Hacker News new | ask | show | jobs
SpaceX is buying communications smallsat operator Swarm Technologies (fcc.report)
93 points by caiobegotti 1773 days ago
5 comments

I have a theory they're buying it more for the engineering talent and personnel, than any intention to run a narrow-bandwidth cubesat M2M communications network as a separate thing to starlink.
My theory: they are buying out a potential competitor, and thier IP, before they become a threat. Killing them in the cradle as it were. It sends a strong signal to anyone else talking about satellite-based network services.
Starlink and Swarm don't meaningfully compete. Starlink is a V-band (GHz-territory) system suitable for high bandwidth applications, but requiring good line of sight, lots of power, and a big, complicated phased-array antenna (or I guess you could maybe have something actively steered instead, but SpaceX doesn't seem to be pursuing that).

Swarm is a VHF system (hundreds of MHz) that can't come anywhere near Starlink in terms of bandwidth, but can operate with much lower power and a small, simple, dirt-cheap antenna. The tradeoff is that you'll end up paying way more per byte. But for their intended applications (basically, embedded IoT-type sensors), you don't want to transmit/receive much anyway, and keeping BoM cost down is key.

In other words: there's basically no application where someone might be choosing equally between these two options.

That roster includes O3B mPOWER (SES), Inmarsat, Kuiper (Amazon), OneWeb (UK et al), Guowang (China) and Sfera (Russia). I doubt they are so easily intimidated.
Intimidated? Or maybe if you do the right things SpaceX will bring you trucks full of money to stop.
Umm, they are already pretty intimidated. As Starlink stands right now (in beta with partial constellation in service), it's superior to all of those combined. Those guys are shitting their pants right now scrambling to either block SpaceX from moving forward (by whining to authorities) or setup their own competing constellations.
And Telesat (Canada) also raised money to build a LEO constellation.
> sends a strong signal

A signal of encouragement, you mean? Like, "invest in this space" -- no pun intended -- "because we'll offer you a great exit"?

> they are buying out a potential competitor

I think it's pretty clear SpaceX doesn't have any competitors in the satellite constellation business.

SpaceX will spend something like ten billion dollars to build out the full Starlink constellation using their own launch services, which are vastly cheaper than anything anyone else has.

Potential competitors would have to spend an insane amount of money to launch their constellations, which basically means it will never happen.

So they are in violation of antitrust rules? If space is the future, SpaceX is the railroad and should not grant cheaper rates for cargo from its other businesses.
Good luck arguing this to regulators. They are cheap, but not the only option, both in the US and globally.
That logic doesn't work with current court interpretation of Anti-Trust. Read Judge Richard Posner on modern Anti-Trust law. Its not the late 1800s where it was basically politicians driving Anti-Trust train.
Can argue SpaceX is a global telecom company which is selling excess transport capacity to fund their transport cost center.
So they're pulling a Facebook/Instagram?

Here's hoping someone with investigatory power is reading this.

That sounds likely. SpaceX is a lot further along in development than Swarm appears to be -- it's unlikely that they see much value in the specific technologies Swarm has developed.

Interestingly, Swarm had launched some of their microsatellites on SpaceX vehicles.

My theory is more that they will add the functionality to Starlink. Worst case, you just plug the RF part of the swarm cubesat on the outside of a Starlink. This way Starlink immediately supports low power devices, has a ground segment for the end-users readily available, and solves the latency/backhaul problem that all these cubesat constellations have.
It's worth noting that it won't be as simple as plugging in Swarms' payload - there will be singificant necessary engineering work to validate the RF and power characteristics of the integration.

Interesting work I'm sure.

Other than the sat-to-sat communications (which starlink is also planning) is there any real difference between the product offerings? Seems like a reasonable theory.
Very different in the details. I commented above, but tldr: Starlink is suitable for high-bandwidth applications but requires a complicated, high-power antenna. Swarm is only suitable for very-low-bandwidth applications (IoT), but can get by with very simple hardware. They're not really competitors.
Very complementary.

They are going to have synergies obviously around launch (swarm sats are relatively tiny so cost now with spaceX for launch is going to be very low).

Product is also differentiated. You want global low bandwidth (primarily M2M but can also be search and rescue etc) along with the higher bandwidth / power / footprint starlink stuff. Businesses often strugle with being competed against from the bottom - I hope SpaceX keeps the swarm idea going.

Obviously a way to pick up capable engineering talent. If they can do sat to sat comms (I think still an area of some development to really dial in). that's going to be big for everyone - and if swarm can interoperate / get backhaul in space from starlink - amazing.

Some details on the technology/licenses being acquired: https://fcc.report/IBFS/SAT-T-C-20210806-00096/12345289
>Coincident with this filing, Swarm will be withdrawing its Petition for Market Access inthe 399.9-400.05 MHz and 400.15-401 MHz bands and concluding its remaining experimental license under the Experimental Licensing System call sign WI2XYY

This is an indication that SpaceX has no intention of expanding Swarm’s business, but rather acquiring the company for its engineering talent as suggested in the other thread.

Very curious about the price. And whether this is a hopeful acquisition to expand on what Swarm is doing, a fire sale of a company that couldn't make the financials work, or some monopolistic move to snuff out a competitor.
This. I am interested to see exactly what SpaceX does here with IP they acquire. It should be a shame if it was all mothballed.
The Swarm Tile looks easily integratable into PCB designs. And, it's very low power. Although, I wonder why they need about 3 Watts power consumption when transmitting. At their low transmit frequencies (148-150 MHz), I would think they could lower that number to less than 1 Watt since the wavelengths are so long. But hey, if you deploy your sensor in a sunny location with solar panels, then why not?