is not something that can be done as it just makes them absorb more sunlight and so over-heat.

What about using the same flat panels, etc, as used on [1]stealth aircraft to reduce reflections and so visibility ?

[1] https://en.wikipedia.org/wiki/Stealth_aircraft

Supposing your satellite doesn't radiate anything into the line of sight. It still occults the background. So was that a planet passing in front of the star or a satellite?

Probably you can separate them. But in reality, your satellite will radiate something, somewhere in the spectrum. And that will create edge cases. (Literally: cases where the edge of the satellite clips what the target.)

If a panel's not reflecting, it's absorbing.

It's about time someone clever invented a black paint that didn't absorb!

They're all busy inventing the battery that never needs charging.

They're being very busy inventing non-reflective white paint!

"Stealth" aircraft do have radar absorbent paint but they're primarily designed to reflect radar signals away from the plane in which they arrived, so that the radar station or aircraft that emitted them (or others in the area) don't receive a loud return signal. This usually means reflecting them away from the horizontal.

In the case of these satellites, the transmitter (the sun), the reflecting object (the satellite) and the receiver (ground telescopes) are all at wildly different and constantly changing angles, which makes dealing with these reflections a much more difficult problem.

Perhaps, we need an international study to determine what numbers of satellites would be required to actually benefit us earthlings while permitting astronomy to continue.

I can see a point being reached where the only way to easily see past the fleets of sat's will be to subscribe to 'Astronomy as a service' from astronomical sats like Hubble launched by the new commercial generation of astrogarchs.

"Perhaps, we need an international study to determine what numbers of satellites would be required to actually benefit us earthlings while permitting astronomy to continue."

Except:

1) It would probably be set up/controlled by Americans who would have vested interests in ensuring that they could put up as many as they like, while subduing other countries

2) How would you "police" this? And what is to stop some country selling their allocation (for many shekels) of satellites (that they were allowed to launch) to some big bucks company like Amazon/Spacelink/OneWeb

3) Putting more satellites into orbit, and NOT having a fool proof (and internationally recognised) way of de-orbiting them makes little sense and for now, the chances of further collisions by wayward orbiting space debris is going to increase to the point where it might mean we cannot launch anything soon, if too many fragments/shards are orbiting at 20,000+ mph.

I believe most of these are "low orbit", ~ 600 km, by necessity - to keep the signal flight time short. At that altitude, without active boosting from on-board propellant, the orbit will decay in a few years. They have on board propellant so they will up for a while.

The plan, apparently, is to deliberately de-orbit them before propellant runs out, so they come down quickly and at preferred locations.

With 40k+ satellites you have one approx. every 250 km apart all over the globe in each direction in the sky. That is simply excessive. No need to study or investigate. It is a bad plan, plain and simple, a bad plan.

> you have one approx. every 250 km apart all over the globe

For a short period... Then one of them will experience a catastrophic failure (given the numbers it's unavoidable), and the ensuing chain reaction of debris collisions will quickly turn them all into an all-compassing cloud of shrapnel, angrily orbiting earth until many years later it eventually sinks into the atmosphere.

With such a density, all it takes is one single accident to quickly take them all out: It's like shooting fish in a barrel. And given those are not handcrafted high-end birds, but cheap & cheerful disposable satellites, something is bound to happen sooner or later: A missed orbit, a satellite unable to hold its position, any technical problem with one of them will do, and when there are some 50000 of them up there, the chances of it happening are very high.

All that schrapnel might cut down sunlight by a few percent and buy us some time with global warming where the cuts aren't deep or quick enough.

Just because one of them lets fly a bit of shrapnel, doesn't mean they all immediately explode. I think you may be underestimating how easy it is to miss a target the size of a satellite at a range of several hundred kilometers.

Let's imagine a solar panel flies off in a random direction. The area of the orbital plane at 600 km is approximately 2.5 billion km2. 40,000 satellites, each filling about 40m2, will fill about 6.4e-10 of that space. That's the chance,per orbit, of an accidental collision, and that's assuming the debris flies off in exactly that orbital plane. If we allow for a bit of variation in orbital radius, it rapidly becomes way less likely.

my honourable et4dsbreheate5u86wags makes a fairly common error of overestimating the earthlings, calling this a 'bad plan', because there's no plan.

I'd be in far greater favor of installing fiber to the community, with a proper cell network to provide the citizens access, and hardwired connections for those close enough to the termination/entry point of the national fiber.

It wouldn't pollute the skies, it would provide LOCAL jobs for the populations around the globe, and it would produce LOCAL infrastructure instead of another excuse for some American megacorp to rape and pillage the pockets of the global population...

They won't even do that for rural America - their own backyard.

And yet the States just passed a bill to do so. It is all but guaranteed there will be some group raising environmental concerns when trenching up the countryside to accomplish this.

What if the community is in the middle of nowhere, surrounded by mountains (secluded valley) or even ocean (islands)?

Worse, what if the community in question is moving , like an airliner in flight or a ship out at sea?

Lastly, who's going to pay for all that cabling and installation? Why isn't rural America wired up? There's your reason right there.

I'm quite sure putting some 40k satellites to orbit plus the ground infrastructure will cost way more than laying some fibers. No matter how remote, if you've got electricity and water there is nothing preventing you from adding a bunch of fibers to that same trench.

Now who is going to pay for all that cabling? Well, who pays for all those satellites and rockets and ground infrastructure? This is a stupid question, isn't it. The only difference is that in one case nobody thinks there is "enough" profit to be made connecting the boonies, while in the other they are convinced they can sell enough high-value contracts to rich clients to make the whole thing profitable. Funny enough their marketing argument is those very same unprofitable boonies...

I'm quite sure you're wrong about that.

However, I am too lazy to look it up as well.

Laying some fibers, yes, but laying enough to connect all of them at once? That's the key with the satellites: it's a "blanket" approach able to cover all those rural and isolated communities in one shot. This also helps solve the cost issue. Running cabling to each individual community for little return (not enough customers) adds up, thus the sticks stay in the dark. A blanket approach is easier to spread the costs to all the communities (not to mention the ships and airplanes, which you did not address and are impossible to wire up because they move ).

And look, I lived on an island once: Guam, a dinky little speck of land the size of DC in the middle of the blankin' Pacific Ocean, where the nearest major city is a several-hour flight away (and isn't in the US) . The power plant was local using shipped-in bunker fuel and the water was either shipped in, collected from the rain, or desalinated (using more power). None of them required trenches. Undersea cables are a major undertaking in themselves. At least with places like Hawaii and Guam they have the excuse of military presence to justify government outlaw for undersea communications cables. The rest? They're in the same boat as everyone in the sticks.

I have a philosophy and it goes like this: if you can run power lines to a house, you could also run telephone/DSL service. If you wanted.

Sounds great if you want to email someone else at the other end of Guam.

I recall reading an article I can't find now that described an inexpensive community networking utilizing cell phone technology that would provide internet and enable people to communicate with local cell phones directly bypassing the cell phone companies with a proof-of-concept demo. There are proposals like the following that as far as I know have never been built:

https://dil.berkeley.edu/technology-portfolio/current-projects/village-base-station/

https://arxiv.org/abs/2009.05966

with something like that you wouldn't need to run fiber to every house and would still provide a lot of connectivity.

It would probably be harder to charge people for using a network like that, and would likely be harder for a government that felt in crisis to shut down communication among the populace.

> Guam, a dinky little speck of land the size of DC in the middle of the blankin' Pacific Ocean

Still had long distance telephone, didn't it? So the infrastructure was there, they just didn't chose to expand it to offering internet to the inhabitants. (Given it's mainly a military base, I guess things are somewhat special over there.)

I don't think your "cover all at once" argument stands: Don't forget it needs more than just satellites zooming overhead, it requires building ground stations feeding them, lots of ground stations all over the globe, and those are not only expensive but will also heavily depend on local governments and laws. In your Guam example, where would be the ground station which is feeding the satellites over Guam? It has to be somewhere near enough to actually see those satellites.

IMHO much like cell phone, those constellations are meant to primarily cover the well-heeled customers (Caribbean cruises for instance), the rest of the world will remain an eternal "work in progress", simply because it's financially not reasonable to spend millions just so some crab fishers in the north Pacific can enjoy cat videos between work shifts. You'd never get that money back.

I'm quite sure the only reason all those constellations cover the whole earth is because those darn birds won't stay put. The profitable parts they are interested in covering are quite localized, but well, LEO satellites have that irrepressible need to orbit...

"Still had long distance telephone, didn't it?"

I wouldn't really call it that. It was expensive as hell, too, and often unreliable and unavailable. That should tell you something about the costs involved with laying down that kind of infrastructure.

I don't think your "cover all at once" argument stands: Don't forget it needs more than just satellites zooming overhead"

Sure you need ground transceivers, but it's not like it's new technology. The only new wrinkle is the lower orbit and thus differing logistics: a difference in degree rather than kind versus already-existing technology. Okay, some backhaul would be needed, but you wouldn't need that much, it wouldn't have to be expensive, and by spreading them out, you reduce the amount of backhaul you need at each point, not to mention adding redundancy.

"IMHO much like cell phone, those constellations are meant to primarily cover the well-heeled customers (Caribbean cruises for instance)."

Why not both if the incremental costs are low enough? You hook the rich customers first to cover your costs. Once that's done, all the rest is gravy.

"I'm quite sure the only reason all those constellations cover the whole earth is because those darn birds won't stay put. The profitable parts they are interested in covering are quite localized, but well, LEO satellites have that irrepressible need to orbit..."

Sure, but why not turn a bug into a feature...?

> Why not both if the incremental costs are low enough?

Because the ground station which covers the profitable Caribbean doesn't reach up to Alaska (for instance). And given the density and average income of rural Alaska (towns won't need this), the Alaskan ground station will never pay for itself. Not for the initial investment, not for the operational costs.

Besides, I admit I don't know anything about how they plan to do it, but I don't think you can't feed a constellation of fast-moving satellites zooming overhead with a single biggish dish on a rooftop. You'd most likely need several different dishes to feed several different satellites at the same time, and since those satellites are moving rather fast, you might need tracking of some sort to keep them in focus 180° from one end of the horizon to the other. What I'm saying is the ground stations are certainly more elaborate (and thus expensive) than a tin shack with a big dish and a fiber connection.

Undersea cables are pretty extensive, though. Niue makes Guam look like a humming metropolis, and it's way more remote, but even it has its own cable connection.

I'm not so sure about that. Trenches to install fiber in is not a problem, but that is a lot of fiber to install.

In any case, I think the rural America market is a target only in that they can collect subsidies for it in the early stages. Once up, they can sell to everyone and I would not at all be surprised if urban users exceed rural users in short order. Think about it. One service that does it all, wherever you go, even during international travel. Internet and cheap mobile communications for every commercial ship from shrimp boat to giant cargo ship. Same service in your car, wherever you go, etc.

Well if you live in the middle of nowhere then what do you want internet for? Seems counter-productive. If you want to talk to your neighbor, guys, hamster radio. Get real.

I contend that any house that can be hooked up to power could also be hooked up to DSL. You just gotta believe!

The why is the telephone monopolies do not want to.

The problem with latency is that it is directly related to the refractive index of the medium through witch the electromagnetic wave propagates. The refractive index of of air/vacuum is close to 1.000 (EM waves travel at ~300,000 kilometers per second ; 186,000 miles per second), and for glass it is closer to 1.5 (light travels at ~200,000 kilometers per second ; 124,000 miles per second). So potentially 50% more latency for a fiber installed over the same path length.

So I can see why LEO satellites constellations are so attractive. Dropping 80 millisecond ping time between the east coast of the US and western side of the EU down to nearly 50 milliseconds is very attractive and can not (easily) be achieved any other way.

But on the flip side when a Coronal Mass Ejection next hits earth directly and kills ALL the satellites in orbit (the 1859 Carrington Event took down the entire US telegraph network ), any under ground/water fiber cable should be fully immune (except for possibly the repeaters every 100 kilometres ; 62 miles and the end points, if not electrically isolated before the CME hits). If a large enough CME hits the earths magnetosphere could temporarily collapse, I wonder what would happen to the Van Allen radiation belt ? That may cause some additional unexpected problems that might take a good few years before the dead satellites can eventually be redesigned for a harsher environment and be replaced. And that would be assuming that the catastrophic consequences of such an event is already planned for ahead of time, I wonder ...

In the event of a carrington class event, not having any working internot satellites, or internot cable is going to be the least of your problems

Working out how to repair the power system will be the priority, because no power means no pumping which means no water... no sewerage and no gas supply (unless powered by a small generator fed by a tap on the gas line)

But then replacing about a 1/3 of the worlds transformers and associated switchgear is going to be fun when you find out we dont make that much in the first place...

The bitter irony is that upto about 150 yrs ago , a carrington class event would have ment "look at the pretty lights in the sky" and then everyone would have gone back to bed and carried on normally the next day...

I'm picturing people powering up the relays of the under-gound/under-water global fibre cables with large banks of D-cell batteries and then tapping out Morse code with blinking LED's asking for spare transformers, switchgear and other parts to exchange in barter! Or asking for the other end to relay the message on to the next hop!

radio + batteries/generators are more likely to be used

That would assume that the ionosphere is in a stable steady state. I wonder what a temporarily collapse of the magnetosphere would do to global HF propagation. It could be better or worse. Maybe all those charged particles might raise the overall noise floor a lot, but the ducting could be amazing. It might be swings and roundabouts.

I'm thinking of when (October 30, 1961) the US bombed the ionosphere with a Nuke and blocked global HF communications for a while.

Sorry I mean Hardtack Teak was the one 3.88 Mt (1958-08-01) and then they scaled down to kt soon after that international faux pas which caused a communications blackout that lasted 9 hours in Australia and at least 2 hours in Hawaii.

> Dropping 80 millisecond ping time between the east coast of the US and western side of the EU down to nearly 50 milliseconds

I'm not sure you can simultaneously see a LEO satellite both from the USA and Europe. And if you can, it's certainly not for very long, which means you'll have handshake delays as the following satellite takes over, then the following, then the following.

People, please don't succumb to magical thinking. "Internet from the sky", yes, all satellites will have an internal reservoir of "Internet" they can instantly pour over the poor isolated users... Not. In the real world they depend on simultaneous line-of-sight with an (existing) ground station and the end user. Even if satellites are dense enough for us to live in a perpetual twilight, without line of sight to a working and well-connected ground station they are just useless flying trash cans.

Have a look at this:

https://news.satnews.com/2020/10/14/leo-speed-when-milliseconds-are-worth-millions-an-nsr-insight/

According to this, using inter-satellite linking, you can still get a signal from London to New York some ten milliseconds faster than through fiber due to difference in speed of light. And as noted, high-frequency traders will pay a fortune for a jump on the competition since they could potentially rake in millions from this latency arbitage.

High frequency traders trading in London aren't in New York, they are within a few hundred yards of the LSE. For that very reason.

But events in London CAN affect things in New York and vice versa. Especially in a cutthroat market like HFT where even the tiniest scoop can mean latency arbitrage and getting that narrow window worth millions. Information is ammunition, timing is everything, and coordination matters.

Here's some grist for the mill (searching for "latency arbitrage" can bring up plenty more):

https://www.cnbc.com/2020/01/27/latency-arbitrage-trading-costs-investors-5-billion-a-year-study.html

https://www.forbes.com/forbes/2010/0927/outfront-netscape-jim-barksdale-daniel-spivey-wall-street-speed-war.html

Even if this is true, the solution is to ban high-frequency trading, like they've done in Europe already, not spaff the sky with thousands of pieces of wannabe space junk.

That is the thing, they plan to relay information between satellites within range of each other with auto locking tracking lasers! The tracking systems are fully counter balanced for Newton's 3rd law (every action has an equal and opposite reaction) to avoid unwanted pitch, yaw, and roll. So travelling in a constellation helps reduce the targeting once an initial link lock has been achieved. Messages can be relayed around the globe in space using fricking lasers!

Some of the repurposing of military tech is ingenious, but for a bulletproof long lasting solution it most definably is not the best choice. Ground based fibre has an expected useful lifetime of about 25+ years, these constellations need to be fully replaced every 7 to 10 years. I guess you could look at that as the glass half full, the system will constantly be upgraded.

On the bright side, light from the sun takes about 8 minutes 20 seconds to travel the ~149.4 million km ; ~92 million miles to earth so we will should be able to see a critical CME (CME velocity is usually between 20 km/s to 3000 km/s) before it knocks out all the satellites by worst case maybe 14 hours to best case maybe 2 months warning.

They don't care, it is all about money and publicity now, not what might happen......

Your latency calculation is erong. The satellite path is two sides of a triangle, or worse if any routing is involved, ehereas the fibre link is closer to a direct path. For most geometries, that will wipe out your speed difference. Also, each fibre is a private universe whereas your orbiting wi-fi is a shared medium, so the scale-up is horrible.

SatComs is the best solution only where it is the only solution. There are very few areas where that is the case.

"For most geometries, that will wipe out your speed difference."

Most? Fiber tops out at .7c whereas a satellite link can go almost the full c, at least a 50% speed advantage. The longer the trip, the bigger the latency arbitrage. Peg the length of a fiber optic cable from New York to London at 6,000km. In a good fiber with a refractive index of ~1.467 (meaning light's going right about 200,000km/s), a single trip in the fiber would take...30 milliseconds, not including any stops and conversions along the way (which a sat link would also experience). In 30 milliseconds, light can go half again as far (~9000km) in the same time span in a vacuum.

"Also, each fibre is a private universe whereas your orbiting wi-fi is a shared medium, so the scale-up is horrible."

Just a little bit of information can go a long way. Plus there could be more dedicated channels for those willing to pay up. As for being in its own world, that usually doesn't happen at the trunk level. It'd be wasted bandwidth otherwise.

Ah, but the likes of Musk, Bezos and the Zuck would never do anything like that because they may end up being forced to provide something like the Universal Service in the UK or share things.

It is much easier to it this way and is seen as providing a useful service for billions of people who cannot afford it.

The only beneficiaries of this are the companies putting them up and the data that can be collected. If they decide to be philanthropic and make it free to say Sub-Saharan Africa, then the only reason they will is to monetarise the user's data.

Genuine question....

Would anyone trust Musk, Bezos (or Google) etc. to handle their Internet access?

Even the though of Sky brings me at in a sweat......

"Would anyone trust Musk, Bezos (or Google) etc. to handle their Internet access?"

Which leads to the secondary question, "Would anyone trust their government to handle their internet access?" And the tertiary one, "WHO would you trust to handle your Internet access?"

At some point, you find you're simply Up Crap Creek.

Did anyone else look at that picture and think of the Tholian Web?