>Yes, the only e-mail I got was a credit note giving my money back.
That's great news ! They don't have nearly enough staff to deal with support issues, so they default to reimbursement. Which means if you do this every month, you get Claude for free :)
>The first move in the coming WWIII, where the emperors try to expand their empires militaril,y will be to wipe out any orbit with Starlink satellites.
I find this highly unlikely, given Starlink is soon to reached 10k satellites and will continue to grow. Why expand 10 000 ballistic missiles to bring down one of many communications networks ?
If it's WWIII, and you're using ballistic missiles against satellite constellations, then either:
- You are not targeting individual satellites; you're setting off nuclear warheads in space, and relying on the EMP to disable all satellites within a large radius of the blast - https://en.wikipedia.org/wiki/Nuclear_electromagnetic_pulse
or
- You're nuking the ground-based command & control centers for those satellites. Again, nothing like 10,000 missiles needed.
(Or both.)
To target 10,000 satellites directly, the "obvious" weapon would be a few satellite-launch rockets, lofting tons of BB's (or little steel bolts, or whatever) - which would become a sort of long-duration artillery barrage shrapnel in orbit.
> - You're nuking the ground-based command & control centers for those satellites. Again, nothing like 10,000 missiles needed.
With Starlink's peer-to-peer capabilities, hitting every single ground station and keeping the satellites from working through new ground stations may actually be quite difficult.
Starlink orbits close enough that they're looking into offering LTE coverage from "space". You don't need a giant dish to access the satellites, which means building new ground stations and reprogramming the network from an unassuming-looking ground device to use them is quite feasible.
The paths of the satellites are rather predictable, though, so your shrapnel attack executed with some precision should clear out enough of them.
The moment you launch a nuke (even if just to set off an EMP), you can expect nukes to come your way in retaliation before your nuke even detonates. Unless whatever war is going on has already gone full nuclear, I don't think nuclear weaponry is a viable move to take out satellites.
> With Starlink's peer-to-peer capabilities, hitting every single ground station and keeping the satellites from working through new ground stations may actually be quite difficult.
Yes-ish? I was thinking the command & control facilities - far scarcer than the (probably unmanned) StarLink-to-Internet Backbone connection ground stations.
> The moment you launch a nuke...
Yes-ish. The (great-)^n grandparent comment posited WWIII starting, and the nukes flying at scale. Between the widespread obliteration of ground-side infrastructure, ground-side EMP damage, and very likely EMP in space - I'd assume that Starlink would quickly go down. Plus, the ionosphere could become opaque to Starlink's radio frequencies. Finally, the ionosphere's upper layers might expand enough (due to nuclear detonations in or near space) that the orbits of the Starlink satellites started degrading very quickly.
With how easily any major space power could set off "small n" nukes in space during a major crisis, to knock out satellites - I would not rule someone doing so. The responsible parties need not claim responsibility. And sane leaders might hesitate to go full nuclear in response.
The BB idea doesn’t really work either- if they are in orbit they circle with the satellites and don’t hit anything, if they are at different speeds they are in different orbits and fly above and below the satellites and miss, if they cross the orbit SpaceX just moves the satellites to miss.
"Circle with the satellites" is not how orbits work. Do a Google image search for satellite ground tracks, and observe how those tracks repeatedly cross each other. In LEO, a 90 degree orbital crossing represents a relative velocity of >10km/s. (Normally, collisions do not happen because the satellites are under control, and everyone is making ongoing efforts to avoid collision. Kinda like how cars & trucks normally don't hit pedestrians.)
Bottom line - a "3 tons to LEO" satellite launch vehicle could put ~10,000,000 untrackable little metal objects into orbit, crossing satellite orbits at lethal velocities. Trivial methods, such as dispersing the BB's with small explosive charges, could randomize their individual orbits.
The satellite operators have very good reason to be concerned about such "low tech" anti-satellite weapons.
> Kessler syndrome means you don't need to hit all 10k yourself
Kessler is useless for LEO constellations. The timeframes of the cascades exceed the useful lives and dwelling times at those altitudes.
I am not aware of a military solution to prompting a cascade over even a limited area. Instead, you’d use repeated high-atmosphere nuclear detonations to fry birds in a region.
That depends on how you define risk. If it means the probability of a collision, then you'd be correct. But if a collision does happen, the consequences will be worse than being in the same orbit. Based on an oversimplified model, debris in orbit is likely to have low relative velocities with respect to an intact satellite in the same orbit, since a large deltav would change the orbit. (It's not as simple as this, but it's good enough in practice.)
This is actually what asat weapons take advantage of. They usually don't even reach orbital velocity, just like ballistic missiles (of course, there are exceptions like the golden dome monstrosity). The kill vehicle just maneuvers itself into the path of the satellite and lets the satellite plough into it at hypervelocity.
I remember a short story about Canada preventing total global annihilation in WWIII, by deliberately triggering Kessler syndrome. My google-fu is failing me though.
Ah, so I should have replied with "Yeahhh bitch, science!", because that's a reference to the Breaking Bad meme? That's how we're discussing things now?
Or why try to shoot them down when you can also go to the command center and turn them off? Or do a targeted strike on said command center. The sattelites are plentiful and redundant, but the network will collapse very quickly when they're no longer controlled from the surface.
In fact, if SpaceX can no longer do any launches due to whatever reason, Starlink will no longer be feasible after a few year - if I'm reading it correctly, the sattelites have a lifetime of only 5 years, meaning they will have to continually renew them at a rate of 2000 new sattelites a year.
You could launch some missiles, blow a few satellites into smithereens, and gradually over the next few months they would take out the others. That's a poor kind of war weapon. An effective weapon is one where you can inflict damage continuously, and are able to stop immediately upon some concession. If you can't offer to stop in return for concessions, you won't get any.
Its not really that easy, to cause such a chain reaction, specially if the other person reacts.
And its also really expensive, each sat you take down costs you far more then what you hit. So unless you can actually cause a chain reaction its a losing proposition.
Not really. That’s more science fiction than reality. You should try some Kerbal Space Program and explore how orbits are affected by thrust = collisions, in different directions.
As soon as a satellite is hit the rest of the fleet can start thrusting and raise their orbits to create a clear separation to the debris field.
Following such an attack the rest of the fleet would of course spread out across orbital heights and planes to minimize the potential damage done by each hit, leading to maximum cost for the adversary to do any damage. Rather than like today where the orbits are optimized for ease of management and highest possible bandwidth.
Imagine using a rocket and blowing up one car on a highway - how many other cars will actually be affected? How many cars on other highways will be affected?
Blowing up something in the same orbit as the targets isn't an effective strategy. The explosion disperses the fragments into different orbits that intersect the original orbit only at one or two points. And even if some of those fragments find their targets, the collision velocity will be low (relatively slow).
It will be like getting hit with with shrapnels from a grenade. Depending on how they collide, the target may survive. If you think that grenade shrapnels are fast, you need to understand the 'hypervelocity impact' that happens when objects in different orbits collide, or when an interceptor hits a satellite. Hypervelocity impacts are impacts where the impactor moves faster than the speed of sound in the solid target. What that means in practice is that the debris/interceptor may have hit one end of the satellite and vaporized already, while the other end of the satellite doesn't yet feel the shock and vibration from that impact. That end doesn't yet know about the carnage that's about to hit it in a few milliseconds.
I imagine you could just send a rocket to the specific orbit and just start metering out 100s of thousands of small steel (or whatever) BBs like seeding a yard; seeding a Kessler event.
What kind of pictures can starlink would take? When I look at pictures of starlink satellites, I don't see a camera. Maybe they have one, but if we can't see it, it is most likely useless for observation, except for taking pretty pictures of the Earth, or maybe other passing satellites.
Spy satellites are more like space telescopes, but pointed at the Earth. As an example, Hubble is designed after a spy satellite, the "camera" is pretty massive and obvious.
Starlink can probably be weaponized for a variety of thing, like for communication, obviously, but I don't think earth optical observation is one of them.
It's also been used for regime change attempts - part of the internet that's harder to shutdown, though apparently jamming GPS currently appears to be quite effective.
Looking at the price of industrial lasers, right now the only thing stoping a random 3rd world terrorist cell from being able to afford to destroy all of them is the adaptive optics to compensate for atmospheric turbulence.
Well, that and the fact that so much of the stuff on Amazon etc. that's listed as "welding laser" is actually a soldering iron.
I think you severely underestimate the amount of power you would need to damage some Starlink satellites in the 4 minutes they would be visible while tracking them at ridiculous speeds.
Modern welding lasers are pretty powerful. Hard part is focussing them, especially given need for adaptive optics for atmospheric turbulence. Movement is almost irrelevant, they only deviate from a fixed path while their engines run, and even then those engines are pretty weak.
I'm saying the only real limit now is the adaptive optics.
When people attack satellites with lasers, what they're trying to do is blind surveillance sats. To actually physically damage a satellite would take enormous amounts of power and accurate tracking tied into powerful radars. That's something a state might do, but too many resources are involved for terrorists.
I don't think that's necessarily true. I think with modern lasers the power to get permanent damage is there.
I don't think powerful radars are required either. The satellites will probably reflect the laser. At 4000 km this is 26 ms, so you would probably be able to use the laser itself for the adjustments.
I'm saying the only limit to damaging them now is the optics.
Radar is pointless here. Both for crude and precise positioning. For crude, we already know roughly where the satellites themselves are because those are well-advertised.
For precise, you still wouldn't use radar, physics prevents high enough resolution ever. Even for tracking, angular resolution is k(λ/D), i.e. you care about aperture size in wavelengths, and radar uses wavelengths much much larger than visible or IR laser light. But even with arbitrarily large equipment, you get a spot size ~= wavelength, which wastes a lot of power as the wavelength is much larger than the necessary spot size for a critical component on a satellite.
So you'd use optical targeting and tracking, i.e. you'd look through the exact same system that the laser also fires through, with the exact same adaptive optics, and say "this specific point on this satellite".
The hard part is focussing a spot size order of cm scale (it can't be the same size as you find in a welding system for same reason radar is useless, k(λ/D) gets you ≥300m telescope and that's obviously a no). This requires adaptive optics (and also a wide telescope). Adaptive optics is the really hard part here.
Getting a 12 kW industrial laser is relatively easy, and putting that power into a spot on the joint between the PV and the main body, that has a decent chance[0] of weakening or severing it while also causing catastrophic loss of control, even with just a few minutes over the horizon. Weakening is still important, see all 9/11 memes about steel beams and why they miss the point. Severing is plausible but only because of space design constraints, see [0] again.
As I understand it, PV cells themselves have a much lower threshold for catastrophic damage, a 12 kW system is basically guaranteed to cause irreparable damage to that even in a few minutes even though the spot size here is much much larger than you'd find in a welding system.
The prices I see for 12 kW industrial lasers are significantly lower than the estimated cost per missile for most of what the Houthis used to attack shipping last year, and they fired quite a lot of those missiles.
[0] can't say for sure without detailed plans that it would be genuinely insane[1] for me to have access to; but do consider that everything in orbit is mass constrained, even with SpaceX pricing, and designed without expectations of e.g. wind or needing to support its own weight, so the thickness of structural elements is likely much lower than you'd expect from anything you see on the ground
[1] the world is currently going insane, so if it turns out they are available, either deliberately or via a leak, this is just more evidence of insanity rather than a contradiction
Why would you not bend the key ? There's a reasonnable chance Trump is gone in less than 3 years and MAGA tears itself apart. Better bet on that than try to make a stand now and lose everything.
>The flip scenario: AI unlocks massive demand for developers across every industry, not just tech. Healthcare, agriculture, manufacturing, and finance all start embedding software and automation.
I find this one hard to believe. Software is already massively present in all these industries and has already replaced jobs. The last step is complete automation (ie drone tractors that can load up at a hub, go to the field and spray all by themselves) but the bottleneck for this isn't "we need more code", it's real-world issues that I don't see AI help solving (political, notably)
I tend to agree with your assessment. The increase in demand cannot possibly equal the loss from AI.
Given projections of AI abilities over time AI necessarily creates downward pressure on new job creation. AI is for reducing and/or eliminating jobs (by way of increasing efficiency).
AI isn't creating 'new' things, it's reducing the time needed to do what was already being done. Unlike the automobile revolution new job categories aren't being created with AI.
Germans were so quick to revert back to paper after COVID that it felt like one of the only reasons they came out of lockdown eventually was to get paper back.
The Gewerkschaft tactics to resist AI is what I’m really interested in seeing.
Agree, people were already worried about the excessive focus on software over physical technology well before LLMs significantly reduced the barrier to entry
Gaddafi was trying to establish a gold-backed "arab" currency system and wanted to sell his oil using it. This was a threat to the US dollar so Obama was very happy to see Sarkozy knock at his door asking to go get the oil themselves lol
>Your job isn’t to complete tickets that fulfill a list of asks from your product manager. Your job is to build software that solves users problems.
While I agree in spirit, when you reach a certain amount of people working on a project it's impossible. The product manager's job is to understand real user problems and communicate them efficiently to the engineering team so the engineering team can focus on engineering.
No. The product manager has to understand the big picture, but when you're working on a team that big, it follows that you're going to be working on a product big enough that no one person is going to be able to keep every single small detail in their mind at once either.
You wouldn't expect the engineering manager to micromanage every single code decision—their job is to delegate effectively so that the right people are working on the right problems, and set up the right feedback loops so that engineers can feel the consequences of their decisions, good or bad. In the same way, you can't expect the product manager to be micromanaging every single aspect of the product experience—their job is to delegate effectively so that the right people are working on the most important problems, but there are going to be a million and one small product decisions that engineers are going to have to have the right tools to be able to make autonomously. Plus, you're never going to arrive at a good engineering design unless you understand the constraints for yourself intuitively—product development requires a collaborative back and forth with engineering, and if you silo product knowledge into a single role, then you lose the ability to push back constructively to make features simpler in places where it would be a win/win for both engineering and product. This is what OP means when they say that "The engineer who truly understands the problem often finds that the elegant solution is simpler than anyone expected".
That's great news ! They don't have nearly enough staff to deal with support issues, so they default to reimbursement. Which means if you do this every month, you get Claude for free :)
reply