Amazon Seeks Permission to Launch 3k Internet Satellites | Hacker News Reader

Amazon Seeks Permission to Launch 3k Internet Satellites

tomshardware.com

172 pointsartsandsci245 comments

At what point should we start worrying about Kessler Syndrome?

> The Kessler syndrome (also called the Kessler effect,[1][2] collisional cascading or ablation cascade), proposed by the NASA scientist Donald J. Kessler in 1978, is a scenario in which the density of objects in low Earth orbit (LEO) is high enough that collisions between objects could cause a cascade in which each collision generates space debris that increases the likelihood of further collisions.[3] One implication is that the distribution of debris in orbit could render space activities and the use of satellites in specific orbital ranges difficult for many generations.[3]

* https://en.wikipedia.org/wiki/Kessler_syndrome

Those dense internet satellite constellations will be in low earth orbit, which have relatively fast decay times. So even in the worst case it would only set us back by a few decades while either cleanup solutions are developed or we wait for them to decay naturally.

You can still launch through the debris field into higher orbits because the collision probability will be quite low at any given moment, it just adds up over time for things that want to stay there.

What about debris from LEO collisions moving into higher orbits?

Gravity is directional.

A single impact can't put something into a higher circular orbit-- it can create an elliptical orbit with a higher apogee (max distance), but the correspondingly nearer perigee (closest approach) means that it experiences more atmospheric drag and falls out of orbit faster.

Granted, I don't know the math and this is intuition from Kerbal Space Program, but I'm not sure that's right. Wouldn't two objects colliding at 0° and 90° orbits result in a higher apoapsis and the same periapsis? Wouldn't it be equivalent to a normal impulse? Some pieces of one of the satellites would have more energy than they did before the impact.

You are correct. The result would be an elliptical orbit, though, not the circular orbit that the post you replied to mentioned.

I don't know if the person edited it after you replied, though.

Kerbal Space Program taught me so much, I now realize. I was able to follow and visualize everything in this thread due to that videogame. Neat.

I don't think the perigee needs to be lower? Consider a collision fragment that breaks off tangentially to the orbital path and moves faster than the speed needed to maintain a circular orbit: its new perigee should be the radius of the circular orbit, but the apogee will be higher.

I thought so too, but apparently this is not universally true. Some of the debris can be put in an orbit with a longer lifetime than the original satellite. I got into a long twitter argument with experts in this field, and I was proven false: some of the debris can have longer orbital lifetime and go to higher orbit.

EDIT: Here: https://twitter.com/SpaceJosh/status/1119721780254371840

Always embarrassing to make a fool of myself on the Internet, but losing an argument IS a very effective way of learning.

You said:

> That seems unlikely, as that requires like 11km/s which is more than the sum of the two velocities[...]

Isn't the first problem here that you're assuming a spherical cow in a vacuum physics problem? Two satellites colliding aren't two indivisible pebbles colliding.

They're going to pulverize on impact, and some of the now-expanding debris cloud might even contain combustibles the satellites were carrying.

The debris also won't only collide once, there'll be a series of rapidly occurring re-collisions. Some of those might impart extra velocity on some of the expanding debris.

Galilean cannon?

Higher orbits require higher velocities, which aren't going to be reached by collision debris.

It’s not likely but definitely not impossible.

https://youtu.be/2UHS883_P60

For the sake of completeness, did a little digging, looks like it did happen in fact - https://celestrak.com/events/collision/

Collisions can absolutely eject debris at a higher velocity than either of the initial impactors.

>only set us back by a few decades

Uhhh, that’s quite some time.

It is interesting to think about. Assume for a minute that Starlink and Amazon's project and a handful of others put tens of thousands of satellites into 500-600 km LEO. Two crash, bits of those crash into others, and runaway Kessler syndrome ensues, resulting in space at those altitudes effectively closed off for 25 years. Assume the ISS at 400 km is no longer safe for human habitation, and it's too risky to try to punch larger payloads through to higher orbits. Human space exploration is effectively dead for 25 years. What happens?

On the scale of human history, that's a moment in time. Today, as we learn about the history of colonial Europe, there's lots of exploration, then the 30 Years War in 1618 slowed that down a lot for a while, then things resumed. In 2500 AD, astronauts will learn of the early history of 1900s first flights, and the 2020-2045 setback, and then 400 years of further development.

But in the short term, that's a full career for everyone who just entered the field. Space companies would publish 100 quarterly earnings statements with zero progress, just waiting for the window to open again. I just don't know that modern society has the ability to think long-term for an event like that.

FYI it appears that for a unpowered satellite at 500-600km the average lifetime is around 5 years.

https://space.stackexchange.com/questions/13851/how-long-can...

I've heard that at least with the SpaceX satellites, they have full plans for decommissioning with these satellites re-entering the atmosphere.

I do agree that this is a real concern though, and we'd hope there's pretty careful consideration, considering the sheer number of satellites being launched. Still, space is big .. really big, even the space just above our planet.

I don't put it past humanity to not fuck up space. After all, look at all the plastics and waste in the oceans. Hopefully all these companies and FCC regulators are thinking far ahead, and require the right designs and safety measures to make sure we don't end up enclosed in a float garbage pile.

Correct on the decommissioning plans. SpaceX lists out in their FCC filings a number of things including how long it will take them to actively de-orbit satellites, how long for them to de-orbit if they lose contact/control, and also how much of the material will survive re-entry.

I have read conflicting things so don't hold me to this, but on article I read said they delayed adding the laser links to the current batch of Starlink satellites because of concerns over how much would survive the de-orbit maneuver.

What I heard they launched them ASAP because otherwise they would lose spectrum license.

This is relevant: https://www.youtube.com/watch?v=yS1ibDImAYU

> I've heard that at least with the SpaceX satellites, they have full plans for decommissioning with these satellites re-entering the atmosphere.

It's not just that they have plans to do so, they're low enough that they'll de-orbit fairly quickly on their own without active control (boost from their ion engines). Something pretty strange would have to happen for them not to eventually burn up.

Isn't the Kessler effect partly about creating further junk from a cascade of collisions? Even if the "junk" were to re-enter the atmosphere eventually, how long would that take? How long would access to space be cut off before the junk were to clear?

I'm not an expert (and expect to be corrected on details here), but I think it depends on the orbit altitude - if it's low enough to suffer significant atmospheric drag, then it'll de-orbit fairly quickly. But if they're at high altitude (geostationary and geosynchronous orbits spring to mind - around 22000-26000 miles), then that's much less likely, although there's a much bigger area to cover at that altitude. But given the damage that even a tiny particle can cause at those velocities, I think it's a real concern with such mass launches.

The smaller any given piece of debris is, the greater it's surface area:mass ratio is, and therefore the faster it deorbits due to drag.

Project Kuiper and Starlink both put the very large, thousands-of-satellites constellations low enough that trash gets cleared quickly enough that even a set of multiple catastrophic collisions is very unlikely to threaten most of the network, and even if they all just blow up, the pieces will fall down shortly and clear the orbit for reuse. For Starlink, drag is actually so high that they are using the solar array as a sail for attitude control.

With right lobbying strategy and targeting lawmakers during campaigns, I have no reason to believe that Amazon cannot influence enough heads in their favor. Are government entities such as FCC are the absolute gatekeepers that we expect them to be? I am highly skeptical. Look how FAA handled Boeing's 737 Max and we ended up with two crashes. A company of Amazon stature and resources can easily influence and overwhelm FCC into giving them green light. For Amazon this is a two fold win; Blue Origin will finally have a purpose, and further domination of Amazon in farthest corners of the world.

I'm more worried about the future sudden need to stop supporting the existing internet once a few companies decide we should just use their new space internet, that seemsnlike the more profitable longterm play.

It’s the same internet, just like when we switched from dialup to broadband.

Not without net neutrality it isn’t.

The bigger issue is all these satellites outnumbering the total number stars visible from Earth. It has astronomers worried.

https://www.nytimes.com/2019/06/01/science/starlink-spacex-a...

Astronomy is fine and all but I think this is more important.

I can't tell if you're being sarcastic or not.

Worldwide satellite internet access provides more value to humanity than identifying stars. Maybe some telescopes should be put on the moon/mars anyway.

I'd tend to disagree with ya on that!

Without astronomy, this whole enterprise couldn't have happened.

All the necessary physics could have been developed just fine even if there had always been a dense fog of debris surrounding the earth past the moon.

The Big Bang vs Cat videos?

One obvious solution is to put them in an orbit low enough that they'll fall back within a few years.

Think about it in terms of cars. Imagine spread out 3k cars around the surface of the earth. Each car will be thousands of Km away from each other. For example New York has 5 million registered cars. And they manage to use only a fraction of available land dedicated to roads. So I don’t think 3k is anything for a orbit around earth, which is also much bigger than the surface of the earth itself

Edit: I am not saying a collision would not cause the effect that OP referred to. I am saying that 3K satellites in Earth's orbit will be very far apart, so the odds of them colliding in the first place is incredibly low.

It’s not 3000 completely self-container items. It’s 3000 satellites with hundreds and hundreds of parts inside, each of which can splinter into thousands of pieces.

I hope you know about how dangerous space debris is for space exploration. Each of those tiny pieces move faster than bullets around the Earth.

Sandra Bullock in Gravity would disagree: https://www.imdb.com/title/tt1454468/

(SPOILER: The debris from one satellite wiped out all the others, including the ISS)

;-)

But satellite collisions can, and indeed already have, happened: https://en.m.wikipedia.org/wiki/2009_satellite_collision

Private corporations owning more and more utility infrastructure really concerns me.

Is there a world where nations band together and launch this as public infrastructure rather than private corporations lobbying a USA government agency for permission?

Get specific about the concern, otherwise it's easy to get afraid about everything.

Govt's hardly own telecom infra, it's all in private hands. And when orgs get too big for their boots, things like this happen- https://en.wikipedia.org/wiki/Breakup_of_the_Bell_System

> Get specific about the concern, otherwise it's easy to get afraid about everything.

There are many reasons to be afraid about a private corporation from a country with little regulation providing a utility to areas with no other choice.

If Amazon was sponsoring this, that's one thing, but them owning the network raises red flags for me surrounding: 1) Privacy 2) The potential for anti competitive practices 3) Quality of service / expected lifetime of service

I simply don't trust Amazon (or any other large private company) to do anything else than look out for their bottom line when push comes to shove.

> Govt's hardly own telecom infra, it's all in private hands

I'm old enough to remember when Australia did: https://en.wikipedia.org/wiki/Telstra#Privatisation

Since Telstra was privatised, any impetus to upgrade Australia's struggling networks has gone out the window and when they do end up upgrading capacity, it is usually as part of a Government project. All in all, tax payers still foot the bill to improve utilities while a private company furthers their monopoly.

Same thing happened in the UK with BT.

Usual privatise the profits, socialise the costs.

This is false. When BT was privatized, the UK government sold the shares and collected the proceeds. That’s not “socializing the costs” any more than in any other divestiture transaction.

Additionally, these companies went from requiring huge operating subsidies to becoming net contributors to the exchequer through paying taxes: https://www.cps.org.uk/research/the-performance-of-privatisa....

> When the Conservatives came to power in 1979, the major nationalised companies were receiving large sums of taxpayers’ money. NERA’s report reveals that in the year to March 1980, the 33 companies it examined were contributing nothing to the exchequer: in fact they absorbed a total of £483 million between them, including £1,199 million in loan finance. British Steel was one of the worst companies requiring £1,020 million in the financial year 1980/81 on a turnover of just under £3 billion (thereby earning itself a place in the Guinness Book of Records).

> This dismal state of affairs has been reversed. In 1987, the 33 companies examined by NERA contributed £8,374 million to the exchequer. Net contributions have continued at a high level in each of the last eight years.

Privacy: NSA already exists

Competitive practices: there are laws for that, and it's going to be competing with Elon's

Quality of service: if it stinks then people won't subscribe to it, and it's going to be competing with Elon's

The traditional justification for the government to own utilities is because they are a natural monopoly (e.g., the sewers), not because they are utilities. We should celebrate when technological advances changing things that previously were natural monopolies into competitive markets, like satellite internet.

At the very least, this has to be be better than the terribly regulated private monopolies we see in the stagnant, captive-audience-holding cable internet industry.

It could be argued satellite internet is also a natural monopoly / duopoly.

How many n-thousand satellite networks can there be?

How many companies / countries can afford to launch n-thousand satellites?

Two this year, of which one is a very young and not particularly huge company. That suggests there's room for a couple more.

There can be a lot. Given the immense volume above Earth, satellite collisions are only a worry because of the speeds that satellites travel and the current difficulty with coordinating them. As the technology for satellite coordination improves (tracking, low-propellent maneuvering, etc), the number of satellites that can fit will keep rising for a long time.

Geostationary orbit is a good example.

There, every satellite is following the exact same orbit, but different positions around it. There are 500+ satellites on it.

Now imagine how many possible orbits there are. If we could locate satellites accurately enough, we could give every 1 meter shell of altitude to another 500 satellites, giving us 2500000000 between earth and geostationary orbit.

The fact that there are only O(10^3) satellites in geostationary orbit is (I believe) mostly a limitation of ground antenna pointing accuracy than collision risk. There are many, many miles between each satellite along the arc.

Sorry, but I don't agree with your premise in this case. How is adding another competitor a bad thing? If Amazon offers Internet via LEO satellites, and SpaceX does the same, and you have your regular telecoms using copper lines, and some wireless carriers offering 4G, and perhaps some community operators offering wireless -- how is all of this competition worse than a utility being provided by the government, which has absolutely no incentive to compete -- and can by law stop new entrants into the market?

It seems like no coincidence that Amazon is also building a ground station network as part of AWS. https://aws.amazon.com/ground-station/

Anyone wanna offer me their hypothesis on what Amazon and other tech companies gain from offering internet to people who probably aren't valuable consumers?

It s a ruse. The US will win the landgrab for controlling the communication networks of up and coming regions like Africa before the Chinese do

Not a bad point at all, especially with Amazon joining at the hip with the U.S. government lately.

Better the US than China, honestly. For all its faults, the US is still a (flawed) democracy with a free press. Its capacity for abuse is at least moderated.

Satellites are outside the regulatory authority of countries they operate over. They can do far more of value than merely provide Internet access to poor people.

Satellites also have to be specialty built for their purpose. You can't just take a small communications satellite, attach a small camera, and have a spy satellite.

Well you "could" but it would be so crappy that it'd be useless.

Also LEO satellites lose their altitude rather quickly, not a great investment for spy satellites.

> Satellites are outside the regulatory authority of countries they operate over.

Except the United States (Assuming the satellite was made/operated by a US entity such as a citizen or business), so they still have PLENTY of regulations to follow.

But they would be good for measuring graphical changes and some weather. Though NOAA has a lot of this already covered and better.

These would make poor imaging satellites you are correct, but what could possibly be more valuable intelligence than a global scale firehose of data. Just the metadata of packet routing across the constellation would have huge value to intel agencies.

To say nothing of the implicit value in denying an adversary access to that information.

There's far more efficient ways of tapping global data than launching satellites.

Just build undersea fibre if surveillance is your goal.

Are you imagining they'll provide uncensored internet in China or something?

I agree it's mechanically possible, but Amazon isn't the company I'd have expected to do it.

It's also dependent on consumers being able to purchase the receiving dish, hide its installation, and the government not to jam the signals. That's a lot of ifs.

They probably are valuable customers. Tech companies have literally no reach where the internet does not touch. If they want to expand to their full (terrifying) power, they would need to give internet access to everyone, and then control it.

Terrifying power? Amazon has succeeded by getting customers what they want, and doing it cheaper, faster and more conveniently than their competition. Of all the tech companies to be cynical about, they most obviously make people's lives better, and expanding their reach (along with the "ancillary benefits" of giving people access to the internet...) is a moral good.

Really, this knee-jerk paranoia is difficult to understand. If Amazon stops serving its customers well it'll go out of business. If it pisses people off it'll go out of business. If it tries to break the law it'll go out of business. It has limited power -- effectively none compared to any kind of government -- and no ambitions, incentives or clear paths to acquire that kind of power.

"If Amazon stops serving its customers well it'll go out of business. If it pisses people off it'll go out of business." This is the central myth of the free market. Part and parcel of a massive internet company is having control over the flow of information, how can people protect their rights if they don't even know they're being violated, or if they're being distracted from protecting themselves by lies and manipulation? That is essentially the stated goal of a company's PR, and the sort of technology that goes into actively manipulative practices that we just accept out of tradition (the entire concept of marketing is a violation of "invisible hand") has now reached the level where people stand no chance, the amount of money and research that goes into making you make decisions that are in a large corporation's interests over your own are something any particular individual could never possibly match.

> This is the central myth of the free market.

Tell Sears and Kmart that it's a myth.

Top US retailers in 1970: Sears, Pennys, Kmart, Woolworth, McCrory's, WT Grant, Genesco, Allied, May, Dayton-Hudson (Target)

Only one of those remains on the list now (in an entirely different form at that).

The Internet has changed nothing fundamental about whether a retailer can be replaced by competition. And no, Amazon's warehouses don't make it impossible to compete with any more than Sears previously having thousands of physical stores supposedly protected them via reach / scale that others couldn't match. For retailers, the Internet is a better catalog. Sears rode the last version to temporary dominance, Amazon rode this version to temporary online dominance.

and more generally than retailers:

http://www.aei.org/publication/fortune-500-firms-1955-v-2017...

The myth here, though, is not that companies are taken out by competitors with better technology or more ruthless practices or better PR, that happens all the time. The myth is that when this happens it is always in the customer’s best interests, or that the customer will even be allowed to become aware of what their own best interests really are. For a company there is no difference between genuinely benefitting their customers and manipulating their customers into believing that they have been benefitted. Making that determination falls on the customer themselves, and in the competition between global corporation and median consumer the consumer is wildly out-matched.

This is literally Big Data in practice. If a superhuman level of intelligence is achieved (as in, comprehension beyond that which any individual human could possibly contain) it will be in the field of controlling groups of humans' behavior.

We could already be there. We won't know, because by definition we won't be able to grasp the scope of it.

My own theory is that this is the strong claim to a-life as practiced by collective organisms. To a cell, a human is a bunch of nearby cells, but to the human, the cell could be a potential fingernail clipping. To collective a-life, humans are no longer the point.

> If Amazon stops serving its customers well it'll go out of business.

You have no evidence for this, and the fact that a court case to hold Amazon liable for bad products was even required provides a counterfactual.

Define "valuable".

If you gather a lot of people with just a bit of money each, they'll collectively have a lot of money.

Telecoms are among the most tightly regulated and jealously held national level industries. If you can put up a global network in a few years that pushes down 4 and 5G speeds, the national telecom companies become obsolete.

There will be some holdouts, but eventually it will become stupid for many places to invest in terrestrial infrastructure just to protect a technologically inferior domestic telecom company.

There will still be bandwidth limitations. Same with 5g whenever that comes. Fibre isn't going anywhere.

Oh, they will be valuable customers: poor people are too poor to afford choice. If Amazon offers "internet access" for x¤ monthly, and "Amazon internet access" for x/3¤, what option will poor people take? After that, all of their online spending will be as profitable as selling groceries in the company town store.

I live in West Virginia and have no household access to broadband internet service. (No cable or DSL available here) The square mile around me has dozens of homes with incomes above the US median. Our area is not especially rugged, but the West Virginia government is both corrupt and incompetent. We are all cheering on Starlink and this new effort from Amazon. The only question is whether 5G will get here first!

Why not use 4G? Starlink's bandwidth (price per gigabyte) is probably not going to be great.

Not valuable? Ships, oil rigs, airplanes, remote vacation destinations (e.g. ski lodges), etc have lots of money and a good concentration of rich people with lots of money.

Yeah and there's already a lot of providers of satellite internet. But, Starlink has not published their pricing, so it's unclear how much of the market will they capture.

Sure, there are competitors. Starlink (or any other proposed LEO constellation) offers a fundamentally better product as a result of low latency (ballpark 640ms vs 20ms). Which they get as a result of being in low earth orbit instead of geosynchronous orbit. Starlinks costs per unit bandwidth are almost certainly far lower, meaning the ability to out compete on price.

I don't see the existing constellations remaining competitive for long.

> Starlinks costs per unit bandwidth are almost certainly far lower

Interesting, how do you know that?

It's not true. LEO systems are far more CAPEX intensive than GEO, and neither Starlink, Oneweb, Telesat, or Kuiper will be cheaper than Viasat-3 and will not be able to compete in terms of cost per Gbps.

Do you have any good sources on that? Not that I doubt you, but would like to read more on the subject.

See the cost of phased array, which Leo needs to work. They're significantly more expensive than standard parabolic dishes, and there's no evidence SpaceX has changed that.

Based on Elon's standard playbook, I'm guessing he's going to try to bring down unit economics with much larger scale in phased array manufacturing than we've seen in the past. If he didn't think he could do this, I don't think they'd be launching Starlink.

That said, I know almost nothing about phased arrays, so I don't know if it's expensive due to some fundamental reason, or if it's partly because it's been a niche/low volume device before now.

Right, Elon needs that to come down to work. The satellite part is very cool and all, but the ground station cost is where the uncertainty is.

Well, look at Oneweb as an example. They have raised 3+ billions, still need more funding and their total system throughout is not that much larger than Viasat-3's,which has a cost of ~1 billion (?). I don't think that Starlink is gonna be significantly (maybe a factor of 0.5x) cheaper in terms of cost per Mbps.

It's an opinion based on:

- Number of satellites launched at once

- Stated bandwidth numbers of satellites

- Lack of concern over intentionally deorbitting satellites for testing and sending experimental prototypes to space with an apparently expected not negligible failure rate.

- Stated cost of the phased array groundstations (in the hundreds)

This is completely incorrect. There is no evidence whatsoever that starlink costs are "almost certainly far lower".

I've been wondering about remote cell towers as well. When I'm out in rural areas I'll regularly get 'LTE', but my downloads will be considerably slower than when I'm in a city (despite a much lower population density). My assumption is that it's very expensive to get highspeed internet out to these locations.

This should enable cell phone companies to place a tower as long as they can run power to it (and solar panels might be an option for really remote towers).

In a lot of rural areas the cell towers are linked together using some sort of wireless tech, usually microwave links or similar line of sight technology.

So in effect all those towers share their Internet bandwidth.

It's not likely this. Microwave transport is fast. The real difference in rural sites is a combination of spectrum (ie. 5mz of 850) vs 20mhz of 1700/2100.

850 has better propagation, hence its usage in rural.

The other factor is antenna patterns and site sectors. Many rural sites have an omni antenna, and with no tilt.

In cities you may see 6+ sectors on a site with 12 degree tilts. So your site has more bandwidth(20+ mhz), and way more capacity.

Long story short.. They give you less of the interwebs in rural because it's cheaper

There's a few, but mostly AWS.

If you're going to put up satellites to do low latency internet, they need to be in LEO. (Speed of light is slow and GEO is far away) This ends up allowing you to provide internet in most locations as well. Kinda a spin-off.

So they want to provide high speed internet to any first world without having to build tons of physical towers (you still need them). I also wouldn't be surprised if Sl Americans who live in sparsely populated areas use Amazon as a primary place to obtain things (I know I did).

So now we can also do things like provide high speed and low latency internet to ships, islands, and people all over the world. Remember that most of Amazon's business is AWS, not Amazon.com. Less than half the world population is on the internet. More users means more servers. Easy to see the the major advantage. Amazon doesn't care if they're using Amazon.com but if they're using AWS. Which there's a good chance they will.

Tldr: AWS is where they currently make their money and where more users will just increase that capital.

One of Starlink's planned markets is global high frequency trading. The latency of radio and laser transmission in atmosphere and space is much lower than undersea fiber optic cables. (Light moves significantly slower through glass than through air.) Even with the longer distance and retransmission they stand a good chance of being lower latency across oceans than undersea cables.

I did the math on this. It probably wouldn't be good for hft. The highly profitable hft is done on the floor of the nyse. There's some from Boston but they typically use directed radio instead of fiber-optic. Satellites will be even slower, but maybe cheaper if you're sending date from California or Australia. But it's still probably cheaper (and more profitable) to put a machine on the floor or even just in NYC.

I'm not sure why hft would be a major pitch. These people fight for floor space at the nyse because that much latency matters.

Think London to NYC, or NYC to Tokyo. Fiber will be slower. https://youtu.be/QEIUdMiColU?t=196

What I'm saying is that it's faster to just buy a server in NYC.

The idea is you profit on acting on cross market trading before anytime else can. It's not about latency to any single market.

Seeing something before someone else can IS latency. That's why servers on the NYSE floor (or ANY SE floor) are really expensive. Because it gives you an advantage. It isn't humans reacting and buying in HFT, it is computers. The humans are constantly updating algorithms, but HFT means it is the computers doing the reactions (based on algos written).

They have servers in both locations. If you see that a stock starts falling in London you have a time window where you can still sell it at a higher price in say New York, until the price drops to the same level thus closing the opportunity for arbitrage.

How quickly can you "see" the price falling in London and get that message to NYC? Whoever "sees" it first wins, right? That's what they're talking about.

I don't think you understand what many HFT firms do. They arbitrage between exchanges. For example the option or futures price of a security in Chicago and price of the underlying in New York. Also arbitraging between equities listed on multiple exchanges.

Co-location isn't a solution to these problems, and low latency lines between exchanges in different parts of the country and world is a huge huge factor to successfully implementing many/most HFT strategies.

GP is obviously not talking about HFT on the NYT, but HF arbitrage between different global exchanges.

s/nyse/arbitrary exchange/g

It seems useful, but tricky to monetize. But if you gave priority to hft traffic then you could charge more as it would still have a small advantage over a standard StarLink plan.

How would that be hard to monetize? You literally just charge people for internet access.

* HFT firms will pay for it if the latency is lower

* Online gamers that play with other people around the globe would also probably be willing to pay to get their ping down.

* Maybe they could have a LEO CDN.

* People like me who travel a lot would be willing to pay for it. It would be especially sweet if I could somehow ditch my phone plan and just have global satellite coverage and use VOIP via Google Voice or something.

Because HFT firms can pay for a standard StarLink and get the same results. There are few HFT customers and you have to offer them something that can't get from your consumer service if you want them to pay commensurate to the value provided.

Your other points have nothing to do with my HFT specific comment.

It's very likely that standard Starlink plans will downlink traffic as close to the customer as possible to reduce lateral utilization, but HFT needs to route traffic on the constellation all the way across the Atlantic.

It will have to be something like this that gives best possible latency for a higher price.

"standard StarLink"?

StarLink doesn't exist yet, and when it does, it doesn't seem unreasonable to think that two satellite constellation providers could serve the HFT market and both be profitable.

They could just charge extra for the “last ms”. So add a ms of latency for the standard plan, and remove it for the HFT plan.

Still totally missing the point. You have to offer something extra to the HFT traders to sell them a $10000/month plan instead of a $49.99 plan. Discriminate on latency somehow for their traffic. Prioritize packets, prefer shortest links over most expedient links, minimize hops. Otherwise if they will just buy your standard plan and you don't monetize them effectively.

Where did you get the idea that satellite constellation internet will cost $10000/m? Or that HFT firms aren't already paying much more than $49/m?

... They're just numbers picked out of the air for one last illustrative attempt to explain why you need something special to entice HFTs to pony up way more than a standard internet plan so you can monetize them effectively.

The point is that HFTs aren't going to be ponying up "way more"...

It's going to be a bit more, for a bit better latency.

Your comments are predicated on the assumption that this internet service is going to be significantly more expensive than existing fiber subscriptions, and therefore needs many benefits to outweigh the increased cost. But I have seen no reason to believe that this will be the case. Every statement I've seen indicates that these services will be quite competitive on price.

No, you've still completely missed the point. I give up.

This is not true as far as I can tell. HFT does not appear to be part of spacex’s strategy at all.

Specially now that the satellites have no crosslinks.

My understanding was that was just a limitation on the first batch, as they're still developing the design, but they're intending to cross link in future. Is that not the case?

Sure, but that's significant feature. Other satellites will be built and launched by the time they get there, and they absolutely need to OSL for capacity.

I originally though this would be a good market, and have voiced that opinion on HN, but afaik SpaceX has never said anything about it.

Someone has since pointed out to me that HFT is currently done by bouncing microwaves off the atmosphere. This, unlike fiber, should be faster than SpaceX's constellation making that business unlikely to work out. (Possibly unless high bandwidth almost as low latency is useful).

> Someone has since pointed out to me that HFT is currently done by bouncing microwaves off the atmosphere.

I would love to read up on this. Is this similar to HARP?

Bouncing waves off the ionosphere has different ranges depending on the time of day. Do you have more info on how that's currently used? I'm curious to learn more.

Not really, sorry

High speed trading. You can send a packet faster from London to nyc via a LEO network than an underwater fiber.

Why are there people investing so much money on these things, filling up the space, just for some stupid trading. Why is humanity doing this ??

$$$

"just for some stupid trading" is one of the underlying forces beneath the success of all modern economies. Without it almost everything you buy would be more expensive.

They referred to HFT only - how does it contribute to "success" of modern economies?

It just sucks out value due to stock market implementation.

That argument can apply to any improvement in efficiency of a process.

Admittedly any contribution of HFT to economic success (other than that of the traders themselves, of course) is probably fairly marginal, but more efficient markets are probably a good thing.

Are you considering “having cheaper stuff” actually progress in terms of humanity ?

Yes, when everything becomes cheaper we can pursue greater technology. Having a 5GHz computer would have been impossible some decades ago; now they can be used for scientific research by anyone who has $500.

Ultimately price is just a measurement of how many resources it takes to make something, with the currency added in as the medium of exchange. A product being cheap means it is now easy for us to make; the previous price of the now-cheap product will be taken up by a better product.

Yes, the march of human progress can be measured in the decrease in prices of all of the things we produce.

Sure, why not? Cheaper stuff, from another angle, means you can work fewer hours to achieve the same lifestyle. And that's pretty good!

But you don't. None of us do. At some point theory has to run into a reality check.

You can if you're a contractor or running your own business.

But it would definitely be nice for businesses in general and employment law to stop being obsessed with 40 hours as a magical number. Working 4 8-hour days, or 5 6-hour days, or 4 6-hour days should be just be an option by default.

(And wages have to stop stagnating relative to prices so that the premise of "cheaper stuff" actually happens in the first place.)

you've grown so accustomed to modern conveniences that you don't realize that you're kept alive by things that not even a king could've afforded 100 years ago.

https://www.washingtonpost.com/news/wonk/wp/2017/06/26/the-a...

Being able to trade milliseconds faster is about as pure zero-sum as you can get. The comparison isn't versus having no stock markets at all.

Dude, we blasted from New York to the San Francisco to enable high frequency trading.

If we had a way to remove all oxygen from earth, require people to wear astronaut suits, in order to reduce 10ms off the speed for HFT, you'd have companies start sucking oxygen out of the atmosphere.

Would you pick up a $20 bill on the ground?

I think we all know the answer to that question, and if you know the answer to that question, you know the answer to your question:

To make fucking money (of course).

Won't this offer internet to everyone on Earth?

For those who can afford to spend hundreds of dollars on a receiving, sure.

(starlink, for example, will require a pizza box sized dish at what they hope will be a price of a few hundred dollars)

It won't be that cheap. I'd bet money on it.

Due to bandwidth limitations, most of these systems won't work well in densely populated areas. The orbits also have an inclination that won't take them over more extreme latitudes

Once they have their own satellite-based cell network, they can take another stab at launching a mobile phone. The Amazon Fire2 phone will be able to use this proprietary network. Amazon could then go for entry into mobile markets by selling phones with data service included free, trying to become a third major mobile platform.

Alternately, Amazon is developing their Alexa-fronted Internet-of-Things (home automation, etc.). This network could serve those devices and make their setup easier. (No need to connect them to your wifi network. Amazon could claim their IoT is inherently safer because they control the entire network that they operate on, so your security camera won't get hacked. (Note that Amazon owns the Ring doorbell products.)

Imagine how many customers will jump at the chance to carry a pizza box-sized antenna for a cellphone!

Honestly, if I could get home wifi speeds in my car, I would.

I hate watching Netflix and having pauses as I drive on the highway.

Real engineering made a video on it: https://www.youtube.com/watch?v=giQ8xEWjnBs

I recommend watching. The channel does really good explanations.

I quite liked that video. It was indeed worth watching.

They're playing the (somewhat) long game. While the portion of the world that doesn't internet access right now is generally poor, adding an internet connection to that populace will drastically speed up their development. This then gives Amazon new customers.

Their options are either wait around for 3rd world governments to build the infrastructure to get them those customers, or do it themselves faster.

Lower latency internet backbone for cross continental links. Latency around the world via satellite is less than via optical fiber, since speed of light in fiber is much less than in a vacuum. For AWS, it's a huge selling point for many customers if they can offer lower global latencies than anyone else! Think of all the algorithmic traders, the canonical example.

I wonder if it would ever make sense to set up some "IXP satellites" so that Amazon and SpaceX's (and whoever else enters this space) various LEO networks could switch in space.

Developing compatible systems would be very expensive. There are very few standards in Space Comms., specially for satellite-to-satellite Comms.

I am looking forward to when Huawei put up their global network... Charge zero for the uplink devices if you buy the first 6 months up front using bitcoin. With free Huaflix and sports channels (unlicenced, use at consumers own risk).

And be perma-banned for accessing harmful content.

No, it's just negative social score, so you'll lose your rented place and can't take public transport anymore.

you are going to need to order straight from china though

Surely every country around the world needs to be involved in approving something like this. Amazon does not get my vote to put anything that could potentially become space junk in the air above South Africa.

Not yet at least. What is the benefit to humanity as a whole?

I am with you here but:

1. It's never about what benefits humanity, it's just about money most of the times

2. Every country and probably several companies of each country will do this as long as we are the divided world we have to live in

3. Just like the internet never became the awesome tool one might see in series like Star Trek (benefit of humanity & stuff like that) something like this will never be that awesome and useful. Until there'd be a market that forces people/companies to act like that. So I think our world is a place where incentives lead us to do the wrong things that actually hinder humanity in it's progress and positive development.

> 2. Every country and probably several companies of each country will do this as long as we are the divided world we have to live in

I really hope that a “world government” never happens, it is too much of a risk.

Consider the US for example: once the land of the free, it has become a land of espionage, lack of personal privacy and persecution. This hasn’t begun with Trump, but it became a lot more evident with him as a president.

What I want to say is that the risk of a global government going rogue is too high.

I hope that we will always have a plurality of nations and ideas, living peacefully with one another.

In my opinion this is the only solution that will lead to success of the human race. For me that is not destroying our planet, evolving in the "right" direction (more education, more science and especially more ethics).

At the moment we kind of burn our resources without need and our nations compete which creates more unnecessary costs and suffering than good outcomes.

We live in a world where you wonder every day why things are solved in that particular stupid way instead of doing things the smartest possible way to really serve humanity as a whole. The fact that it's more important than anything else to "grow" money out of nothing speaks for itself.

What is the situation like, right now? Several countries with atom bombs "go rogue". They do what they want and behave like little children who's toys have been stolen (USA, Russia, Iran, North Korea, Great Britain...).

They screw their own people and take away the wealth and stability they enjoyed for some time now, they increase surveillance and take away the (human) rights the people have fought for many decades and they engage in wars and battles that serve no one on this planet but companies who produce weapons and political agendas of maniacs who shouldn't do this job in the first place.

Of course the danger that this one govt could do something bad is also present but at least we'd have a single point of failure that we could get rid of in case of emergency. Now your only chance is to migrate to some other country in the hope that you'll be given a chance to start a new life there and that no one does something that will force you to leave, again.

Wars, Nationalism, tax escaping, finance, poverty .... nearly everything could be solved in a better way when there'd be no seperate govt's who all believed they were the godsend rulers of earth.

I hope some force some day is just strong enough to force everyone into such a system as I don't think nations will be willing to give up their sovereignty that easily. Let it be some aliens - as long as they are intelligent and act rationally it's far better for us than the current unbearable state we have to endure.

Without that there will never be peace on earth. Plurality and ideas come from people and not from states. You can see that today e.g. in countries who are more open minded and welcome people with different believes.

I imagine these are, like SpaceX's, low orbit satellites which decay and fall to earth over time. There should be no Kessler syndrome risk here.

Amazon ring and echo captures a lot of data, but imagine how much more they get as an isp?

Are there any international agreements covering this stuff?

What if China decide they want to launch another 3000 and Russia want to launch another 3000 and India and Japan and whoever else?

not a problem if they are maintained in the way the SpaceX sats are going to be.

> not a problem if they are maintained in the way the SpaceX sats are going to be.

Aka - Not a problem if they are well thought out, well planned, and maintained.

Are there international agreements that specify this is the case? Just curious, real question.

Yeah ditto. And are there agreements to stop them interfering with each other? I get that space is really really big but surely there needs to be a way to coordinate so much stuff up there.

The ITU is the international body that assists with coordination among nations and their satellite systems.

It's a problem for astronomy for example.

Isn't this a solved problem already by putting cameras in space?

The amazing thing to me is that this article does not include the words “spacex” or “starlink” once in the webpage.

Do you have to pay extra for that sort of press hit?

It's a press release

Iridium has global coverage (you can make phone calls from the South Pole) but isn't LEO so there's a latency issue. Starlink is LEO and will have faster rates & lower latency, but won't have global coverage since the orbits don't go over the poles (sorry, Amundsen-Scott!).

What about the proposed Amazon (Blue Origin?) system?

The FCC requires that Starlink cover Alaska, and it will with a later set of satellites in polar orbit.

Ditto for every constellation approved by the FCC.

Iridium is definitely LEO.

Iridium orbits are ~780km (485 mi). Starlink lower tier orbits are ~340km (210 mi).

So what it's gonna be like - latency and bandwidth?

For latency at LEO, we're looking at below 50ms, going as low as 20ms. For bandwidth, it obviously depends on load and usage. SpaceX's satellites each can handle 1 Tbps, so it comes down to how they distribute and price it. At 100Mbps and 3000 satellites, that's roughly 10 million users. Obviously not all users will be using that much data simultaneously.

You underestimate the hunger. It's a new toy to play with for billions of people. Pretty much 3k satellites at 1TB/s each is equivalent of a bucket of water for a stadium full of people. And they plan to give this access for free. It will overload their network and realize they will need 3 million satellites instead, at least.

SpaceX satellites cannot handle 1 Tbps each, they can handle 23 Gbps each according to their FCC filling.

SpaceX satellites cannot do 1Tbps, and it's irresponsible to keep saying this. See the MIT study. They estimate ~20Tbps of total capacity for 4000 satellites.

Can you attach said MIT study?

http://systemarchitect.mit.edu/docs/delportillo19a.pdf

I am one of the authors.

This was donde in Sept 2018 so its a little bit outdated by now.

Awesome work. Are you going to update it for the 10,000 satellites + Amazon? I'm guessing the answer is no.

Also, are you looking for a job after graduation? If so, please email me :)

The Gilded Age had nothing on this. Two billionaires are battling over ownership of communication in the sky.

Is the winner "Prince of The Air"?

The various national governments of the world better wise up pretty soon to the fact that they are losing all of their leashes on these beasts they've been riding. Sooner than we think pure profit-seeking entities will be on even footing with organizations that are (or at least are supposed to be) deriving their just powers from the consent of the governed. This is a deeply sobering thought.

Who else thinks these systems deliver wireless high speed Internet faster than 5G?

Maybe, but it will probably have lower total througput due to the large size of the cells. One of the reasons why 5G is faster is due to smaller cells.

In any case there's no official data from Starlink, on costs, on speeds, total througput or almost anything (if I'm wrong on this please post links!).

It's not really about speed in my opinion, it's about worldwide availability. 5G requires a node at every few hundred meters, so it will probably never go beyond very dense city centers.

These LEO constellations on the other end can provide connectivity in the most rural and remote places. It would be a game changer for planes and cruises for example. Tesla could include a receiver in every car giving you connection anywhere you drive, etc.

These satellite networks require dishes (with starlink for example they're trying to get down to something pizza box sized). It's a different market than 5G and wireless carrier options. You're not going to have satellite data smartphones with these systems as they're currently planned out. They could, however, be used as backbones for cell stations, but then you're back to square one.

You’ll be able to summon Alexa from anywhere... without even owning an Amazon device! :)

Upon whose rockets?

Referring to Hoi An, we cannot help but mention the shimmering old town nights in the brilliant lantern festival, especially on the night of Trung Thu full moon. This has become a feature that takes place monthly by the city and makes many tourists stay when having the opportunity to travel to Hoi An. https://privatecartransferhoian.com/hoi-an-lantern-festival/

They're just copying SpaceX. That's disappointing.

Just as disappointing as that time they copied K-mart and Wal-Mart.

OneWeb launched before SpaceX, but they’re all copying Teledesic/Iridium.

Should anybody be putting satellites up without some credible plan for decommissioning them from the atmosphere?

If this is anything like SpaceX's constellation, the plan is that the orbits will naturally decay in < 10 years and they will burn up in the atmosphere upon re-entry.

IMHO, it would be nice if the decay orbit for each satellite were tracked accurately by Amazon and published beforehand. If these satellites truly do burn up completely -- great. But if they don't it would be great to hold Amazon accountable for any problems and for waste cleanup.

> accountable for any problems and for waste cleanup

These are companies, not people... $10 buck says if one of these things hits your roof, Amazon will just give you a $100 gift card to Amazon.com

+1 lol. yeah. I guess that's better than nothing.

Decommissioning plans are part and parcel of any satellite launch. These small ones will deorbit pretty rapidly if they are told to do so and a bit slower if it happens uncommanded once they are unable to maintain altitude.

No FCC license unless you do, so yeah - it's accounted for. Unless they want to operate entirely without US downlinks. I do ground stations for a living.

Does that mitigate the chances of debris significantly impacting future launches?

Maybe it's time to talk about tech-free preserves established by international treaties.

E.g.: sections of each continent and ocean can be declared tech free where modern technology is forbidden on earth,sea or land.

This is so we don't lose and forget the nature that forged the present.

My knee jerk reaction to this was negative, but I really do think having some sense of preservation of the ways we lived in the past, for educational reasons if nothing else, is valuable. A living museum if you will.

The first thing I thought of is how microplastics are everywhere now. A preserve like the one I suggested would somewow ideally be protected even from things like that. Future researchers will have a good natural frame of reference to study various unnatural phenomena.

Maybe we should each decide what we want to forget or not forget.

How so? It isn't for individuals but for future generations. Ever heard of UNHCR?

There goes my privacy!

I can't even think about a horrible company like Amazon being an ISP too! Think about all the privacy issues that we are going to have!

But if they do this, this is going to be good for the US broadband customers who don't have internet currently. Not to mention they are stuck with monopoly companies without any competition and options. This is going to wake Comcast, AT&T, Verizon, Sprint and Charter!

But Amazon is going to screw your net neutrality in the most crazy way once the initial period of cheap plans gets over and they have a strong monopoly of user base!

It’s not just the idea that Amazon is going to be an ISP that you should worry about when it comes to their effect on internet privacy. They also recently purchased Eero.

Welcome to the dystopian future where no human anywhere on Earth can look up and see an authentic sky, as their ancestors did.

https://www.nytimes.com/2019/06/01/science/starlink-spacex-a...

And thanks to SpaceX (and now Amazon), astronomers will now have to do their observations from space. If only there was a way to get telescopes up there. Oh, wait... SpaceX just happens to have a way to solve the problem it created!

What's particulary dystopic about it? Preserving nature is valuable because we rely on the ecosystem, you can't really make that argument for the dead emptiness that is space.

Ubiquitous space access even is a common theme in scifi utopias!

I think it's close but missing one really key point. Imagine if this was being done by some non profit international consortium funded by many nations.

then it would go over budget by 100x, and if it ever did finish, users would be subjected to the intersection of all of those countries moralizing content bans?

> non profit international consortium funded by many nations

that sounds horrible

If you've ever looked up at the sky at night in a dark area I don't think you could honestly say it's "dead emptiness". It's fascinating to gaze up at the stars and taking that away would be a tragedy

It is very dead and very empty. Sure, it looks pretty and is useful for astrophysics. But compare that to the utility of global, low-latency internet.

A world where two billionaires can take the stars from humanity. It's like a fairytale, and yet it can happen.

Next, one of them will sell the moon!

When humanity consistently votes for people who do nothing with space unless there is pork involved, humanity kinda deserves it.

Thankfully (/s), due to the light pollution given off by the city I live near, I can't see them anyway.

I don't think 3000 satellites are nearly enough to completely blot out the sky across the entire planet.

That depends on how big they are. :)

You're technically correct, which is of course the best kind of correct.

But assuming they're not the size of small moons, the surface area of any orbital plane is still bigger than that of the entire planet. That's a lot of space to fill with objects that have to fit into a single spacecraft.

3,000 from Amazon, 10,000 already approved from SpaceX, who knows how many will go up from various companies in Europe, China, India, etc...

This is the beginning, not the end.

Yeah... still not enough. Millions probably wouldn't be enough. It's a really big sky.

Sounds like that plastics industry talking about disposal in the 1970's. Now look where we are.

Again... we're talking about an area bigger by orders of magnitude than that of the Earth itself, where plastic is still not covering the entire surface of the earth or the oceans.

This is simple math.

I know people here are emotionally fixated on their dystopian fantasies, particularly where Amazon is concerned, but no one is launching so many satellites that they will blot the sun and stars from the sky. Aside from probably being physically impossible, there's no practical reason for even launching that many satellites.

"Welcome to the dystopian future where no human anywhere on Earth can look up and see an authentic sky, as their ancestors did?" No... that's just silliness.

I don’t see the connection.

Yup. Sattelites are roughly car sized. Think how many cars are on earth and yet cars aren’t covering the surface of the earth.

And earth orbit is an even bigger surface area.

> where no human anywhere on Earth can look up and see an authentic sky, as their ancestors did.

The past ~80 years of light pollution has already prevented this from happening.

The past ~80 years of light pollution has already prevented this from happening.

Only if you never leave a city.

Dark skies are so important and desirable to people that there's an entire tourism industry that revolves around giving people the ability to look at the night sky.

https://www.lonelyplanet.com/travel-tips-and-articles/travel...

https://www.visitarizona.com/dark-skies

https://www.visitcountydurham.org/opportunities/darkskies

https://www.travelandleisure.com/trip-ideas/space-astronomy/...

https://cpdarkskies.org/2018/12/13/destination-dark-skies/

https://thepointsguy.com/news/dark-skies-how-light-pollution...

Clearly there are lots of people who value the sky, even if a few billionaires and people on HN who believe in "tech at all costs" don't.

That is true. People go as far as dial 911 the first time they see the Milky Way.

https://timeline.com/los-angeles-light-pollution-ebd60d5acd4...

The video in the article you linked is not a good example of the long term impacts. The satellites were at a lower orbit than their final orbit and not oriented the same way they will be once in their operational orbit. Once they reach their final orbit, which most of them have now, they will be far more spread out, higher, and all around much harder to identify. That and SpaceX is actively working with astronomy groups to try and lower their impact.

Even if they are easily visible, so what. Should we not build roads because then our children wont be able to see things the same as we saw them? It's progress, and really good progress in this case.

All of that aside, I fail to see how it makes a dystopian future. A quick Google says the definition of dystopian is "relating to or denoting an imagined state or society where there is great suffering or injustice." I fail to see how internet providing satellites are making people suffer or feel injustice. If they work out they will bring internet to a substantial number of people who cannot get reliable and cheap access to resources you and I take for granted every day.

SpaceX's whole goal is lowering the cost of access to space. Again assuming they achieve that goal to the degree they want to, and Starship is sure looking to be a giant step in that direction, then deploying a space based telescope very well might just be the better option. You can basically make it as big as you want and shipping it to space could be cheaper than shipping it to some random place on this planet to be built.

There were plenty of studies done by astronomers that showed the effect of the satellites can be detrimental to astrology. No, we shouldn't just build them; that's a red herring. There are plenty of other competitors that have comparable bandwidth and don't light up the sky. Of course, the musk effect makes people ignore those.

Yep, I wouldn't like my daily horoscope be wrong because of these satellites.

Astronomers. Thanks :)

Reply on news.ycombinator.com