> Someone will fulfil the need as there is a high incentive to.
Unless the capital cost to compete is too high and the risk of the existing manufacturers undercutting you is very real. Plus it can take 5-10 years or more to build a new fab, debug/iterate your process, then start shipping product.
Markets are prone to natural distortions. This is one form of that. It can be perfectly natural for all potential competitors to choose not to compete no matter how much demand exists.
Frankly I'd expect nationalization of some of the DRAM makers before we see the rise of useful competitors. The more likely scenario is government pressure, up to and including arresting executives, to rattle the cages of the existing players who are way better placed to expand production quickly for relatively low capex. Not that I think any action is likely in the short term. My guess is the existing players are betting on an AI bubble pop so they don't see the use in really expanding capacity only to be left with idle fabs later. None of us really knows.
I think you have to temper the skepticism a bit though.
SpaceX has dramatically lowered the cost of launching things into space. They are still the leader here. They can put a kg into orbit cheaper than anyone, even heavily subsidized state operations (EU and China).
Their order book continues to be full. Every single launch vehicle they roll off the line was pre-sold years ago, including its re-use flights.
I agree that Elon is their biggest potential problem and a big risk but their launch business is sound and wildly successful. If you believe access to space will be a growing segment of the economy in the future it isn't exactly a bad investment.
I remember all the people putting Tesla down when they IPO'd. I bought $4k of stock (all I could afford at that time). Sold $100k of it a few years ago, still have the other half worth near $220k. Their numbers at IPO time were garbage and it wasn't clear they would even survive. Then they started shipping hundreds of thousands then a million cars.
YMMV, consider all sides and make your own judgement. Just be careful about trusting the anti-SpaceX case. Even if everyone is technically correct about them it can still be a huge miss not to invest! The future is not static and if they can put the raised capital to productive use the IPO could end up being a fantastic deal. And FWIW I also agree the largest immediate risk is they are over-valued. Only time will tell on that front.
Tesla IPO'd at 1.7B and is worth 1.4T today. Giving you the benefit of doubt since it would be closer to a 1000x gain rather than the 100x gain that you didn't buy right at IPO, I will point out that there's a world's difference buying in at 1.7B, because there's still room for the stock to 1000x, but there's not much gain to be had buying in at 1.7T valuation.
Even the most highly company in the world Nvidia is less than 3x that valuation, so it's not a good comparison with Tesla's IPO.
In 2024 66% of their launches were for Starlink. So it’s not quite correct to suggest there’s a vibrant external market for their product, a lot of it is sort of self dealing.
> it’s not quite correct to suggest there’s a vibrant external market for their product
There is a very large demand for launch services. SpaceX balances launching customers and launching Starlink. It's not like they give every launch slot to customers and then launches Starlink whenever there's an opening they couldn't fill.
This is missing the point of their valuation. SpaceX will internally use their launch capabilities to build industries that no one else can. Starlink is already their main revenue stream. Starship will open up new realms of possibilities.
At current launch numbers it may not be worth 1.5+ trillion but valuations aren't about current, they're about discounted future cash flows.
It seems logical that there could/will be far more demand for launch if the price were lower. Prices are quite extreme currently, a standard 3U cubesat (loaf of bread size) is $300k and that's just for orbit.
There could be lots of startups that want to try robotic space mining but launch costs just make that mostly impossible currently so there are only a select few. It's like valuing the Dutch East India company based on the trade volumes in 1603. Of course people are not going to be buying much pepper or nutmeg if it costs them weeks of labor, but build lots of reusable ships, and with each voyage, more people can afford your pepper and nutmeg until it's a common household item.
discounted future cash flows is discounted by risk. There is a lot of risk on growing future revenue is the point.
>seems logical that there could/will be far more demand for launch if the price were lower.
This thesis hasn't played out much in the 10 years since Falcon landed in first 2015.
The non Starlink component of revenue has not massively grown beyond what size the market in 2015 to today. SpaceX isn't lowering launch price to induce demand beyond out being the cheapest just by enough, they would be going lower if cost was the only barrier for more revenue.
It not that businesses aren't possible there at lower launch prices. Starlink is testament that it is.
The problem is that rest of the world is not able to innovate fast enough to take advantage of it even after 10 years. The industry struggles with things like manufacturing satellites at scale or raising money for it, or executing on innovation etc.
What that means for SpaceX is that even if launch costs are cheaper than now, the launch market simply may not grow quick enough for the valuation number to make sense. They would need to enter a lot of new markets directly and be their own launch customer beyond Starlink. This comes with its own set of execution, regulatory and other risks. The data-center[1] in space play is an attempt to do this.
Either DC play or something else, they will need to find and sustain a large business to grow, maybe they will, maybe not.
It is not very clear now and that is a lot of risk so any future cash flow projection has to be discounted heavily.
---
[1] I am not qualified to comment on the technical feasibility, however to analyze the company finances that is not needed, it is just one more risk factor, depending on how you feel you can assign 0 or 1 or anything in between.
> This thesis hasn't played out much in the 10 years since Falcon landed in first 2015.
It did play out: there are many more launches today, it's 5x in 20 years. The 75% of SpaceX starlink launches (which account for nearly 50% of all launches) were quietly financed by their other launch customers, exactly because the real cost to launch dropped so much.
That doesn't mean you're wrong, but you do seem to forget that SpaceX, as its own customer, knows the number of launches is going to rise exponentially. They obviously choose to manufacture for where the market _will be_, while you don't see the market before its there. Which is good for them.
I think you have to temper the glazing a bit though.
These people and their endeavors are thoroughly, irredeemably corrupt. It’s nice you got a taste, but their impact on society has been calamitous, and will take decades to recover (if at all).
It is important to remember that Microsoft created some of this chaos to begin with. Other aspects can be attributed to "the industry didn't understand the value of $x or the right way to do $y at the time". And some of this is "nonsense you deal with when the internet and automatic updates is not yet a thing".
Why did programs overwrite system components? Because Microsoft regularly pushed updates with VC++ or Visual Studio and if you built your program with Microsoft's tools you often had to distribute the updated components for your program to work - especially the Visual C runtime and the Common Controls. This even started in the Win3.11 days when you had to update common controls to get the fancy new "3d" look. And sometimes a newer update broke older programs so installers would try to force the "correct" version to be installed... but there's no better option here. Don't do that and the program the user just installed is busted. Do it and you break something else. There was no auto-update or internet access so you had to make a guess at what the best option was and hope. Mix in general lack of knowledge, no forums or Stack Overflow to ask for help, and general incompetence and you end up with a lot of badly made installers doing absolute nonsense.
Why force everyone to share everything? Early on primarily for disk space and memory reasons. Early PCs could barely run a GUI so few hundred kilobytes to let programs have their own copy of common controls was a non-starter. There was no such thing as "just wait for everyone to upgrade" or "wait for WindowsUpdate to roll this feature out to everyone". By the early 2000s the biggest reason was because we hadn't realized that sharing is great in theory but often terrible in practice and a system to manage who gets what version of each library is critical. And we also later had the disk space and RAM to allow it.
But the biggest issue was probably Microsoft's refusal to provide a system installer. Later I assume antitrust concerns prevented them from doing more in this area. Installers did whatever because there were a bunch of little companies making installers and every developer just picked one and built all their packages with it. Often not updating their installer for years (possibly because it cost a lot of money).
Note: When I say "we" here that's doing a lot of heavy lifting. I think the Unix world understood the need for package managers and control of library versions earlier but even then the list of problems and the solutions to them in these areas varied a lot. Dependency management was far from a solved problem.
I use them when working with sheet metal. They are high-dexterity. Thin and flexible. Steel threads are woven into the fabric. McMaster has a variety of high dexterity gloves - fingerless, insulating, cut-resistant.
Because it's the least offensive. No one can yell at you or claim the color is "dated" if you stick to white/gray/black and a touch of stainless steel.
Sad to see how many houses are painted gray or beige in SF now. We are going to paint ours purple in protest.
The REILs are part of ARFF training. Pilot training on it is also clear. The system is automated. It plots the direction and speed of anything approaching the runway and predicts a conflict. If the REILs are red it is HIGHLY likely there is a conflict that is missed by human error and you should not proceed without confirming. Don't just confirm cleared to cross, explicitly tell the controller "XYZ tower we have red runway entrance lights. Please confirm runway XX is clear".
The system is smart enough that if you get red bars to cross for an airplane departing once it passes your position the red clears because it knows the airplane is already past you. It is not dumb - it was deliberately designed to minimize false positives so everyone would trust it otherwise they might ignore it when it really counts. (AFAIK it very accurate in fact so the firetrucks weren't crossing because they distrusted the red lights).
This is just like all aviation incidents and indeed most incidents of any kind: the holes in the swiss cheese lined up.
The emergency aircraft couldn't find a free gate, creating a massive distraction for ATC, airport, et al. This is probably the primary domino that started the sequence. Had a gate been free this incident would not have happened. One big hole lined up.
Normally the aircraft would visually see the truck or the truck would visually see the airplane. But it was dark and rainy. Another hole lined up.
Everyone involved was rushing because noise abatement requires the airport to close at a certain hour. Thus everyone wanted to take-off or land before that shutdown. Another hole.
Normally the controller wouldn't issue the clearance to cross or their supervisor monitoring behind them would notice the error and override. But the controller and/or supervisor were distracted by the emergency. Another hole lined up.
The controller realized the error and issued a stop command but the fire truck proceeded anyway; they may or may not have heard the transmission. Another hole lined up.
Then someone else decided to jump on frequency during this busy time (we don't know who just yet) which may have prevented the controller's stop and/or go-around commands from being heard (another hole lined up).
The ARFF crew did not obey the REILs, accepting the clearance. Perhaps they thought the red lights were due to aircraft on short final and they still had time to cross? Perhaps it was some other misunderstanding of how that system works. Another hole lines up.
And the Air Canada jet was not paying attention to the chaos on frequency. There's a reason runway crossings are typically done on tower frequency: so aircraft can hear what is going on. But it was late at night and their brains probably didn't process what was happening. Or they were too close to touching down to have the bandwidth. Another hole lined up.
> The emergency aircraft couldn't find a free gate, creating a massive distraction for ATC, airport, et al.
Yes. And I want to add one more thing to this: the airplane with the "odour" issue was kinda ambivalent about the danger. They deemed it dangerous enough to declare an emergency, and request a gate then later ask for airstairs but not dangerous enough to pop the slides and just evacuate right there and then. I'm not saying this is wrong. Obviously they were evaluating the situation as new information was coming in. But it increased the workload of the ATC. They were trying to find a gate, and etc. If it was a clearer "mayday mayday mayday, aft cabin fire, we are evacuating" that might have been paradoxically less "work" for the ATC. Or at least more of a "practiced" scenario.
> Perhaps it was some other misunderstanding of how that system works.
Yeah. That's a big one. Total speculation but maybe they thought the airplane with the "odour" issue was keeping it red?
Is that not an admission that his poor management and lack of vision cannot foresee any way to make profitable use of those people's labor?
A company raking in 5-6 billion per year can't find any profitable bets to make? Possibilities to invest in? All they can do is cut?
LOL. If you're that bad at capitalism then please resign and let someone else give it a try.
Reminds me of PG&E. So bad at being a for-profit electric company they need constant state handouts to guarantee profits. They made bad contracts so they need a PCIA fee for not selling me electricity. Hedging? Severing contracts? Arbitrage? Forecasting? Never heard of those, now make with the free coin! My son... if you are that bad at capitalism shut it down!
I agree with Warren Buffet's take here. A company that cuts or can only pump dividends is basically saying "we can't figure out how to make productive use of people and/or cash". What an unbelievable joke.
Gamblers are the whales of that industry. The industry is well aware of that and well aware of how much harm they can cause. But their paychecks depend on not knowing so they choose not to.
Same as pay-to-win freemium games. Find the whales and milk them for all you can. For every high-spender who can afford it they know full well the other 99 cannot. They know they are ruining some people's lives. They know they use dirty psychological manipulation tactics. Their paychecks depend on not knowing so they choose not to.
There are many jurisdictions where the companies are not allowed to ban 'winners', but the companies often respond by lowering those users' bet size limits.
No different than big tech and their divisive algorithms. Or big pharma and side effects. Or big manufacturing and environmental harm (including harm to the people living around manufacturing companies).
It is an inherent property of unchecked capitalism to externalise and ignore any unwanted costs. Or on the flip side of that coin, profit from causing damage to others, where possible.
Also there’s a bit of a tragedy of the commons. If one entity is scrupulous that doesn’t mean another will. Obviously if they had any morals they’d see the bright line.
Had an interesting case study where a coworker liked to gamble - he was fairly responsible, kept to his budget and treated it like an expensive hobby he enjoyed- but at the same time, he had someone else handle his retirement investments, which is an unpredictable payoff market where you come out ahead on average. I asked a couple times why he didn't replace gambling with investing and never got a good answer. He was certainly smart enough that he could have had fun with the research and chance.
Then there was a market downturn and his investment advisor had to talk him down from selling in a panic, and I was like "oh... It's not an information problem at all. It's entirely an emotional regulation problem"
> I asked a couple times why he didn't replace gambling with investing and never got a good answer
There's a simple answer, which is that gambling is fun, and investing is not fun (my guess.) Your coworker is a pleasure seeker, like everyone is to some degree.
I think your coworker was quite smart. Investing and gambling are close enough that for many parties they are indistinguishable. I've heard investing described as 'gambling for people with more money'. The biggest difference is that if you have enough money you can legally manipulate the market. If it's your retirement investment that might just be over the horizon far enough to get you into danger. Just having access to that with a habit could already be an issue.
That's true, but if that was what was stopping him he could have got a no fees account and just used his casino budget to speculate on penny stocks instead. It would be like going to a casino that has better odds and that he can access on his lunch break... Ok I see your point.
I'm not sure it's irrational to sell in a market downturn. It's a way to pad your emergency savings rather than try to catch a falling knife later when you're already fired. Of course if you sell more than you need to survive a layoff, then that's probably not smart.
If you dig into JS engine implementations they deal with a lot of the same sorts of things. Simple objects with straightforward properties are tagged such that they skip the dynamic machinery with fallback paths to deal with dynamism when it is necessary.
A common approach is hidden classes that work much like classes in other languages. Reading a simple int property just reads bytes at an offset from the object pointer directly. Upon entry to the method bits of the object are tested and if the object is not known to be simple it escapes into the full dynamic machinery.
I don't know if those exact techniques would work for Python but this is not an either-or situation.
See also: modern Objective-C msg_Send which is so fast on modern hardware for the fast-path it is rarely a performance bottleneck. Despite being able to add dynamic subclasses or message forward at runtime.
This is just an analogy but in Swift String is such a commonly used hot path the type is designed to accommodate different backing representations in a performant way. The type has bits in its layout that indicate the backing storage. eg a constant string is just a pointer to the bytes in the binary and unless the String escapes or mutates incurs no heap allocation at all - it is just a stack allocation and a pointer.
Javascript implementations do their own magic since most objects aren't constantly mutating their prototypes or doing other fun things. They effectively fast-path property accesses and fallback if that assumption proves incorrect.
Couldn't python tag objects that don't need such dynamism (the vast majority) so it can take the fast path on them?
It already has a fast path, from (I think) 3.11. If you run `object.x` repeatedly on the same type of object enough times, the interpreter will swap out the LOAD_ATTR opcode to `LOAD_ATTR_INSTANCE_VALUE` or `LOAD_ATTR_SLOT`, which only makes sure that the type is the same as before and loads the value from a specified offset, without doing a full lookup.
Unless the capital cost to compete is too high and the risk of the existing manufacturers undercutting you is very real. Plus it can take 5-10 years or more to build a new fab, debug/iterate your process, then start shipping product.
Markets are prone to natural distortions. This is one form of that. It can be perfectly natural for all potential competitors to choose not to compete no matter how much demand exists.
Frankly I'd expect nationalization of some of the DRAM makers before we see the rise of useful competitors. The more likely scenario is government pressure, up to and including arresting executives, to rattle the cages of the existing players who are way better placed to expand production quickly for relatively low capex. Not that I think any action is likely in the short term. My guess is the existing players are betting on an AI bubble pop so they don't see the use in really expanding capacity only to be left with idle fabs later. None of us really knows.
reply