet (30 meters) wide. The original Starlink satellites each have a single solar array wing measuring about 36 feet (11 meters) end-to-end.
-
They also use an “autonomous collision avoidance system” to help avoid impacts with other objects in orbit. Something previous versions did not have. Plus they boast thrusters 2x more powerful to position themselves in orbit and dodge the space junk.
-
SpaceX plan to deploy the larger variations of the V2 satellites starting next year in 2024.
-
The full-size Starlink V2 satellites, will have more than double the surface area.
For the full list of SpaceX launches, you can find them all listed on NASA's website (in albeit a really awful format) (32).
Ground Stations
Starlink ground stations harness highly sophisticated tracking and positioning mechanisms that are capable of observing with precision every satellite’s position as well as trajectory. They help enhance the quality of the connection given the ever-changing orbitals, by being able to combine the satellites with these complex antennas it allows ever more precise monitoring & speedy communication.
There are 2 in Africa, 5 in Asia, 20+ in Europe, 2 in Middle East, 12 in Oceana and most impressively 78 in the Americas (18 in Central, 17 in South and 35 in U.S alone). (33)
Shareable Bandwidth
All users or subscribers within one beam spot have until recently had to share bandwidth. The size of this spot is apparently determined by physics – essentially its width = (satellite altitude x microwave wavelength/antenna diameter). Which for Starlink satellites works out to be, at best, a few kilometers.
In other words, dozens of satellites were needing to point their most powerful beam at the same area, and subscribers in that area would use ground terminals that would split the demand amongst those satellites.
Laser-Linking Tech
A recent feature-add by Starlink to install lasers on their satellites means that now however, they can communicate spread the bandwidth across each of the satellites. This is why they expect to be able to upgrade their satellite network every 5 years. New innovations are advancing fast.
The value that satellite-to-satellite laser links provide is that they create a low latency, high bandwidth path that stays within the Starlink satellite network (34). Without it, each Starlink satellite is only capable of communicating directly to ground terminals (either consumer, transit or SpaceX control).
For traffic that is intended to move large geographic distances (think transcontinental like flights), this would otherwise require several hops back and forth between ground and satellite.
Fun fact: this is why we have to use airplane mode on flights. Not because it will disrupt the plane's own communications but because our phones have to constantly ping back to earth, and having hundreds of devices do this along the entire flightpath, would disrupt cell coverage for many of those on the ground (35).
By performing this type of transit directly in space, and exiting at a transit node nearest the destination for the data, you greatly reduce latency. Not to mention the requirement for those ground stations.
Related Innovations
I also recently stumbled-across a seemingly unrelated new article about an innovation for a new type of algorithm for networking & flow (36). As soon as I read it, I figured it could have implications for Starlink.
For now, it’s basically a theoretical possibility but its benefits for network computing have been labelled as profound, especially for networks far larger than ones we have today. The intention with that algorithm is to increase latency and speed, beyond the bounds of our existing limits for 'Maximum Flow'.
Eventually, the researchers created an algorithm that runs in “almost linear” time, meaning that as you look at larger and larger networks, its runtime approaches some multiple of m. It’s a “tour de force,” Mądry said.
A network as large as Starlink's could be just the application for breakthroughs like these.
8. Conclusion
Thanks for making it this far. That was quite the ramble! Whether you have skimmed-through or read every word, hopefully it has provided some useful context on the Starlink internet network. The new internet that we will all be using, in all likelihood, in 5 years time.
We got into the technical specs, encryption, the history of the internet and even network flow. The biggest question to answer for me coming into this, was around encryption. I have since learnt that Starlink via SpaceX will become an internet monopoly if left unchecked. They are a powerhouse.
For me, this research doesn't definitively answer the question of whether Starlink's unparalleled influence in space and satellite communications is a net positive or a net negative. I will let you decide this.
But to answer the original question of whether or not we will see Starlink utilising user data or at least our metadata to their benefit, on balance I have to lean in-favour of the status quo being maintained here. Reasons being:
- The laws changed in 2017 to legalise data collection by telecommunications providers, removing any remaining ambiguity.
- Data is already more valuable than oil. (Although this may be put to the test over these coming years).
- It is in U.S. Government to gather more data on more worldwide communications.
- There is a separate 'more secure' instance of Starlink specifically for Governments worldwide, named Starshield.
- When Government subsidies dry-up, SpaceX will wish to fund continued upgrades to their satellite network.
- Leveraging user data would also allow them to be able to potentially subsidise internet usage for their customers, in some regions where equipment is unattainable, if they so chose.
It may indeed turn-out that the level of data-collection could be an afterthought. If Elon's businesses of Tesla, SpaceX and X were to ever become as one. Even if that never came to fruition, I still think we're about to witness one of the biggest monopolies we've witnessed in human history.
Share your thoughts...
Please drop your comments below on what your assessment of Starlink is, and how you perceive the future of our internet to be. More or less free.