Hacker News Reader

Comment by nordsieck | Hacker News Reader

IMO, one of the central problems is the cost of launching stuff into space.

Today, SpaceX offers world leading low prices to launch satellites: $4 m / tonne[1]. But Starlink has access to launch at cost, which is $0.86 m / tonne[2]. Which is a huge advantage when launching an enormous number of satellites.

One thing to keep in mind, especially for these LEO constellations: the lifetime of these satellites is 5-10 years. Which means the operators can never stop launching. It's an ongoing operational cost.

For smaller operators like OneWeb, they don't have to launch that often, but for a serious competitor like Kuiper, they'll be constantly launching some satellite every year.

IMO, launch cost will be a problem even for China. The cost of an LEO constellation is so high that even if it's partially subsidized by the military it'll be a serious cost for the country.

That could change pretty soon, though - various companies and organizations in China are aggressively working on getting reusable rockets working.

---

1. $70 m / 17.5 tonnes == $4 m / tonne

2. $15 m[3] / 17.5 tonnes == $0.86 / tonne

3. The $15 m number is not public info, but it is widely believed that it is in the correct ballpark.

View on HN

In the past this at-cost dealing would have been considered monopolistic enough to force divestment (ie, almost exactly the same as Boeing and United Airlines divestment due to the Air Mail Act).

> Air Mail Act of 1934: > This legislation prohibited the common ownership of airlines and aircraft manufacturers to prevent conflicts of interest and promote fair competition in the aviation industry.

> In the past this at-cost dealing would have been considered monopolistic enough to force divestment

I guess we'll see what happens.

As with most things monopoly related, the critical fight is over how to appropriately define the market. Presumably SpaceX would argue that Starlink is an ISP and that it just happens to use satellites to deliver its service.

And if that doesn't work, then it's a satellite internet provider, but competes with both LEO and GEO services.

If it ever goes to court, it'll be interesting to see how such an argument holds up.

For obvious reasons I think it's pretty safe to say we can count on at least the next four years of zero regulation or government scrutiny of any company Musk is involved in, monopoly-related or otherwise.

You believe the FAA, FCC, OHSA, EPA, FDA, etc., will cease to regulate SpaceX, Tesla, Nuralink operations? That seems pretty outlandish.

Elon Musk was just appointed by Donald Trump as being in charge of firing half of the government. He won't likely do that. But any regulator who gets in his way?

Yeah, not many will volunteer for "the firing line."

I really don't know what Musk has been appointed to do and it's a laughably blatant conflict of interest, but conflicts of interest seem to be what the entire government is built on. Politicians involved with energy and military companies are involved in decisions to go to war, generals get lucrative consultancy jobs at military firms, congress makes billions of dollars insider trading, foreign aid somehow finds its way funneled through "charities" owned by the ruling class, politicians cosy with medical companies block real healthcare reform, etc.

Musk isn't anything new or different here. The idea that he'll just be above the law is fearmongering hyperbole though. Sure he'll get favorable treatment and be able to push his agenda to degrees well out of reach of us commoners. No more than if he'd just stayed in the shadows and bought his politicians and judges and bureaucrats and generals like a normal billionaire.

I follow the same reasoning as you. This is actually nothing really "new". Patronage from politicics is something that is publicly criticized but is quite common among politicians and business.

It's too bad no one in government will ever step up to undo the citizens united ruling. At least we didn't have legal-but-opaque bribery, prior. The difference is substantial when people can't report on where campaign financing comes from without someone first talking too loud about it in a public setting.

The good news, at least, is that Citizens United was only a legal ruling and can be overturned by another ruling. Laws are much harder to undue, with rulings we don't need anyone in the government to step up (other than judges trying the case).

> The idea that he'll just be above the law is fearmongering hyperbole though

Is it? If there is anything the 45th and the aftermath has shown is that there are people clearly above the law. And even without the 45th, Musk himself has escaped justice many many times - especially the SEC whose explicit orders he openly defied multiple times.

Yes I think it is, and I think the rhetoric around Trump is hyperbole and fearmongering too.

Not that you can't criticize them, I just don't see exaggeration being interesting or helpful there. Also I think caring about certain corruption or conflicts of interest when it happens to politicians one disagrees with is fairly easy to be seen as being divisive or politically motivated even if it's not. I thought that wheeling out the architects of the Iraq war to denounce Trump's corruption/incompetence/bad foreign policy/etc was particularly ironic and sad, for example, even if they might have been technically correct.

Musk isn't going to be immune to federal regulators. I'm sure he'll get the kinds of favors that come with buying politicians as all the rest of them get though.

> Yes I think it is, and I think the rhetoric around Trump is hyperbole and fearmongering too.

Well just reading through Project 2025 is very sobering. It's not like old times where what they wanted had to be read through the lines any more, it's right out in the open what they want to do - and even getting a quarter of their plans actually passed through is a very, very troubling perspective.

> I thought that wheeling out the architects of the Iraq war to denounce Trump's corruption/incompetence/bad foreign policy/etc was particularly ironic and sad, for example, even if they might have been technically correct.

A sad consequence of people no longer debating policy on a shared common ground based on facts, but on tribalism, lies and propaganda instead.

I shouldn't have brought up Trump, the subject never goes anywhere useful in an online debate. That was just my opinion, and other opinions and fears are not invalid.

> > I thought that wheeling out the architects of the Iraq war to denounce Trump's corruption/incompetence/bad foreign policy/etc was particularly ironic and sad, for example, even if they might have been technically correct.

> A sad consequence of people no longer debating policy on a shared common ground based on facts, but on tribalism, lies and propaganda instead.

Yep. When they do that it does make you wonder who shares common ground with whom, and who spreads lies and propaganda about what.

I just seriously wanna praise, first without commentary, that you just said >other opinions and fears are not invalid

And then for commentary: I'm the sort of person "intense" enough that I'd want to pounce even on your comment about not bringing up trump, but that overall phrasing was such a perfect way to communicate what you mean, and your good intentions, to me when you at first read (to me) as apologist and unwarrantedly dismissive.

That's all, often in these times it feels hard to even achieve this basic level of communication across personal and political differences. I really admire it.

How can you determine a conflict of interest if you don’t know what he’s been appointed to do?

Because I heard he was going to have some advisory or executive capacity on government operation. There's a significant conflict of interest there if he's running and owning these companies at the same time.

Its hard to determine a conflict of interest when the role isn't clear though, and the problem there is that everyone can really go off of what they heard through the grape vine.

If the role truly is advisory I wouldn't personally see that as a conflict of interest. Regulators are often asking for advise from those they are meant to regulate without it getting flagged as a conflict of interest (for better or worse).

I don't think it's that hard to determine. He has big companies involved in significant regulatory actions and oversight, he would stand to gain a lot by influencing things slightly in his favor. Sure, taken to absurdity everybody in government has a conflict of interest because they are alive on the same planet and have heir own views on things, but for the case of someone like Musk it's pretty clear.

Politicians and bureaucrats can and should consult with the people they govern of course. The "proper" way to do that would be via reasonably open and transparent process that is open to interested parties so competitors, customers, unions, scientists could have their say.

Again I'm fully aware this isn't how things actually work, so I'm not saying Musk is really doing anything outside the norm in American politics by buying a seat at the table. He's just being slightly more open about it than most of them.

I wouldn't be opposed to going after such situations as corruption or conflict of interest issues, but that's going to be a big can of worms.

From the FAA and Boeing to multiple health agencies and pharmaceutical companies, there are a ton of advisory type roles that involve industry leaders "recommending" policy. I'd be surprised if Musk ended up at the top of the list when sorted by impact, counted either by financial impact or number of rules and regulations impacted by industry.

It used to be that even a WHIFF of conflict of interest was treated as "no smoke without fire, better divest".

Carter placed his peanut farm in a blind trust to avoid precisely that - sadly, we have seen a complete erosion of norms, standards, and morals in public life.

He hasn't been appointed to any government agency since DOGE is not a government agency. DOGE has no power to fire anybody. All they can do is make recommendations.

Starlink is the most important military weapon in the world right now. Those civilian organizations have no say when state security is at hand. It's like disarming nuclear rockets because some green guys care about birds. Will not happen.

Starshield != Starlink

Parent should have said "SpaceX", Starlink is just a subsidiary.

SpaceX has more military applications than Starshield alone. For example, SpaceX's assembly line will be pumping out (eventually) a rocket a day. That's the plan.

From a military perspective, Starship is supposed to be able to send 100+ people on long space trips. If that is instead to deliver troops to other parts of the planet, I'm sure hundreds could be packed in. Imagine a fast deploy with parachute capability for personnel and cargo, just as with planes, but with immense range and deploy speed.

You may wonder why, but aircraft carriers and their fleets are considered less usable as deploy platforms, due to increased vulnerability. If the US continues to withdraw from the world stage, its ability to deploy could be affected by a reduction in 'friendly' regional countries and thus leased bases. I don't see any issue with this now, but once a large conflict breaks out, who knows... and this could vastly enhance Starship or equiv as a deploy platform.

I'm sure some reading this will balk at "large war" and "never happen" and so on, but Starshield is an example of a platform for such a large conflict. So considering the use of Starship itself as a lightning speed, emergency deploy platform is important.

There are all sorts of gotchas, such as being shot down, but of course those same issues exist with planes or ships.

Frankly, with the state of AI, the close-to-real Android + military robots, along with drones, Starship would be best served by mass fly-over and deploy of 100k small drones, or hundreds of military robot platforms, or.. well, lots of things.

This really isn't about Starship of course. It's just that we've gotten to the point where this sort of platform is very usable. I can't imagine sending in a large-cost asset like this for general troop deploy, but I can for special ops, weapons platforms in low-risk flyovers, and a variety of other use cases.

And in times of war, things get nationalized too.

Interesting thoughts on the logistics side.

Starship as passenger transport point to point on Earth is a nonstarter most likely, even for civilian applications.

It looks just like an ICBM, because it is an ICBM. I doubt Russian and Chinese air defence forces will wait for them to land and see if it’s full of people or plutonium before launching a retaliatory strike.

Your missed the part about parachutes. Also about flyovers. No landing would happen.

(We've been dropping people and gear for 100 years by plane. And yes, it can be done with Starship.)

On the side of civilian transportation, there is nothing to stop normal passenger planes from having nukes on board. The shape of the object is irrelevant.

Communication is key.

"National security" obviously gets significant concessions from regulators. That doesn't mean military and adjacent industries or significant industries and works are above the regulators, it just means the necessity of the activity and input from military and other interested parties would be duly taken into account by regulators.

That's not unique to SpaceX and I don't think that's wrong as such, although people argue that military interests in general get too much leeway.

They won't cease to regulate, no.

But you may have noticed companies like Boeing getting white-glove treatment from regulators.

You know, deciding that their competitor's cheaper aircraft should be subject to a 300% tariff. Not burdening them with too much scrutiny about whether that modified aircraft should keep the same type rating. Taking their word for it when they say every aircraft has 100% of the door bolts installed. If they have broken some regulations, maybe giving a $150 billion company a $250 million fine.

Not ceasing to regulate - just regulators with broad discretionary power exercising that discretion in line with the will of the politicians who appoint them.

Kind of yes. I mean Trumps Supreme Court changes the rules because they interpreted something differently. So any political or government agency is all vibes based to me. Anything other interpretation is rose tinted.

Safe to say? You sure about that?

I’d say it’s just as likely that six months from now there will be a falling out, Musk will be called a pathetic loser, government agencies will be turned against him, etc.

If past behavior is any kind of indicator, it’s more likely than not. I would not be surprised if we see Musk doing a perp walk within 12 months.

There can only be one Main Character, and Trump doesn’t share the spotlight with anybody.

Elon will fuck up and his money won’t save him from what comes after that.

Great fanfic material

Elon knows well enough to act as a supporting character. He can't run for presidency anyway given he's not a native, and no other political post would be interesting enough. There'll be no falling out.

Trump is not a native either, his family is German and Scottish (just checked). So far as I know a native has never held the presidency in the US.

What silly political posturing. Native-born is the specific reference, and a perfectly valid one. By your logic, if the current descendants of people who have been here for many centuries by now aren't natives, than vast parts of the world's population are also not natives of the places where their families have lived for centuries.

Why not go further and say that the "natives" also aren't natives since they also migrated to the Americas over the Bering land bridge?

People were here, and Europeans showed up and killed nearly all of them, claimed all the land as their “manifest destiny”, and proceeded to subject the few original inhabitants they didn’t kill to lives of desperation.

Call that whatever you want.

  > By your logic, if the current descendants of people who have been here for many centuries by now aren't native.

It's not my logic. The term Native American has an agreed, standard meaning. Trump's family does not meet that meaning. My comment was not an attack on your political views.

Are you suggesting that being born in a place makes one native? I'll accept that definition. Now go convince the rest of the world to update their definition of Native American.

And as you apparently agree, the term native, in reference to someone born in a particular place (regardless of ancestry) also has a standard meaning, which millions of Americans who are natives of the country meet. It also happens to be a good definition, because it helps fight against the kind of idiotic racism by which the descendants of immigrants (especially those who are non-white) still get labeled as foreigners despite being native born.

Again, silly, pedantic social justice posturing because the context of the comment about Musk not being native is obviously in reference to this, not native American history.

As the other reply above states, the first nations, as they're called where I'm from, often suffered terribly at the hands of white settlers historically, and there's no honest way to deny that, but it's a separate matter from discussions about what makes a person American today.

Parent didn’t say Trump is a Native American, just a native (of America, from context). That’s a perfectly normal use of the word ‘native’, and you don’t win consciousness points by pretending otherwise.

By that logic, nobody is native. The supposedly native people did not originate in the US, but migrated from Asian.

The recent U.S. Supreme Court ruling that addressed Section 3 of the 14th Amendment and declared that they're only enforceable by Congress really opens a can of worms.

It sounds to me like the 22nd Amendment and the Article II natural-born citizenship requirement are also only enforceable by Congress. If you're making your plans based on a theory that Congress is able to do the right thing, or that the Supreme Court is not a hyper-partisan institution, I would have a strong C because plans A and B are pretty questionable.

For those who forgot, Musk joined Trump v1.0's advisory council in December 2016 [a], and resigned from it in June 2017 [b]. All of this played out once before.

[a] https://news.ycombinator.com/item?id=13175928 ("Trump Names Elon Musk, Uber CEO to Advisory Team – TheHill (thehill.com)", 92 comments)

[b] https://news.ycombinator.com/item?id=14465667 ("Elon Musk quits Trump advisory councils, saying, 'Climate change is real' (latimes.com)", 4 comments)

Trump is likely to have an entirely unexpected terminal medical event before his term is over.

The nation will mourn his heroic patriotism. Then business will carry on as usual, only more so, with a more compliant leader.

Now that Starlink owner Musk effectively runs the US government from Trump's ear no divestment of any kind will happen that negatively impacts Mr. Musk

And this is why we can't have nice, cheap things. Instead we usually get that "cost" pressure solved by giving it on a silver platter to worker-rights-leading China.

There has to be a better way to prevent abuses in the market without crippling it. But following from that, at what point did we assume this kind of (monopolistic) abuse would happen automatically anyways? I haven't seen it yet, so let's maybe hold off till it happens?

Maybe one day X will host all sorts of government-unapproved content on satellites that are free from US jurisdiction and control? @Elon, do this now, they'll come for you eventually.

it is a fascinating outcome when a vertically integrated monopoly is the cheapest option, and best consumer value. The challenge is figuring out if the firm is really providing the best value, or just a local minimum.

My understanding is Standard Oil provided good service for low prices in most cases. It's not always the case that monopolies provide super expensive or bad service.

Yes, hence why it lasted for over a decade even with >>50% of the market in many jurisdictions.

> vertically integrated monopoly is the cheapest option

I would like to remind you that you can use google, gmail, google maps, google drive and a bunch of other services for free (and the best consumer value even if accounting for their data gathering).

If China where to compete they also need LEO satellites that is just over China for a couple of minutes, that have to take a long round trip around the word to give a couple of minutes access again. That’s why you need a huge constellation.

It’s either selling to the whole world or nothing. If you don’t want to go for slow GEO stationary.

I suspect a China-based constellation could probably serve most of africa, much of asia, most of south america, and some of the middle east. Maybe Russia too.

Not North America or Europe.

In fact, it probably has similar coverage population-wise to what a USA based company can offer.

It's not actually necessary to use reusable rockets to get at a similar cost per launch. Long March 5 is at 2.8M$/ton (so less than SpaceX commercial price).

Also, 15M$/launch is not widely believed to be correct. There is much creative accounting SpaceX could be doing with Starlink (is at-cost account for booster depreciation? If so how, since we don't know how much reuse a booster can be expected to give? Or is it just the cost of refurbishment?), and since the last statement where Elon claimed 1000$ per kg actual cost, SpaceX had to raise their prices, claiming it was due to inflation - is that accurate? Most estimates I've seen are that the cost is 20-30M/launch, which would instead give 1.1-1.7M/ton.

So, it's a big advantage but not an insurmountable one.

Boosters don’t depreciate. They are actually considered more valuable with more successful launches.

They do depreciate, even if the later launches are more valuable, that increase in value is marginal compared to the per-launch capital cost. Airplanes, cars, buildings, everything depreciates.

What do you mean by per-launch capital cost? Maintenance? The increase in value of a F9 booster after use is more than the near-negligible per-launch maintenance cost.

"Depreciation" and "capital cost" reflect the fact the vehicle has a maximum life, even given maintenance.

Imagine if I buy a $200,000 Lamborghini which, with regular servicing, will survive 100,000 miles.

That means for every mile I drive, not only am I paying for fuel, and insurance, and tyres, and servicing - I'm also paying, on average, $2/mile in depreciation.

And sure, the "true" value chart might not be linear. Maybe there'll be a sharp drop when the car ceases to be brand new, or a bump in value when it becomes a classic. But so long as it's worth $200k at 0 miles and $0 at 100k miles, the average cost of a mile must be $2.

The statistics here are inverted. The main marginal cost of a launch is the risk of loss of payload which the customer must insure against. The risk of loss of payload actually goes DOWN with more launches, making costs cheaper the more a booster is reused.

It’s as if your car gained value with every mile driven.

The fact that customers launching exceptionally expensive payloads (the US space force, for one) tend to demand new boosters is not consistent with this.

But even then, it doesn't change that the booster has a maximum lifespan and/or eventually increasing repair and therefore depreciation - we are working on an amortized basis.

> The fact that customers launching exceptionally expensive payloads (the US space force, for one) tend to demand new boosters is not consistent with this.

That used to be true. It no longer is.

They don't anymore. They actually require the use of flight proven boosters for important payloads now. Astronauts too are usually sent up with reused boosters, for safety.

We don't know what the maximum lifespan of these boosters are. There are workhorses that have over 20 launches under their belt, and no sign of deterioration. Obviously at some point something will give, but we're not there yet.

> The risk of loss of payload actually goes DOWN with more launches, making costs cheaper the more a booster is reused.

I don't think that's true. All of the F9 failures[1] have been due to the 2nd stage. And that's new every time.

---

1. CRS-7, AMOS-6, and Starlink 9-3

This furthers the point, does it not?

> The increase in value of a F9 booster after use is more than the near-negligible per-launch maintenance cost.

Do we have anything proving this besides the self-serving word of a privately owned company?

I'm not saying it's false, I have no idea either. But there's a lot of highly specific speculation going on here, based on no reliable source.

Yes, insurance rates for satellites launching on reused boosters go down the more flight tested the booster is.

When Starship starts launching customer satellites, it's possible that even the price for customers would be cheap enough to launch and maintain a LEO constellation. But competing requires massive innovation in cheap and fast production of satellites which are very energy efficient and highly capable. Especially, the technology for starshield protecting against cyber warfare in space and the direct to cell capability. This would be the main reason for not emerging a viable competitor for some time. SpaceX has innovation advantage in the satellites, manufacturing, dishes, base stations, software apart from the launch capability.

Is it really required to be an LEO constellation? It would certainly be good, so that eventually the satellites deorbit and contribute less to Kesseler syndrome risk. But some nation state might chose to not care about that and deploy at higher and more stable orbits.

LEO means less RF power and lower latency.

Higher orbits also mean more latency and slower speeds.

Yes. We had satellite internet services before and they sucked bad due to latency. Space is big and light is slow.

Also, more people per satellite, so less capacity per person.

Yeah in theory China is the biggest potential competitor, having both a space program, a state deeply involved in business, etc. But their space tech is archaic in comparison, using really nasty fuels to blast stuff into space. The reusability of SpaceX's rockets is a feat that is years ahead of the competition - it's been nearly ten years since the first recovered Falcon 9 booster in a commercial launch if my quick fact check is accurate, and no other competitor, private or governmental, has managed it yet. And in a few years they will have a reusable vessel capable of launching 100 tonnes into LEO, at a fraction of the development and launch cost of e.g. the Space Shuttle.

Unless of course Musk's political fuckery ends up dismantling SpaceX. But, Musk didn't do the engineering on these feats, so the knowledge and patents will continue on if he doesn't.

PRC didn't take reusables seriously until a few years ago, really when strategic value of starlink became obvious. They're already making relatively quick progress, as in the expected faster than original catchup mode progress.

Ultimately the issue with simping for SpaceX is that it's still an American company working at American scale. People are conflating SpaceX doing cheap payload advantage at modest scale for actual scale. There's like <20 F9s doing more than 50% of global launches, 80% including starlink. People see 50% and 80%, but ignore that <20 is rookie numbers. Frankly no reason PRC won't have 100 reusables fleet _IF_ demand justifies it (TBH only real justification after megaconstellations is space weaponization). And then like with all PRC catchup, they'll put more than SpaceX lifetime aggregate payload in a few years, and then it won't even be close. Sure Elon can wank about starfactory building 1 vehicle per day, but if there's strategic reasons for it, PRC will be able to build 5 per day at less cost once they sort out the tech stack.

> But Starlink has access to launch at cost, which is $0.86 m / tonne[2]. Which is a huge advantage when launching an enormous number of satellites.

Starlink is a division of SpaceX. So what price it charges/pays itself is only an internal bookkeeping artifact.

Reply on news.ycombinator.com