Hacker News Reader

Comment by samwillis | Hacker News Reader

I don't think this was an unexpected result of the situation with Russia, there wasn't really any other options, but interesting to see SpaceX launching a direct competitor to Starlink. I suppose they always made it clear that they were not going to let their investment in Starlink stop them launching competitors constellations.

I am still somewhat surprised by the UK governments investment in OneWeb, I hope the strategy pays off and it becomes an important national infrastructure project. My thoughts where that it may be about providing a future proof network for the remote areas of the UK but we aren't exactly a big country so I'm not convinced that's the case. The rubbish that was written about it being a route to our own GPS system after Brexit forced us to leave the Galileo program was clearly all fluff to make it sound more impressive and tie it to a post Brexit strategy in some way.

View on HN

> interesting to see SpaceX launching a direct competitor to Starlink

It isn't, exactly, a oneweb terminal is MUCH too big and expensive for an ordinary residential consumer or small business. It's two active tracking parabolic antennas with their own RF chains in radomes, takes up about a 2.5 meter long x 1 meter wide space on a roof or similar.

More than 1.5 years ago oneweb pivoted to a plan to sell high capacity uplink services for regional ISPs and telecoms on a business to business basis only. The oneweb terminal is still much larger and more costly than the recently announced starlink premium.

oneweb in its current plan to sell services to ISPs currently dependent on geostationary is more like a cheaper/slightly smaller o3b terminal.

I had been talking with AST Group about getting OneWeb set up on my boat and the size of the terminals was a blocking issue; only place I could put them would be shading solar panels much of the day.

Unless you have a gargantuan yacht the monthly recurring cost is going to be prohibitive anyways, I'd be shocked if it's less than $1200/mo to start.

Right, but that is cost I could potentially swallow; worth it work from paradise anchorages with little other connectivity. You don't pay rent at anchor.

What I will probably end up doing this summer is a stern-to mooring to some trees with a Starlink terminal sitting on the beach with an Ethernet cable running out to the boat.

> stern-to mooring to some trees with a Starlink terminal sitting on the beach with an Ethernet cable running out to the boat.

How come? What's the problem with having the Starlink terminal on the boat?

It doesn't deal too well with all the movement; it works sometimes but not reliably. I don't know if it is only the rolling or also the lateral swinging around the anchor that throws it off.

In perfectly flat seas it might work fine, but you don't often find yourself in those conditions.

Seems 'marine stabilized platforms' of many specs and sizes are advertised. Surprising if none suitable to mount a starlink on.

A catamaran would be a nice "marine stabilized platform"

Not necessarily.

Catamarans tend the roll pretty aggressively, even if they aren't going to roll over a wide angle.

A SWATH - small water-plane area twin hull - would be a better choice. Basically two completely submerged pontoons with small pylons passing though the water surface up to the superstructure.

Because they have very little boyancy change when the waves change the water surface level, they have very gentle roll characteristics.

It deals REALLY well with movement. I’m very curious what you’re seeing and if you’re on the latest firmware?

My guess would be the moving radio shadows cast by mast, rigging, boom.

Last tried in April of last year

Can’t you set it on a gymbaled gyro to stabilize it?

If you're that close to the beach that you can string a cable, do you not have 4G LTE/5G services available? With a high-gain antenna I would imagine you could get speeds and latency that would rival Starlink in many places. Of course, this will vary wildly and depends on how remote you are mooring.

It just depends; every anchorage is different however it's not uncommon to be cruising in area with great coverage, but once you lay anchor you find yourself without service because the rocks of the bay you chose to shelter you from sea waves also shelter you from radio waves.

What about point to point WiFi instead of a cable? WiFi 6 will handle anything that Starlink can possibly hand out. With a directional antenna, you'll have more range than you would ever want to store cable for.

Then again, I would think you could make a stabilized platform that could keep a starlink connection up.

The advantage of Starlink over point to point WIFI is there is so much less set-up involved; just update your service address and boot your terminal and you're good to go. Conveniently the ethernet cable will also power the terminal from the boat's batteries.

> with an Ethernet cable

Just use a WiFi bridge, eg: https://mikrotik.com/product/wap_60g_ap

Starlink is likely to eventually have a mobile-base capable terminal. Probably not by this summer, though.

Surely Starlink's addressable market is a superset of OneWeb's? A fully-launched LEO constellation will make many of OneWeb's customers' use cases obsolete.

That said, as others have pointed out, whether OneWeb does or doesn't use SpaceX as a launch provider doesn't really change that outcome (whatever it happens to be).

In that case, isn’t it oneweb helping their competitor more than their competitor helping oneweb?

I think OneWeb is backed by the british government, and as we're finding out, global satellite networks are quite useful in military conflicts. While there's definitely a commercial aspect that has value, the military value is likely significantly higher.

If one web requires car sized antennas the military value is probably much lower - though that might just be the commercial offering and not a fundamental issue.

AWS hosts a huge amount of direct Amazon competitors. Explicitly banning certain competitors in an unrelated business would be pretty damaging to the credibility of a platform.

Yeah someone else is going to take that money. Another competitor taking it would potentially be more damaging.

not to mention you can use that as a way to spy and choke your competition out when they rely on you

Avoids accusations of anti-trust.

Basically this, for SpaceX it is all win, they get to say "see? We are agnostic and will launch competitors." And at the same time the profit they make from launching OneWeb satellites can be invested in growing their own business.

I tried to explain this sort of thing to Intel once about opening up their fabs (before Pat Gelsinger took over as CEO). Selling access to your infrastructure for profit lets you invest in better infrastructure without using profits from the things you sell using that infrastructure internally.

Yeah especially if you know that the competitor’s product is inherently worse than yours (OneWeb vs Starlink).

Good analogy. As long as oneweb can differentiate they’ll be fine. And as long as Starlink has vertically integrated launch, they’ll always have price as a differentiator.

I would say that aws is a rich substrate upon which many businesses can be built - many of which Amazon simply arent interested in getting into. Space ISP is one business and the price/bandwidth ratio is one of the few differentiators and that alone may decide the winner.

So oneweb better get creative about what they’re layering on top of their pipes.

For me, I consider the main reason for Starlink to exist is to make use of SpaceX launch capacity. SpaceX focus on assembly line production techniques and reusable rockets to bring down costs per launch only make sense if you have a lot of stuff to launch. With Starlink, they create their own demand.

SpaceX is probably very happy to have a "competitor" paying for launches. By itself, I have some doubts about the profitability of Starlink anyways.

This 100%.

SpaceX is really the only provider who can credibly pick up the former Soyuz business in the next few years. And they need _a lot_ more business. Especially once Starship's capacity becomes available. Right now, there's just not enough up-mass or down-mass demand.

They're at 250,000 customers now, so that's a run rate of $300M a year from Starlink. I fail to see how their cost of operations is above that if they were to stop launching additional satellites which means they have per-unit profitability pretty good already.

I think it makes perfectly good sense for SpaceX. They've always made it clear that their real business goal was colonization of Mars based on optimizing the cost of transport to space via reusability and design for manufacturability. Everything else they've done is basically just a way to monetize their current space travel capabilities in service of funding the design and construction of what they really want to do. Making Starlink as profitable as possible was never a goal, just a way to make some more money off of their incredibly low launch costs and probably also develop technology for communicating between ground and spacecraft. So why not launch a competitor too? It's just more launches and more funding for them.

I have a nice dream that Elon will build rocket and space enterprises, get them profitable, only to break them up and go public to allow them to be independent corporations and allowing real competition to begin (plus remove all his other business interests from each companies mission)

He could still rake in cash from their owned shares it would just be them creating a real public space sector for the good of humanity. This will never happen ofc, but I can dream.

That sounds like it might be even more profitable, but he already has more money than he knows what to do with. Besides, that seems less fun than being absolute dictator of mars.

He has a lot of wealth, but he doesn't really have that much "money". He can't re-invest in SpaceX by selling SpaceX and then buying SpaceX again... He could diversify away from SpaceX and Tesla, but he doesn't seem very interested in that.

You can't rake in cash from owned shares if they don't pay dividends and I don't think he has much interest in selling his shares. He's previously said repeatedly that SpaceX won't go public until there are regular flights to Mars, which will likely be after his death.

> interesting to see SpaceX launching a direct competitor to Starlink.

Interesting, but entirely expected. They'd be investigated in a heartbeat if they didn't.

I don't think so. That's not anti-competive. Its like intel refusing to make amd chips on intel fabs. Perfectly logical

I think that only works when there is other capacity available. If Intel was suddenly the only fab available to western companies but they refused to fab AMD chips I suspect that would result in intervention.

Not so sure. AMD is always free to build a fab to make their own chips. It's perfectly reasonable to refuse to serve a direct competitor...or offer the service at a huge expense that would make it unreasonable for the competitor.

There is extensive legal precedent, and numerous outright laws in many jurisdictions restraining anticompetitive practices such as this.

'AMD is always free to build a fab to make their own chips.'

By that logic there us no limits on a monopoly abusing its power, after all you are always free to create your own water supply company, electric distribution company, etc.

I get what you're saying, but SpaceX is a bit more public/private than Intel, and likely the government would be upset and better fund competitors if SpaceX was monopolizing launch capabilities. Their strategy (at least for now, as I would understand it) is to be neutral for launching cargo/services/etc. - likely OneWeb is paying more to deploy than SpaceX pays itself internally for Starlink deployments as well.

You're saying that SpaceX, a privately owned company that has some contracts with different parts of the US gov't would find themselves in trouble with the government if they refused to launch satellites from a foreign owned company?

And you're saying that the government would go out of their way to fund alternatives to SpaceX because of this?

That seems a bit implausible.

"foreign-owned" in this case is still the UK i.e the closest ally we have today. So I don't think you'll see the same sort of American protectionism you might see with e.g a Chinese competitor.

Therefore... yeah, nothing's stopping the DoJ (guessing the FTC would make the referral?) from pushing an antitrust matter. But I certainly can't say for sure; I'm not a lawyer.

Where's the law that says that an American business must sell their product or service to a foreign company?

> Where's the law that says that an American business must sell their product or service to a foreign company?

As I understand it, it's covered by the various antitrust laws in the United States. And it's not so much a "foreign company" thing so much as it's an unfair advantage for any one company thing. https://en.wikipedia.org/wiki/United_States_antitrust_law

Again, not a lawyer. But my lay reading would say that it would feel comparable to the action the feds took against Microsoft for trying to stifle Netscape. But I'm sure there are far better analogues.

Foreign owned maybe, but the satellites are manufactured in the US. There are US commercial interests at stake in Oneweb, and the services it intends to offer to US clients too.

Under antitrust law, SpaceX would theoretically be made to split off StarLink from the launch business so that the launch business would have no incentive to prioritize StarLink over other sattelites.

In practice though, antitrust laws aren’t enforced very strictly and government contractors are treated leniently, so probably nothing would happen.

How?

SpaceX has no stranglehold over the market.

Presumably 'internally' Starlink is seen as a major recurring customer who has agreed to an exclusive purchase contract for better rates; and also to take on higher risk mission slots (like the 12th launch of rockets which are making new records for launches).

Is SpaceX a subsidiary of Starlink? Cause if not then it's not that logical to me...

Starlink is not a separate company, it's a product of SpaceX.

Intel doesn’t sell fab services, so it’s not an issue. However if you offer and advertise a service at a listed price (as SpaceX does), you can’t always refuse to provide that service to specific potential clients just because it serves your commercial interests. It very much depends on the specifics of the service and market competitive situation.

This is factually incorrect. Intel has in the past and is currently trying to sell and expand their foundry services. They just happen to have done a horrible job of it so far and have repeatedly lost foundry customers (like Achronix) as a result.

SpaceX is an american company that is subject to the Sherman act, which deals with domestic and interstate commerce. OneWeb is an international competitor based in UK that has no standing to make a complaint under the Sherman act.

No similar antitrust provision or treaty exists in international commerce; charges against a monopoly must be brought within a certain national jurisdiction, SpaceX does not have partners or subsidiaries that offer launch services outside the US.

I’m sure Oneweb will have a US subsidiary and offer service to US customers through it. They manufacture their satellites in the US. There’s no way they wouldn’t have standing.

But that subsidiary would need to build and operate a satellite constellation, not simply distribute internet services or act as a purchasing agent of a foreign competitor.

US DoD and NASA now have effectively SPOF reliance on SpaceX...they aren't going to be "investigated" for anything

I don't see why the government contracts would act as a shield for SpaceX. It's not like being forced to carry a competitor is going to make them unable to carry the government's rockets.

Stone Brewing does distribution for a lot of smaller local breweries

I'd like to think that cutthroat competition is not the only way of doing business

In the beer space, Sam Adams used both their massive purchasing power and stockpile to provide hops to small brewers during a shortage.

Funny, I think I recall someone at Stone citing that story, specifically as a model for their own corporate behavior, when I was on a tour of one of their breweries.

I thin Ukraine is proving to national governments exactly how important LEO constellations are. I was pretty critical of the decision of the UK gov't to bail out OneWeb, I think we may look back and say that it was a dramatically fore-sighted decision by the government.

The UK still has a insane amount of geographical dispersion, even in their post-empire state. Having the ability to ensure that no one can turn off the lights on their communications is important.

It was definitely a problem in the past, hence why they originally went with the Russians. But since then OneWeb has been bought by the British government so while its a similar tech I don't think they are strictly competitors anymore.

It's a positive outcome for SpaceX/Starlink. The former will increase revenue while the latter will have a cost advantage regarding their main competitor. There's no need for then to draw bad publicity.

Perhaps the government are using it as a cheaper way to provide super fast internet (flexibly defined) to the last 1% or so who can’t be affordably reached by standard exchange - green cabinet - premises measures.

Reply on news.ycombinator.com