Hacker News new | past | comments | ask | show | jobs | submit login
SpaceX tests black satellite to reduce ‘megaconstellation’ threat to astronomy (nature.com)
244 points by pseudolus on Jan 16, 2020 | hide | past | favorite | 363 comments



The satellites can be seen with the naked eye. There's a visible pass over San Francisco tonight around 6:15 PM: https://james.darpinian.com/satellites/?special=starlink-lat...

The viewing window is actually pretty small. For most of the night the satellites are not visible because they are in Earth's shadow. There is an impact on astronomy but it is being overstated by journalists hungry for yet another "Big Tech bad" story.


No, it’s not. This hurts survey telescopes and it’s an orders of magnitude issue.

A rule of astronomy is that if you can see it with your eyes near a city than it’s really really fucking bright, if you can see it with your eyes in the wilderness after your eyes have adjusted for 10 minutes, than it’s still extremely bright.

Bright, fast-moving things are pretty terrible.

Even if Starlink doesn’t kill astronomy, the next 4 companies with similar deployment will definitely exclude types of sciences and ruin billions of dollars of investments in new observatories.


First, satellites are only visible in the optical when they are illuminated by the Sun, which is only for a fairly short window after sunset and before sunrise. The exact size of the window depends on how high they are, but most of the good "dark" time should be unaffected.

Second, astronomers already use image stacking to reject any number of transient artifacts, like satellites, airplanes, cosmic rays, etc.

Third, while these constellations are going to greatly increase the number of satellites in orbit, it's only by about an order of magnitude. Yes, that's worse, but it's not like it's a problem that hasn't existed before.

The type of observations that are likely to be most affected are surveys that actually search for moving near-Earth objects, especially near the Sun (and radio astronomy which apparently can detect terrestrial emissions scattering off of satellites) but I fail to see how this will mean anything like "the end of astronomy" (and I have a Ph.D. in astronomy.)


Just to add to this, the satellites can still cause problems during the night after sunset, as they can occlude other objects being observed as they pass overhead.


This is extremely unlikely, and if it were to happen it is for an extremely small time. I have never heard of this being a practical concern.


Half a millisecond of occlusion is enough to cause problems?


Depends on the telescope, but yes.


I thought most 'real' telescope exposures were over minutes to hours?


What happens from birds, soot, bugs, airplanes, frisbees, and clouds?


They don't emit giant amounts of EMR.


Also, almost all of those serious observatory telescopes don't have to worry about. Which is what we are talking about.


I'd love to read an article on how Frisbees are the biggest threat to observing the universe :D


Could describe what such a problem would be in layman's terms?


Terrestrial telescopes, which we've invested a bunch of money in terms of observatories in, receive interference from Starlink satellites. Telescopes in space, like the Hubble, don't have the same problem, but they're on the magnitude of 100x more expensive and its difficult to have the same missions.


Noting that the orders of magnitude increase in cost are directly caused by infrequent launches by launch systems with tight mass and volume constraints, and the need to build a space telescope on the ground, have the equipment survive launch, and then survive with no intervention for a decade or two in space.

As new launch systems like Falcon 9, Starship, New Glenn and New Armstrong eliminate these constraints it will eventually be cheaper to build telescopes in space than on the ground.

Other “pusher” motivations include worsening access to real estate, reduced control over light pollution, recognition of indigenous rights, etc.


correct - I didn't think I meant Starlink would kill astronomy (I thought I was addressing some of the hyperbole rather than extending it) - but probably even at the level to be deployed in the near future we would see severely degraded quality of observations for solar system objects. Furthermore - I do think SpaceX is the "good citizen" here.

I don't think the next 5-10 companies launching constellation satellites will be quite as considerate, so one order of magnitude could creep on to two over 15 years.

I think image stacking has been simulated to work somewhere from okay to unusable even for 15s exposures - I think the trails are actually pretty bad due both due to quantity, velocity in the sky, and brighter-fatter.


It's an area that is plainly in the need for global regulation that currently has none.


Earth shine is a thing.


Not during the night. When the satellite is out of direct sunlight, it also does not see any illuminated part of the Earth.


Thanks for showing up to be dismissive of concerns that you understand! Nuance is a thing.


Saying it will "Kill astronomy" is pretty hyperbolic.

It will likely affect some earthbound astronomy significantly, but much (most?) of the most important work in modern astronomy is satellite based.

I'm on the fence on this whole issue. It's not exactly clear what impact it will have on astronomy. Nor what impact it will have on making the internet pricing and availability. Where I used to live, the only options for internet access were expensive and really bad, the positive impact this might have is potentially quite big.

It's hard with a story like this to suss out what the long term effects will be so it's a big grey area.


>It will likely affect some earthbound astronomy significantly, but much (most?) of the most important work in modern astronomy is satellite based.

I'm not an astronomer but I'm not convinced that that's the case. The number of space telescopes pales in comparison to the number of terrestrial observatories. Moreover, advances like adaptive optics have done a lot to close the gap between ground and space capabilities (for optical telescopes). Even once you've invested in expensive tech like adaptove optics, a telescope in space still costs an order of magnitude more than one on the ground (thats being really conservative. JWST has already cost ~10x more than the most expensive ground telescope ever, and the thing hasn't even launched yet)

There were a few space based radio telescopes in the past, but I don't think there are any now. Imagine building something like the Arecibo Observatory[1] or VLA[2] in space. And speaking of the VLA, some of the techniques for getting high quality results (e.g. interferometry without physically conjoining the receivers) are difficult if not impossible to do in space.

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

[2]: 27 of these guys, which are able to move around precisely on rails to generate constructive interference with the different waves they're receiving: https://public.nrao.edu/wp-content/uploads/2016/04/vla_panor...


> I'm not an astronomer but I'm not convinced that that's the case.

I'm not entirely certain what the balance is here. I don't think my post made it clear that I'm not certain how big the effect would be, only that there is a tradeoff here which is hard to quantify.

I do know there is a lot of significant astronomy done by space telescope and that the importance of space astronomy is only growing.

Seems to me the answer is that those who benefit from the new satellite constellations (SpaceX, etc) should finance additional investment in astronomy to mitigate the effects.


> I do know there is a lot of significant astronomy done by space telescope and that the importance of space astronomy is only growing.

Former astronomer here. I would argue that the importance of space astronomy is actually decreasing. Satellite telescopes are most important at wavelengths that the Earth's atmosphere absorbs like IR, UV, and X-ray. At visible wavelengths satellite telescopes have been much less important over the past couple of decades due to the development of adaptive optics.

The main reason to put a visible wavelength telescope in space is that you avoid the smearing out of the source by the atmosphere. Adaptive optics solves this problem by correcting for the distortions induced by the atmosphere and bringing images to resolutions that are comparable to (though still not quite as good as) what you can achieve in space.

Since space telescopes no longer have the advantage of much greater resolution, there's much less of a compelling reason to use them for most research problems. (Some questions, of course, can only be tackled by using the highest resolution available.) On every other dimension space-based telescopes are inferior to ground-based telescopes. They are more expensive by many orders of magnitude, much, much smaller, and they cannot be changed once they're up (at least not easily). By contrast a ground based observatory can get substantially better over the course of a year by installing better detectors.


Adaptive optics have zero issues if you’re looking to make a cool picture, but they do introduce bias which is an issue for scientific instruments. That’s largely offset by more and larger telescopes for the same price, but it is very much a tradeoff based in launch costs which are falling.

IMO, over the next 30 years space based telescopes are going to play an ever larger role.


What about amateur astronomy? Are we now saying that only those with the means to afford space telescopes or expensive software can participate?


Are you an amateur astronomer or just trying to be the devil's advocate?

I'd be curious to know what the impact is on impactful amateur astronomy (as opposed to backyard hobbyists who are just engaged for personal pleasure). I know some amateurs use images created by public telescopes, I'm not sure how much meaningful work is done by amateurs using backyard equipment anymore.

Not being dismissive, genuinely curious.


Variable star observing is actually a place where amateurs really help, since it's about observation time, and not necessarily some hugely powerful telescope. And the more powerful a telescope is, the more people want to use it to point at all sorts of things, which makes it kind of expensive to use those telescopes for variable star observations which take a lot of time.

"Since professional astronomers do not have the time or the resources to monitor every variable star, astronomy is one of the few sciences where amateurs can make genuine contributions to scientific research." [0]

[0] https://en.wikipedia.org/wiki/American_Association_of_Variab...


And the Betelgeuse deathwatch!


In recent news, the only confirmed interstellar comet in our solar system was found by a dude who worked at an observatory (not as an astronomer) and decided to build own telescope for fun in his spare time. It's one of only two known interstellar visitors we've had, and because he found it early enough the pros were able to make very detailed observations.

A decent number of asteroids and comets are still found by backyard astronomers.


If SpaceX works out, we'll probably see a lot more telescopes in space. Wouldn't be that expensive at that point for, say, a school class to launch a telescope cubesat.

I'd trade amateur ground astronomy for that.


I'm not entirely sure how you think looking at an image on a computer screen compensates for the experience of doing observations with your own eyes.

There are plenty of good pictures on wikipedia already, do looking at those compare to going out on a clear night and looking at the stars yourself?


A LEO satellite constellation will absolutely not prevent you from looking into the night sky with a telescope and seeing stars.


"Seeing a star" is a low bar to pass, especially if you get a telescope to help you.


The dead of night will be fine. But during (astronomical) twilight, the time when you are most likely awake and looking up, it will have ugly extra dots.


Also I'm sorry but I really, really don't care about a hobby if it gets in the way of the progress of the entire human race.

Imagine if a painter in Greece in some century BC convinced his fellow Grecians that building a harbor in the cove by their town was a bad idea because it ruined his paintings of the cove. How was he supposed to look at the marine life if you stuck wood and industry in the water???


This. Every change has some negative effect. The question is whether the net effect is positive or negative.


What makes a dot ugly?


Can adaptive optics be applied to filter out objects like Starlink satellites?


Don't think so. Adaptive optics works by changing the physical shape of the mirror. If you measure the atmosphere continuously and bend the mirror the right way, you can remove atmospheric distortion from the image. If anything, AO just gets you sharper images of the starlink satellites (or significantly worse ones if a satellite flies in front of the guidestar that you're using as a reference point to measure atmospheric distortions).


You can filter them out computationally without adaptive optics with enough frames captured.


Modern astronomy works at the edges of whats possible, a single frame might take hours. Needing multiple frames would mean multiple days per observation. It would mean no longer seeing faint and distant objects, limiting us to younger, closer and frankly far more boring objects. It would make observing variability impossible for certain timescales.

You need old distant objects for cosmology, observing the structure if the universe, big bang and stuff. You need variability for finding exoplanets, measuring distance and observing transitions such as supernovae.

Oh, and then a huge part is taking spectra, which means bouncing the light directly off a grating. Filtering transients is hard to impossible there. You need spectra for relative motion, magnetic fields, composition of matter and radiation and of course temperature.

Needing such filters would set astronomy back a few decades


[flagged]


> pay for a resource they’ve been getting for free (unobstructed sky)

Wow.


If someone doesn’t manage or govern it with economic incentives, no one does (see: most of the “public good” rainforest slashed and burned in South America and Asia that we desperately needed managed better, through private ownership and conservation management, to avert climate change).

You’re shocked at my statement, as if StarLink is entitled. Consider the alternative argument: why should astronomy slow global comms progress? If astronomy is more valuable than the ~$30 billion/year global comms market, shouldn’t someone bear that cost?


Because ultimately, arguments relying solely on economics are moot.

Technically, a human life has an economic value. And yes such value is used every day to decide health policies and insurance coverage. Would this mean that SpaceX killing N humans (where N=30B*number of years of operation planned for starlink/value of a human live) is ok? Absolutely not.


That's exactly the calculus a company like Tesla does before taking a mass consumer product like a car to market.


No it doesn't. If you know something will kill people, economic arguments are not considered sufficient. This was famously decided when it was revealed Ford elected not to recall the Pinto on the basis of paying compensation was cheaper then the fleet maintenance they had to do.


Not considered sufficient to avoid penalties in a court of law. But even that example can be reframed as a failure to calculate total risk in the cost / benefit analysis (i.e. they should have factored in the risk of a lawsuit and federal fine).


Hot take: Because Starlink is a shitty answer to ISPs being run by shitty people.


Well then, we should also have Elon pay rent for use of space...


I think it’s hard to argue it belongs to you if you can’t get there. If you can’t defend it or control it, you can’t stake a claim to it.

https://en.wikipedia.org/wiki/Extraterrestrial_real_estate


This is actually one of the more interesting questions in space exploration in general. It has been ever since someone tried charging NASA for parking on the section of the moon they claim ownership of.


Defense would actually be quite easy. Just undesirable. There are working antisat weapons in the major powers' arsenals. However, they would produce nasty debris clouds.

And then there is also the vulnerability of the lauch site and control center...


And it is probably significantly cheaper to take down a sat than putting it up. So it would be a losing battle.


Not in general. A kinetic striker would require as much energy to reach the satellite as the satellite used to get there. Something fancier, like a targeted laser, to my knowledge doesn't exist with anything like the wattage needed to punch through atmo and still have effective kill on anything that far out (assuming you could solve the targeting problem to resolve the beam that tightly in the first place).


The satellite is in orbit. It requires significantly more energy to put something in orbit than merely reachin that altitude. Satellite killers are relatively small rockets that can use a small supersonic fighter jet as first stage, while it requires a much bigger rocket to put a satellite in space (it would depend on the orbit of course).

It is true that spacex can put a significant amount of satllites in orbit witha single launch though, so it probably evens it out.


That makes the relationship sound adversarial, when really there's mutual benefit to SpaceX and the international astronomy community. Astronomy can use cheap lifting capabilities.


And why does SpaceX get the privilege to rule the sky and the radio waves simply because they could stream Seinfeld somewhat cheaply?


It's a complicated question, but there are echoes of it in "Why does the US get the privilege of having their flag on the moon, just because they could get it there and nobody can take it back down?" or "Why do people have any right to the land they occupy, merely because they got there before other people did?"

First-mover advantage is real.


It’s not complicated. The answers to your questions are, “Because they can.” Possession, control, and force carry much more weight than some would like. It just makes those without power, authority, or any other stakeholder equity uncomfortable.

The US is not a super power because it asked politely and a committee granted it permission. Similarly, SpaceX will move forward because anyone with the authority to challenge them allows them to proceed.


What is the SpaceX constellation enabling that others aren't? Hint: theirs is not for consumers.


You're not wrong! The US DoD is already paying StarLink for proof of concept work for low latency, high bandwidth connectivity services.


Hey you're back! I asked you previously "That might work for imaging, but what's your fix for spectroscopy?" here: https://news.ycombinator.com/item?id=21577081

You replied "Optical, RF, or Xray?"

Then I said, "Let's start with optical." but then you disappeared!

So, what's your solution for long-slit optical spectroscopy?


Hey! I'm still waiting to hear back from the astronomer I paid to do the research. I still intend to report back with their DOI from the resulting paper. I recognize when I've exceeded my domain knowledge and need an expert.

To be clear, I think we're well in the realm of academic discussion; no one is stopping any global constellation due to astronomy concerns. As I've said many times in this thread and elsewhere, I think it's a growth opportunity for the astronomy community to move towards space based observation platforms, and with the plummeting cost of lift (ie SpaceX Falcon and Starship) I don't think that's an unreasonable position to hold.


No, they are too bright and will saturate the detector with even very short exposures. Best you can do is just discard the affected area, at worst you have to discard several exposures because the saturation effected the readout electronics.


Arecibo and VLA aren't affected by starlink at all. Those are radio telescopes. They don't even need to wait until nighttime to observe. It's only an optical/ir problem.


Starlink uses radio to communicate with the ground. A common complaint from radio astronomers is new radio satellite operators promising to not interfere with radio astronomy and then making no effort to follow through in that promise.


“It will likely affect some earthbound astronomy significantly, but much (most?) of the most important work in modern astronomy is satellite based.”

Not true at all. Ground based is much cheaper and easier to maintain and reconfigure with new and different equipment. Just look at the cost of the upcoming >30m telescopes vs the James Webb or Hubble.


On the plus side, SpaceX is making launches much bigger and cheaper, which will make space-based telescopes much more affordable. Cheaper launches = more, cheaper satellites.


That's good but I am not sure if the launch cost is where the money goes for telescopes.


You are right, but also wrong. Because the launch cost so much you spend even more move on the space telescope because you want it to work for a very long time to get the best value for your money.

And since you are going to spend so much money to launch it, you also add extra hardware to do other type of observations, so the cost of the telescope balloons upward but you are get more bang for the buck.

If on the other hand you have cheap launches then you can send up the minimum telescope needed since you can always send up a better one or even a very different one if you later discover the need.

Today because of high costs, we need to design the scopes to meet all the possible needs we can think of and afford, result very expensive space scopes.


I guess we would have to look at the actual costs and trade offs. Personally I am still not convinced that launch cost is the biggest problem but I am sure astronomers are crunching budget numbers right now and will adjust plans if it makes sense.


A ground-based telescope will always be way cheaper than one in orbit.

Why make this more expensive than it needs to be? Researchers are already strapped for cash.


> but much (most?) of the most important work in modern astronomy is satellite based.

You're going to have to point out which source you pulled that out of! As a simple example, more than half of the Nobel prizes awarded for astronomy since 2000 are for discoveries made with ground based detectors. And of those awarded to satellite experiments, none was actually competing with a ground based experiment, so satellites mainly bring different, not better, capabilities.


...ground based detectors...

When I encounter this phrase I think of things like the neutrino detector in Antarctica, which of course isn't affected by satellites at all. How much contemporary astronomy relies on visible-light telescopy?


A very large fraction. Optical is still the most important band, probably followed by radio, x-ray, IR and gamma ray in that order. And less then 10% of optical is done using satellites.


It will kill astronomy pretty good.

The key point is usable observation time. Our current handful of satellite telescopes provide 24h of time a day. Each terrestrial telescope provides maybe 8h. However, there are a magnitude more telescopes on Hawaii alone than in space. You would need to get a hundred satellite telescopes to begin to replace earthbased observation time.

And that doesn't even begin to talk about the possible instruments, mirror sizes, astronomical costs of buulding and running satellites, etc.


It really won't.

Satellites are invisible if they're in the Earth's shadow. By the time useful observation can begin, there's an enormous swath of the sky that will have zero visible satellites in it.

Diversify your observation targets. There's more than one interesting thing to science at any given moment. If that means spreading out your observations so you spend two hours a night observing four targets as opposed to four hours a night on two targets, so be it.

At the very worst, this means astronomers will need to do more work during the day shift scheduling and prioritizing. This isn't a new problem. I remember reading a back page article in Astronomer magazine sometime in the late 90s with a page of BASIC code. Someone wanted to optimize a computer controlled telescope to make one observation of 100 or so stars every night. You need to roughly minimize the total Manhatten distance between every two observations, while also eliminating observations below the horizon and penalize observations low in the sky. So the author wrote a program to roughly approximate the traveling salesman problem with the additional constraints. In the 90s. In BASIC. In one page. It might be NP complete to get the perfect solution, but good enough is pretty good.

Add an additional constraints penalising observations where they might be impacted by satellites. Hire an intern working on their degree who's taken an optimization course and have them do it for you.


>And that doesn't even begin to talk about the possible instruments, mirror sizes, astronomical costs of buulding and running satellites, etc.

We're going to have to start thinking bigger if we want to leave this rock. Might as well start now.

The real risk is Elon's companies dissolving and the satellites adding pollution for no real gain. If I could trade observatory time for (working) world-wide internet, I'd take the latter with no hesitation.

As much as I sympathize, most of the cosmos aren't really going anywhere. It'll be there after we build hundreds of telescopes.


Starlink is not for consumers, so you cannot get it anyways. It's for enterprise at this point.


The amount of important work in modern astronomy that is satellite based is limited, as pointed out elsewhere on this thread.


For now. Do the Elon Musk analysis from first principles. The future of astronomy is clearly outside of Earth's gravity well and atmosphere.


Yes, but the current budgets clearly are not.


He's working on that ;)

/s (jeez, your sense of humor folks...)


Well, he is. For the Luddites in these threads, it's enough to cite the JWST as a reason to keep doing things the same old way, over and over, in space and on Earth.

Of course, the reason why people who enjoy lecturing others about why something can't be done differently or shouldn't be done differently seem to gravitate to a site called "Hacker News" is one of those universal mysteries that can't be answered with a computer or a telescope.


How is that clear? What principles did you use?


Because access to space is insanely expensive.

Don't fight the people who are trying to fix this.


Also dangerous and inflexible.


Most astronomy is long exposure which the eventual mega constellation is particularly harmful too because there's no way currently to deal with the light reflecting off them.


You know they don't use film plates for those exposures anymore, right? They stack images digitally. Removing satellite trails is the easiest part of the post processing flow, or it damned well should be.


Digit and film single long exposures are functionally identical and for very dim objects you still need long exposures because the amount of light reaching the sensor is very low. You can't image stack if the object is too dim to appear in the shorter exposure's you're stacking.


[flagged]


Astronomical sensors typically are 12 bit. So anything 4000 times brighter will overflow the sensor and can not be subtracted. And of course you have perfect solutions to Poisson noise, bleeding pixels, and increased dead time caused by shorter read out intervals as well, right? We are done here. But not for the reasons you think. You did not win this discussion.


A simple median filter will take care of it. Have you seen what an unprocessed image from the Hubble Space Telescope looks like? It's riddled with cosmic ray tracks, and they all get filtered out.


There are any number of ways to correct observations for satellite trails, not the least because their appearance and duration are both transient and 100% predictable in any given field of view, and can be gated out of the exposure series. Cosmic rays don't even give us the latter break. (To be fair, the rabbit hole does go deeper than the usual places where a median filter alone will save you, but the general problem is far from intractable.)

But never mind, we've been told that digital imaging is "functionally identical to film," and we're "losing the discussion."


My impression is that a lot of interesting "leads" are generated by ground based telescopes and only then examined in more detailed by space based telescopes. If you kill off the generating mechanism, then effectiveness of space telescopes could also decline a lot.


It's nothing really new and the same dynamics can be observed throughout history. Imagine there was a time one could look over and see the bay without the Golden Gate bridge or high rise buildings. And now, its even an iconic thing!

A generation or two in the future astronomy will probably get redefined quite a bit via remote telescopes and adventure vacations to space and so forth.

Even now there are plenty of enthusiasts who are excited to take pictures of rockets during take off or explore satellites as they pass by. So it actually get more people interested in space and astronomy.

And there will always be people on either side supporting or opposing technological advances - as we have seen thought history. The one thing that is certain in my opinion though is that things will change.


OP is talking about affecting astronomy the science, not astrophotography the hobby. Land-based survey telescopes are incredibly useful cheap tools that do not have alternatives.


I guess the argument becomes this:

Does the proliferation of high speed Internet to every part of the world outweigh the cheapness of current land based astronomy tooling?

If not, maybe spacex can compensate by providing some launch load for space bound telescopes.


We already have cheap internet where people live via fiber optics. Last mile market failures are solveable.


No, we don't. There are tons of places right around me, in central Europe, where people will benefit greatly from Starlink. A lot of rural-based people are excitedly talking about it, even older people who you would not expect to know what it is - and they don't, but they're excited about their access to the Internet.



Ok, then Tesla should really prove they can provide that or find ways to compensate those with land based astronomy tooling. They have done neither and have not communicated in any form that they want to.


tesla?


Sorry, SpaceX, it's easy to get Elon Musk's three card monty of companies confused.


Also, asteroid detection may be hampered - there is still a significant contribution of 'amateur' observations to detecting them.


Staring at a night sky in a designated dark sky area is amazing. Given the pace of SpaceX launches amount of time you can do that without sky crawling with LEO constellations is pretty limited. So days of humans experiencing night sky raw are pretty limited (some would argue it is already impossible). However, I think astronomers will find a way to operate even in a noisy environment of mega-constellations. Given that the objects are bright they can be tracked extremely well. Filtering out them out of the datasets does not seem intractable. Am I missing anything here?

Moreover, I think mega-constellations will actually be a boon for astronomy. Think of it as a platform, while initial iterations will be focused on communication systems, what is stopping them from adding sensor packages looking both inwards and outwards? They already have the bandwidth to downlink all of that. Once on the ground those streams could be combined to produce datasets of unprecedented coverage and fidelity.


I think you're missing OP's point:

Once the sun is set in their orbit, they should be invisible both to the eye and telescopes.


"Invisible" isn't quite correct. They won't be visible as white spots, they'll be black spots blocking out the things behind them. Not a big deal for human eyes. Quite a big deal for astronomers.


This is totally false, the solid angle subtended by the satellites is tiny so the effect of their obscuration is totally negligible.


If it's flying in front of the sun, you are correct. If it's flying in front of an object 100 light years away, you are incorrect.


It'll block a star for ~1/2500th of a second (assuming a three meter blocking radius, and 7500m/s) - I wouldn't be surprised if that went undetected 99% of the time. At most it would read as a minor brightness fluctuation, right?


Not only that, but it first has to actually pass in front of the star. This gets increasingly likely as you look at fainter and fainter stars, but the sky is still mostly empty and is not as common as you might think. (Olber's paradox and all that.)


The satellites are not only small, they are also fast. A fast moving satellite can only occlude a 100 light years away object for a couple of microseconds before it passes. This is absolutely negligible for an astronomical observation lasting seconds to minutes. And even if you launched billions of opaque satellites they couldn't come close to blocking a big enough percentage of the sky for this to matter at all.


There go my hopes for an effective planetary sunshade in LEO...


That should probably be at L1 anyway.

That's where it can shade 24h/day.


It'd be in front of an object 100 light years away for an absurdly tiny amount of time. The percentage of the sky occluded by these satellites is absolutely minuscule. This is not a rational notion.

These are not geosynchronous satellites, but instead are 200-500 miles above the Earth, moving many, many thousands of mph. No, it isn't blocking anything 100 light years away unless your shutter speed is in the single-digit nanoseconds.


The relative (apparent) motion differences and the fact that the locations of the satellites are known should take care of that, no? It’s unlikely that the motions are going to track each other exactly in most cases.

In fact if, as you believe, these satellites would actually occlude things beyond them, they sound very useful for calibration and education.


It's bright because it's in a low orbit and a lot closer to us. Because it's in a low orbit it's in the earth's shadow for most of the night. It's not bright while in shadow.


> Bright, fast-moving things are pretty terrible.

Why is that? Almost everything astronomy studies isn't fast-moving. Shouldn't it be easy to filter out that noise?


Bright = overexposure or unmanageable dynamic range

Fast-moving = more likely for it to traverse your field of view while the shutter is open


This is what I'm wondering. It seems pretty trivial to filter out with some basic image processing.


Everything is "trivial" when it's someone else's problem.


That’s 100% true!


I think there are 2 issues here, and it's making the conversation around it difficult.

First is the impact to "professional astronomers", and from what I've seen this group won't be impacted nearly as much as the other. This group has the ability to use satellite based telescopes, or has the tech already to be able to filter/post-process the images to remove satellites, planes, meteor showers, and any other stuff that might get in the way.

Then you have "amateur astronomers", this groups is likely to be impacted by starlink. This group doesn't have access to the digital filtering stuff that the "big guys" do, from what i've seen, most of the people in this group just use normal long-exposure setups and adding in a processing step would mean a pretty significant change to their process, and probably a lot of additional costs.

Even still, the impact to amateur astronomers seems limited to when the sats are in sunlight, which traditionally isn't a super popular time for stargazing (although I may be wildly wrong on this, as I've read that these sats are removing up to 1/3 of the normal viewing time for some astronomers, so don't take this as gospel), and I still think the impacts will be a lot less doom and gloom than some are saying, but I still hope that SpaceX can work with the astronomy community to see if there are solutions or mitigations that can help everyone out.


We absolutely use short exposures. It's actually more important for us than the pros because the mounts are a lot less stable and polar assignment is far worse. Doing a lot of short exposures is a lot easier than nailing your polar assignment. It also help with bad seeing; you throw away the blurriest quarter of your images and the stack looks a lot better.

Filtering is less advanced, but masking outliers (and their neighbors) and replacing them with an average of the others isn't advanced filtering.

It's annoying the way getting cut off in traffic is annoying, not the way getting T-boned is annoying.


So, you clear your CCD, collect photons for 30 minutes, then some satellite sprays you with a multiple of the number of photons collected thus far, and then ... what?


Instead of exposing for 30 minutes, astronomers take lots of shorter exposures and stack them. Otherwise you'd get tons of noise from CCD hot spots, cosmic rays, aircraft, meteors, satellites, etc.


Well, but then, you also get noise from CCD readout, so it's not like you can just take 18000 100ms exposures and stack them to get the same result as a 30 minute exposure minus the "broken frames". Stacking frames is a workaround for disturbances, but it comes at a cost in the form of readout noise. The shorter you make the exposures, the more disturbances you can filter out without wasting observing time, but the more readout noise you get. So, if you have more disturbances, you either need more telescope time so that you still have enough frames left to see your signal after you have thrown out all the bad frames, or you need to reduce exposure times so you can throw out frames as higher granularity, but then you have to accept more readout noise, which makes it harder to see the signal.


There are already enough disturbances that a 30 minute exposure is impractical. Astrophotographers tend to use exposures around 30 seconds. It's true that more satellites mean that astronomers will have to tweak their observation methods and/or reduce their observation windows, but it's not a showstopper. At worst it's an inconvenience to them.

If we were to employ the reversal test[1], the concern about astronomy would be a non-problem. Imagine if the entire planet was covered by the equivalent of 4G cell networks. And imagine if some astronomers asked us to destroy those networks so that some of their work could be made more convenient. Imagine all of the people affected by this network. All of the lives saved by emergency calls, all of the remote locations made digitally accessible to the rest of humanity, all of the scientific experiments in jungles, deserts, tundra… cameras and microphones and sensors reporting data through this global satellite network… imagine all of that destroyed so that some astronomers could be relieved of an inconvenience. That is absurd. Yet that is the world that some people want to live in.

It's so clear to anyone whose view isn't so parochial, so local in time and space, so blinkered by where and when they were born. Does anyone think that astronomers will still be preventing the launch of satellite constellations in the year 2100? In 2200? In 2500? Clearly not.

I'd rather the improvement happen in my lifetime than after. Launch away.

1. https://en.wikipedia.org/wiki/Reversal_test


Is there any way, perhaps with software, that we could remove a moving object, from an otherwise mostly static set of images?


Absolutely, in fact it's pretty easy and is routinely done by amateur and professional astronomers alike. We've had satellites, airplanes, and meteors obstructing astronomical images ever since we began taking them.

A very simple and common way is to stack all of the static images aligned on top of each other such that you have a set of values for each pixel corresponding to the same region of sky. Then sort the values of each pixel by brightness and keep the median value for that pixel. Or throw out that top and bottom 10% and average the rest, or throw out the top and bottom N, etc.

This is a standard feature of essentially all astronomical image processing software and has been for a long time.


Sure. But increasing the number of frames with tracks increased the number of datapoints you need to throw away, reducing the effective time on target and therefore sensitivity of your observations. Astronomers are not idiots. They know how to deal with obstructions. But that is not free of cost. Neither in observational limits nor computer time nor required brain power.


Sure it's not without cost, but in most cases the cost is not very high.

Remember you don't need to throw away the whole frame, only the pixels that were obstructed. One track across a frame will obstruct less than 1% of the data in that frame.


One of the key assumptions that's being made, throughout this thread, is that astronomy is mostly imaging. Taking or stacking images. A ton of astronomy is spectroscopy, which is nothing like a stack of static images with one that has a streak in it.


Could someone explain what the actual tangible negative implications are besides astronomers needing to adjust?


Personally, I don't want to be limited to observing the stars from planet Earth for too much longer. The development of commercially viable space technology will fund and fuel investment in interplanetary expeditions and help us move to the next stage of human progress.


Could you explain how something that orbits 500km up could be seen at night?


Like most other satellites, it can be seen when it reflects the Sun. Satellites tend to be shiny. For an extreme case, check out "Iridium flares" - a brief but bright flares on the sky caused by old Iridium satellites, whose antennas were essentially large, rectangular mirrors.


Depending on the type of measureme t taken, a dark transient is also harmful. However, by far not as bad as a bright transient.


To reflect sunlight you need the sun. If a satellite is 500km up (and 300km for later revisions), there is no way for sunlight to reach it.

There is no way to see Starlink at night. You can potentially see them for a short period around twilight.


Here is the actual geometry at the equator, ignoring axial tilt (the blue/black circle is Earth, the red/black circle is a satellite at 500km, and the radial lines are hours): https://www.desmos.com/calculator/3br1snafmj

Starlink satellites directly overhead are in sunlight for about 1.5 hours after sunset, and the same before sunrise. Including observing on an angle, you probably lose 4 hours per night out of 12. With 300km satellites, you probably lose 3 hours. I think the situation will be worse away from the equator in summer.


You've really never looked into a night sky and seen a satellite? Unless you live underground or never leave a big city, it's quite easy and there are dozens of apps that will help you understand what you're looking at.


I guess the point they are trying to make is that 500km (or 300km) is essentially hugging the planet. Which means that, if it is night, they will be in the shadow for the majority of the time.

The satellites you can easily see at any time are much higher up.


> There is no way to see Starlink at night. You can potentially see them for a short period around twilight.

The time window is limited, yes. So it is not the end of the world. It is still a problem.


A shortsighted attitude, IMO. Terrestrial observatories are in a good position to clean up satellite artifacts digitally. They already have to do more image processing than the staff at Playboy. Meanwhile, cheaper access to orbit can only be a good thing for astronomy, given that we've picked most of the low-hanging fruit in terrestrial-bound optical observations.


Depending on source class and telescope location it ruins between 1/3 and 1/2 of the available observing time. Are you willing to pay $1 more in taxes to to fund the extra telescopes we need to make up the short fall? Please write to your representatives and senators if you do.


Yes, I am for building more telescopes, and especially space based ones which should become much cheaper to launch when SpaceX's Starship is ready.


Satellites and space-based telescopes are still monumentally expensive even if you ignore launch cost. The James Web Space Telescope has cost almost 10 billion dollars in development and won't even launch until (unless the schedule slips again) next year.


One of the reasons JWST is expensive is its complex folding mechanism. Starship could launch a mirror bigger than JWST's in one piece.

Another reason is that they're only launching one and it has to work perfectly the first and only time it's launched. That level of reliability in a one-off product is incredibly expensive to achieve. With dramatically cheaper launches it would make sense to launch a much larger number of less reliable but much less expensive telescopes.

Another reason is politics, but SpaceX can't solve that one.


> it would make sense to launch a much larger number of less reliable but much less expensive telescopes

What if... SpaceX made up for the pollution they introduce by making the Starlink satellites look up the other way and push the captured data back to earth? The lens and CCD would be small, but with the massive volume (and clear sky) it could add up. Like the amateur set ups using an array of consumer grade cameras.


Space-based telescopes are expensive in part because launches are expensive, making it more effective to spend a lot of money on single telescopes.


Space based works for optical, but not all other wave length. For example it will also reduce the sensitivity of air cherenkov telescopes that detect gamma rays and that can not be moved to space.


Would air cherenkov telescopes be significantly affected by additional satellites in orbit? It sounds like they're designed to detect high energy gamma rays striking the atmosphere, so probably not?


In practice they are looking for very weak flashes of light, adding extra background light obviously reduced sensitivity.


is cherenkov radiation even observable during dawn/dusk +/- 2 hours? I imagine that the atmospheric refraction of the sun would seriously hamper those observations during that time period.


On traditional telescope using photomultipliers such as Magic, Hess or Veritas: no. But systems with solid state detectors such as FACT are moving in that direction.


Can you explain that for me? I'm not sure how better detectors let you delete the noise from the sun.


I'm struggling to see how starlink will affect those telescopes. Please explain?


Why can't those be moved to space?


Because cherenkov radiation is only observable in a transparent medium like air or water. Space doesn't work because the speed of light in space equals the speed of light in vacuum, therefore no cherenkov radiation.


Couldn't they look down?


Cherenkov telescopes for neutrinos do look down for reasons of schielding. Neutrinos can pass the earth, other particles not so much.

But those are a special case, air cherenkov telescopes are looking for "less weird" particles like photons or protons. Those can only be seen looking up, since the primary particles moving down focuses the cherenkov light down in a narrow cone.


The Cherenkov emission is boosted in the direction of travel, i.e. downward. And there are too many lights on the surface of the Earth.


While this certainly requires non trivial large scale space engineering, you should be able to build in in principle.

Basically a big bubble filled with the most useful gas for this + bunch of photodetectors inside at the appropriate places. You could also make the whole detection chamber much larger, than the ~30 km (?) of reasonably thick atmosphere you get on Earth.


They are imaging the flashes as the relativistic particles slam into our atmosphere. No atmosphere, no flash.

https://en.wikipedia.org/wiki/IACT


The viewing window is actually pretty small.

Tonight. And what about when SpaceX gets all 11,000 satellites up there? And then Amazon's constellation. and then all the other American companies planning to do the same thing. And the European companies. And the Chinese companies and the Indian companies, and on and on and on.

Dropping a piece of plastic in a lake isn't a big deal. Until it's 11,000 pieces of plastic. And then hundreds of other people do it, too.


The viewing window is the same length no matter how many satellites there are at the same altitude, because Earth's shadow is the same for all of them.


One step closer to a sky that looks like a television tuned to a dead channel.


That's an amazing site, thanks!


hopefully the clouds clear up before then!


Which telescopes are observing at 6:15 pm local time?

The satellites are not in sunlight unless the sun is not that far below the horizon.


> The impact on astronomy is being overstated by journalists hungry for yet another "big tech bad" story.

Indeed, and let's not forget that Elon has upset a lot of applecarts with SpaceX and Tesla.


Astronomers are already criticizing such ideas, and have already created an international appeal by professional astronomers open for subscription to ask for an intervention from institutions and governments. See: https://astronomersappeal.wordpress.com/


Astronomers will have to get used to this. If we ban these for such nonsense reasons, it will be Chinese satellites blocking their view.


Isn't this exactly the same view as "if we regulate the use of fossil fuels, China will use way more than us and outcompete us economically" and "if we limit the number of nuclear weapons we create, there's going to be a gap between our nuclear firepower and Russia's"?

Creating a livable planet is not easy, but nationalistic thinking makes it a hundred times harder.


Which do you think has more capability to improve and save lives? 1. Cheaper internet available everywhere for anybody willing to pay 2. Better data on the cosmos

I'd go with #1.


1. The response isn't really relevant, because I was pointing out an argumentative mistake in my parent comment. They said, "if we don't do X the Chinese will"; I was not making a point about the relative positives / negatives of doing X.

2. I'd go with #2.

3. Even if #1 is right, jumping to the conclusion that Musk's project is a good idea is a complete non-sequitur. As far as I know, no one has argued that putting 10000 satellites into space is the only way to provide widely available cheaper internet. As you can see from discussions elsewhere in the thread, it's not likely to even do that. Will SpaceX provide service to China?


Then we get blasted by a gamma ray burst from Betelgeuse and we're all dead. Studying the cosmos is extremely important, and studying threats (asteroids, supernovae, etc.) are just one part of that.


What is the point of studying the cosmos if we never leave our planet? Space is coming. Admittedly much slower than we expected. However SpaceX's satellites are just the beginning. Astronomy will continue, just much more so. Imagine a telescope array the size of the earth scanning the stars from orbit. Why look at an asteroid when you can GO there and do something with it?


How should we prevent this gamma ray blast? Or will astronomy just tell us its coming?


How do satellites make the planet unlivable?


You believe there's no use in developing global cooperation to regulate issues like this? Because if so, we have far more to worry about than satellites.


"nonsense reasons"

You mean..."Science"?


Any "Science" being done can correctly account for it.


What useful science has been achieved in the last 100 years thanks to astronomy?

Meanwhile, what kind of useful science do you think can be facilitated by globally accessible high-speed internet? The value generated by such a network is clearly orders of magnitude more useful than observing the cosmos, at this juncture of human endeavor. Not too mention that Starlink will allow SpaceX to re-invest more and more money into space launches / space travel. I'd much prefer humans actually visit other celestial bodies rather than just staring at them.


The internet was created in a particle-physics laboratory: https://en.wikipedia.org/wiki/World_Wide_Web

In the last 100 years, advances in particle physics have been aided by our study of high-energy mechanisms in the universe (nuclear fusion in star systems, supernovae, acceleration of the expansion of the universe, etc.)..

so you have internet thanks, in part, to astronomy..

what a silly comment..


We did not discover nuclear fusion from the stars, we realized the possibility of nuclear fusion and subsequently realized that this is what must power the sun (and every other star).

Ditto for everything else on that list. We made discoveries by looking at things and performing experiments here on earth and then realized that these things must be what is causing X and Y out in the cosmos.


Again you make wild false claims..

The first major breakthroughs for understanding nuclear fusion came because we wanted to understand what powered the Sun and Stars: https://en.wikipedia.org/wiki/Timeline_of_nuclear_fusion

.. and these events began 100 years ago.

Part of what you say is true, we make discoveries by looking at things.. Astronomy gives us more things to look at, and in energy regimes that cost many $$ to replicate in experiments on Earth..


Your claim is not substantiated by the source, while mine actually is.

Without Einstein's E = mc^2, Arthur Eddington would not have realized that the fusing of nuclei could power the sun. In other words, he realized that fusing two atoms would release a lot of energy thanks to terrestrial research, and then hypothesized that this is what must power the sun. He did not come to this theory from staring at the sun through a telescope. Astronomy clearly did not lead to fusion - atomic theory did, which did not come from astronomy.

And neither did Einstein formulate E = mc^2 because he was an avid astronomer who wanted to understand the sun.

I am well aware this happened more than 100 years ago, not sure why you would be under the impression that I thought otherwise or needed to hear that for some reason.


The discovery and confirmation of relativistic physics has resulted in more efficient materials research and led to space technology (communications) and microprocessor improvements here on earth.


The World Wide Web is not the internet, but your point is taken.


We pretty much have globally accessible high speed internet though, and poverty is the main factor blocking significant further access, not mountains.


We absolutely do not have globally accessible high speed internet.

Starlink means you can be doing research in the middle of the Amazon and have good internet. Antarctica. Middle of the ocean. Wherever. Sure, population centres are all connected to the internet by now, but that's really not what I'm talking about.

Also, a good way to make internet cheaper (so that those in poverty have access) is to provide more alternatives to access / competition.


Third time I'm saying this, but starlink is not for consumers. It's not a "cheaper" alternative to your current ISP. None of your comment is based on any of the information put out by SpaceX.


I did not say that Starlink is cheaper. I said a good way to make something cheaper is to increase competition / competing ways to deliver something.


None of those noble cases are actually what it's being built for though, the 100 scientists in the Amazonas aren't the reason you put up 1500 satellites in one year.

Likewise, those in poverty living far from population centres (by the way, those are the places with most of the people) are better served by being gifted 10 bricks so they can build an actual stove.


What is it being built for, if not globally available (as in, actually anywhere in the globe), high-speed, low latency, low cost internet?

It's built for whoever gets utility out of it and will pay for it, those "noble cases" included. It's not for them, but it enables them.

The internet can tell you how to build a better stove, farm better, raise animals better, discourage you from barbaric rituals that don't actually have the effect you believe them to have, etc. Arguably the poorest, most ignorant of the world are the same people that need the knowledge the internet can provide the most.


I agree with all but the last sentence. The Internet is not always correlated with a happier life and the Internet isn’t sure to educate the ignorant.


I did not claim either of those things.

Internet access gives you access to actual knowledge. Yes, it also gives access to fake news and farmville, but I think those latter concerns are less important when you are having sex with babies to make your AIDS go away.



What is knowing about the CMBR useful for?


Umm? Knowing about the Big Bang?


The discovery of the CMBR did not lead to the theory of the Big Bang. But astronomy in general certainly did, for sure.

That doesn't really answer my question though, but that's probably my fault for using a word as subjective as "useful", so let me rephrase:

How has astronomical knowledge of the CMBR/Big Bang/Black Holes/Pulsars/Galaxies/etc. fundamentally improved the human condition? What specifically have we been able to do (not "know") that we would not have been able to do without terrestrial astronomy?


What useful thing has been produced by the internet? I'd much rather have humans enjoy nature than writing about it on the internet.


Surely you can't be serious.


The internet produced thousands of wonderful things. Your argument is 100% baseless.

Astronomy is nice too, but I will choose the internet every time, if forced to choose. We have wonderful images of the whole sky in all kinds of spectrum in incredible resolution.


Minority individuals who would normally feel outcast from society have access to a culture of their peers.


> What useful thing has been produced by the internet?

Is this Hacker News? This is the type of nonsense I expect on the NYTimes comment section.

> I'd much rather have humans enjoy nature than writing about it on the internet.

Why are you here instead of enjoying nature?


the Space Force will take care of that


Well put.


SpaceX could provide a handful of at-cost launches for astronomy projects per year. It wouldn't fix anything for the ground telescopes but could be a bit of an olive branch for the community.


Even if they were free launches that wouldn't rank above a taunt: most of the satellite cost is in the development and construction anyway.


Starlink satellites cost < 500k each, compared to hundreds of millions for traditional sats.

If you mass produced space telescopes they wouldn't cost the ridiculous sums spent on JWST, which is a terrible example of the wasteful cost plus contracts of nasa. They've been planning it since 1996, and costs have risen from 1 billion to 10 billion.


I wonder how much it would cost to get a decent camera on one of those Starlink launch missions just so it would beam down some distortion free, high-resolution pictures to the masses.

Although I doubt that would provide meaningful data to professional astronomer.


Wonder if they're using VantaBlack, Black 3.0, or some other variant of "super black" for this experiment?

https://en.wikipedia.org/wiki/Vantablack

https://culturehustle.com/collections/black/products/black-3...


It depends - coating on a low orbiting satellite needs to conform to various requirements:

- needs to survive vacuum

- needs to survive atomic oxygen, that does show up at low Earth orbital altitudes

- needs to handle the thermal cycling as the satellite goes in and out of Earths shadow

- needs to survive unfiltered sunlight without any atmospheric convection to normalize temperature

- needs to avoid overheating the part of the satellite it is covering

- should not emit particles, that could collide with other satellites

- needs to keeps doing this for about 5 years (design lifetime of individual Starlink satellites)

- should harmlessly burn up on satellite reantry

If the given material can do all the above, while still keeping it's desirable properties, then it can be used on a Starlink satellite. :)


One of the targeted markets of VantaBlack was the aerospace industry to paint the inside of baffles on telescopes and optical sensors. It is currently being used in some star trackers on satellites [1].

Overheating is definitely a big concern for painting your satellite back. There is a lot of work that goes in to thermal design of satellites and that surfaces have the proper optical properties for absorption, reflection and emission.

[1] http://www.eurekamagazine.co.uk/design-engineering-news/worl...


It's possible, although it's kind of tricky to use in practice.

From the Vantablack wikipedia article:

When light strikes Vantablack, instead of bouncing off, it becomes trapped and is continually deflected amongst the tubes, eventually becoming absorbed and dissipating into heat.[7]

Dealing with heat in space isn't easy, since you have no air to dissipate heat into.


Black satellites are nothing new. Just ask any Surveillance Agency how they handle the heating problem due to black paint.

Oops that's classified. But probably such satellites need a dedicated cooling system.

It's never as easy as, just paint it black and hope for the best.


Right, but that means that stealth costs mass (weight) and means less satellites per launch. So that means making them less visible costs more money.

For non spy satellites where stealth isn't one of the top goals, that might mean people don't do it.

Also, typically one does not simply ask a surveillance agency how they do things. They would probably respond with something like, "NO SUCH PERSON AT THIS ADDRESS, RETURN TO SENDER."


From what i read they are going with "partially painted black" Makes sense to only have the underside facing earth black. Also the satellites underside is only exposed to the sun with an steep angle because most of the time when it would be exposed to sunlight it's going to be in earths shadow. Reducing the extra heating even more..

Could work without extra cooling...


Black radiates heat faster than white as well as absorbin it faster, so ostensibly the side in shadow should dump heat quickly if you have something like heatpipes to move it there.


Is this voted down because it is wrong? Educate me.


They are probably using something like Aeroglaze Z307, it's a black thermally and electrically conductive paint used on many aerospace applications.


That's a juicy detail, you'd have heard about that if they were.

But odds are they're only going as far as is required, which would be "no brighter than typical satellites" - right?


There will be many more than typical satellites, so it is only fair to require "much dimmer than usual".


Question: If SpaceX is successful in developing the Starship, couldn't they launch huge space telescopes for a very low cost? I'd imagine that SpaceXs efforts will be a net positive for astronomers in the end. If the satellite problems becomes too big, maybe they should offer discounts for launching space telescopes.

Another question.. if you are building a radio telescope in space, could you just use a thin foil that folds out like origami for the reflector?


> If SpaceX is successful in developing the Starship, couldn't they launch huge space telescopes for a very low cost?

If the BFR (the rocket behind the starship) is successful then yes it could mean the ability to launch very large telescopes in to space. The scientific community would be very exited about this possibility. However, this doesn't necessarily make it very low cost. One launch of the BFR would still likely be much more than an a Falcon Heavy launch.

> if you are building a radio telescope in space, could you just use a thin foil that folds out like origami for the reflector?

Yes! This technology already exists and it is really pretty amazing to see in action. Right now most of them are used on communications satellites or for synthetic aperture radar satellites. See the videos below:

Animation of the radar antenna on SMAP:

https://smap.jpl.nasa.gov/resources/83/smap-antenna-deployme...

Actual video of a large communcations antenna (12m diameter) being deployed. Skip ahead to ~2:15 for the actual unfurling.

https://www.youtube.com/watch?v=_mFnNDzxKFk&feature=emb_titl...


Couple quick clarifications:

BFR isn't a name that's still in use. Poster you're responding to was correct in calling it Starship: "SpaceX's Starship spacecraft and Super Heavy rocket (collectively referred to as Starship)" (from https://www.spacex.com/starship).

Starship projects to be significantly less expensive than Falcon Heavy _or_ Falcon 9. With total reusability of both stages and a construction built toward little to no refurbish or rehab, the cost per launch is nearly completely dictated (order of magnitude) by fuel costs, and project to be ~$2 million. This is an order of magnitude reduction in $/kg over the Falcon 9.

https://www.space.com/spacex-starship-flight-passenger-cost-...

https://www.thespacereview.com/article/3740/1


The article you pointed to said that is would be $2 million that SpaceX would have to spend on each launch. That would not be the amount for someone to purchase a launch with that rocket. Considering Elon estimated that development would cost $5 billion to $10 billion [1], the cost of launch would likely be much higher based on recouping the intial development and manufacturing costs.

As a side note, I don't really believe the $2 million price tag either based on my own experiences. Mission specific planning/services/verification tend to push prices of launches 10s of millions of dollars above the "sticker prices" that SpaceX puts on their website.

Nothing against SpaceX, I am a fan of everything they have done to decrease launch costs. They have significantly changed the game in terms of lowering launch costs. But it is really hard to take Elon's wild numbers that he gives the press at face value.

[1] https://www.theverge.com/2018/9/18/17873332/spacex-elon-musk...


I know it’s extremely far fetched, but part of me hopes that if Starship is successful and they have they room, they could capture the Hubble and bring it back down to put in a museum.


Launch costs are a relatively small part of the cost of space telescopes. Pretty much always less than 10%, sometimes as low as 2% of the program's cost. The expensive part is building something that can run for years at a time with no maintenance. Things that are relatively easy on the ground (keeping some parts at cryogenic temperatures, having enough electricity) get significantly more complicated (and expensive) in space.


But if launch costs go from $400M to $4M maybe it's ok to put stuff up with dramatically shortened mission expectations.

What really needs to happen is the ability to build mirrors in LEO so they don't have to be built to survive the launch.


Saving 8% of the total budget will not make it easier to make reliable cooling systems, nor easer to get the remaining 80% of needed funds.



Is there a specific reason we're still doing ground based astronomy? With satellites becoming ever cheaper, sure we at some point should be able to get a significant telescope up there right? Are we waiting for the bigger rockets to accomplish that?


Yeah. Depending on the spectrum, the telescope might need to be really big. Not to mention at the moment a telescope in orbit would hard, if not impossible, to service.

Perhaps a review of telescope design is in order: https://en.wikipedia.org/wiki/Telescope https://en.wikipedia.org/wiki/Optical_telescope

Building large telescopes is hard enough. Putting them in orbit just adds to all the costs. Look at the James Webb Telescope (which still hasn't been launched). https://en.wikipedia.org/wiki/James_Webb_Space_Telescope

It seems possible to launch multiple small telescopes and operate them as one large scope using aperture synthesis. I don't know if there are any existing designs or plans for this.

Also: somewhat ninja'd, see other replies as well.


> It seems possible to launch multiple small telescopes and operate them as one large scope using aperture synthesis. I don't know if there are any existing designs or plans for this.

We know how to do that in radio (VLBI), have some experience in IR (ALMA), are doing research on how to do that in optical. But in practice that is much harder than you think. The relative distances of the telescopes have to be known and constant to within a few fractions of the wavelength you are using. Hard when you are using centimeter radiowaves, insanely hard with optical light that has 600 nanometers wavelength.


We've been doing optical interferometry for quite some time. Homodyne techniques (see facilities, CHARA, VLTI, COAST, NPOI, SUSI) in which the light is interfered with itself are quite common. Heterodyne methods (one facility, ISI) in which the incoming light is mixed with a stable laser and downconverted to longer wavelengths are uncommon though.


Launch a bunch of small telescopes, connect them with a rigid structure. Keep it shaded. Shouldn't be impossible. And with no gravity and no atmosphere, even a light structure will keep it all positioned very precisely.


Innumberable reasons. Data transfer, stability, power, heat dissipation, ability to use large area parts, accessibility for fixing and upgrades, and on and on.


It's very expensive to build (especially large) space telescopes (JWST is already costing more than 8 billion at this point), and astronomy is not very well funded.


It's fascinating to me that the telescope costs so much more than the launch.

A high-quality 24" or 1-meter university-grade observatory telescope can be had for well under $1 million. If you multiply that by a factor of 100 to mount it on a satellite, you're still at 'just' $0.1 billion and can buy a whole Ariane-5 launch just like the JWST to put it at your desired orbit for $0.15B, for a total of $0.25B (a Falcon Heavy runs about half the cost for a launch). You could launch 30 of those (hopefully improving your factor-of-100 cost increase to something more manageable) for less than what the JWST will cost.

I get that JWST is a 6.5 meter telescope, not a piddly backyard 24" device, but why do we have to launch the best single scope possible?

There are only 7 visible-light space telescopes listed at https://en.wikipedia.org/wiki/List_of_space_telescopes#Visib.... I wish there were 70 or 700, with live Internet feeds.


JWST is being built to do things that are truly impossible on Earth. It is an infrared telescope that will image things that cannot be seen through Earth's atmosphere.

JWST is not the first IR telescope (Spitzer Space telescope, retiring this month comes to mind, 0.85 m diameter primary), but its size will allow both improved resolution (diffraction limit falls like 1/diameter) and improved collection efficiency (grows like diameter^2). Without constellation-flying and interferometric telescopes (see Keck Observatory), one cannot get either one from an array of small telescopes.

There are a lot of scientists grumpy about JWST because of its huge budget, but as long as JWST works, the view it gives of our universe will be spectacular. At this point, I think everyone really wants JWST to work, as so much has been sacrificed to make it possible.


Your $1 million university telescope probably wouldn't work long in space without maintenance, the ability to regulate temps, and rad-hard electronics.


See https://doi.org/10.1117/1.2031216 for the scaling of cost with size. It is very much not linear. Rather if you double the diameter the cost goes up by about a factor of 3.5 (closer to the area that went up by 4).


The university grade telescope, not sure what the entails, but I'm guessing it isn't hardened for radiation exposure and heat cycles seen in space. You can't use some off the shelf product unfortunately.

Making things survive in space is hard. Working functionally for the lifespan of the telescope is really hard. There's not a lot you can do once it's up there.


Aren't they mainly large so they have less hindrance of earth's atmosphere? Couldn't telescopes that are in space be smaller and have similar performance?


The are mainly large to increase resolution and sensitivity. Atmosphere limits how large you can make the mirror before hitting diminishing returns, but that is mitigated by good site selection, adaptive optics and lucky imaging (taking many exposures and keeping the least blurry ones).


Telescopes in space could have better performance, because you are not limited by atmospheric conditions for your angular resolution. But you still need a larger aperture to increase your theoretical angular resolution (see the Rayleigh criterion), and increase light gathering power.


Can't you have a fleet (something like StarLink, but telescopes) and have a bigger effective aperture?


Theoretically yes, through interferometry. But that is really difficult when there is relative motion within the constellation such as with starlink.


For radio we might be able to make it work. But we currently don't have the ability to position satellites to within 10 nanometers, which would be required to make this work in the optical range that you have just killed off on the ground.


Large aperture means more light makes it into the telescope. JWST has a mirror assembly bigger than most ground based systems. It comes down to cost vs performance. For the cost of putting a small-mid sized satellite in orbit you can build one hell of an observatory that will function longer and have a much bigger tube. At a certain point it makes more sense to use orbital systems but only for certain types of experiments.


$500m for the telescope. $7500m for the pork.


Is there a specific reason we're still doing ground based datacenters? With satellites becoming ever cheaper, sure we at some point should be able to get a significant datacenter up there right? Are we waiting for the bigger rockets to accomplish that?

That thought exercise should give you 90% of the answers to your question. The atmosphere and light pollution from cities are pretty easy to counteract with location and bigger optics.


Launching big telescopes into space is really expensive.


I have no experience in astronomy or satellites, but here's my naive idea. The article suggested erasing trails from the images using software. If SpaceX made an open API that detailed the precise location of every satellite at every point in time, could the imaging software use this to know that at this location in the image there is definitively a satellite that can be erased? I'm not sure what kind of sensors these telescopes use, and it probably wouldn't solve the issue of the bright spot messing up the exposure, but at least you could get rid of the trails?


SpaceX is publishing the precise location of each satellite continuously. It's really cool actually; most satellite operators don't do this. The raw data is available at Celestrak here: https://celestrak.com/NORAD/elements/supplemental/

I'm using this data to power my site that shows when you can see Starlink yourself: https://james.darpinian.com/satellites/?special=starlink-lat...


That's a really cool piece of software dude! Will definitely use this in the future to show some friends the starlink satellites.


The overlay on a nearby Streetview showing where the satellite will likely appear, with a timeline-based trajectory, is especially impressive.


This site is awesome! Have you thought about extending it to check for the position of the ISS? I would assume that information is probably readily available, as well.


Thanks! Yes, it can track most of the other bright satellites in the sky as well, try the base page link: https://james.darpinian.com/satellites/


Wow, this is great. Thank you!


What's the difference between the SpaceX supplied TLEs and ones derived from other sources? More accurate? Updated more often?


> More accurate? Updated more often?

Yes. There's only one other (public) source, https://space-track.org/ which publishes tracking results from US military radars. But the tracking results aren't as accurate as the satellites' own telemetry data.


That "where" feature based on street view is really nice


This is really cool


I've mostly dealt with amateur and semi-pro equipment and I'm not certain what the state-of-the-art is so someone else might be able to answer this better than me.

These telescopes essentially work by capturing photon counts on a sensor. The individual pixels on the sensor have a limit to the number of photons they can count. You could theoretically subtract the satellite pixel-count values from the photon counts to get rid of the trails. The two problems I see are: 1) You don't know the correct counts for the satellites and I'm not sure how you could get them. 2) The trails will probably saturate the pixels anyway (which can also cause bleeding into other pixels), in which case you just don't have the data of what's 'behind' the trails.


Tl;dr: Perfect subtraction is impossible due to physics.

Problem is: With photon count the uncertainty in the number of photons also goes up (the relative error goes down). So even if you know that you should have received 100 photons from the satellite (and have not reached the overflow of 256 in this example yet), Poisson statistics means you will actually get anything between 90 and 110 photons. So if you subtract 100 you have an uncertainty of plus or minus 10 photons left. That is deadly if you astronomical source only gave you 2 photons in that pixel in that time.


Are the sensors recording a timestamp with the photon counts? If so, it should be trivial to filter out bright fast moving objects with zero impact on image fidelity.


Photons are not individually time tagged (that is much harder and reduces quantum efficiency due to dead times), but rather are collected for some exposure time (think 30 seconds) and then their count is read.


one gets time signals from photomultipliers, but those are too bulky for traditional pixel sensors. And something as bright as a satellite flare might fry them at an unexpected moment. They are usually built and tuned to be able to observe few to single photons.


could they put an LCD in front of it, and block the relevant pixels as a satellite passes?


That would limit your resolution to the resolution of a LCD panel. Also just not how cameras work


it'd limit your dimming resolution, not your visual resolution.

I mean to use it like auto-dimming mirrors, or smart headlights. Block the bright stuff that you can easily predict, so you can protect the sensitive image sensors from over-saturating on garbage.


Nearly every object in orbit is already tracked and any specific object of interest can be tracked even more precisely, so no need for an additional API. That won't work as they use multiple hour exposures that accumulate the light, you can't "erase" something, as most of the info is already discarded, unless it's a deep stack. You can try recording the video of the same field of view with a smaller auxiliary telescope, accounting for the satellites that crossed the view of the big instrument in the resulting exposure. But that's still artifact-prone, and won't work in many cases (saturated pixels, very faint objects etc). You have to pause the exposure for every satellite that crosses the view, if that's possible at all.


It's possible with a shutter. Not easy, but feasible.


I mostly meant observational possibility, not a technical one. Astronomers rarely have photons to waste.


Telescope images are usually long exposures so that won't work. AFAIK there's no method of selectively dumping or damping the pixels only ass the Starlink satellites pass through the frame.

To do it you'd need a screen in front of the sensor that could occlude the pixels that the Starlink sats were passing over..


Astronomers do put a screen in front of the sensor to occlude pixels. This is to block the light of natural space objects that get troublesome.


Those are static though right? For starlink you'd need to have it moving.


They can be dynamic, with pixels.

One such technology is the optical vortex coronograph. Several types exist.


If you're doing image stacking, could you dump some frames that were ruined? It would take a bit longer to collect enough frames.


During twilight our section of EVERY frame would be ruined. Massively reducing the useful observation time during twilight. And keep in might that some sources are only visible to some telescopes during those hours. And we have neither the money to build arbitrarily many new telescope, nor the suitable locations with good conditions.


Going further, put a small space facing camera on each satellite. The images can be stiched together to improve resolution and maintain continuous observation of space at all angles. Image downloads can be limited to save bandwidth.


Are you joking?


Probably not. I see a lot of ridiculous "startup-hacks" for a hard science problem in this thread.


There is SW already to subtract satellite trails from pictures. And astronomers have stated that it's not a big deal yet. But 42000 extra satellites might be too much. Nobody knows, everybody is worried, lots of talk.


A couple years ago I moved near Lowell Observatory and our city, Flagstaff, has a number of Dark Sky ordinances, which are great. I've seen so many more stars than ever in my life. We just bought our kids a Newtonian reflector telescope.

Most things have trade-offs. Hopefully, the reduced cost of access to space will allow launching of more space-based telescopes, which don't have problems with atmosphere. Any astronomers here that care to explain what only ground-based observatories can do?


Not an astronomer but things you can do on the ground:

1. Bigass telescopes. Some radio telescopes are 100 or even 1000 feet across. You can't put something like that easily in a rocket fairing. JWST has a 20 foot mirror and they probably spent hundreds of millions of dollars figuring out how to fit that thing in an Arianne 5.

2. Easy maintenance and upgrades. There are 80 year old telescopes still doing useful science since you can make incremental upgrades over the years. (shoutout to Hubble: at 30 I think it's probably the longest lived space telescope, even if it's getting a bit long in the tooth)

3. Interferometry. This would potentially be really cool to do in space, since you could theoretically make a telescope whose diameter is tens of thousands of miles. However, interferometry requires you to be able to position yourself really accurately (amount of accuracy you need depends on wavelength and a few other factors, but potentially at the micro-meter level). That's really hard to do in space.

4. You can build a bunch of ground based observatories for the cost of a single space one.

5. Data downlink. Some telescopes generate a lot of data. It's not easy to get a terabit per second of data down to the ground.


You can make really really really big mirrors,. Space telescopes mirrors are limited by Fairing size, so barr folding mirrors ala James Webb, there is a fairly hard limit.

https://en.wikipedia.org/wiki/Extremely_Large_Telescope


That limit won't exist forever. Sooner or later we'll get space-based industry, and I'd rather that be sooner.

I'm as much a fan of astronomy as anyone, but I'm not willing to let it block the best chance we currently have of becoming a space-faring species. The comm satellites aren't important, but the launch capacity scaling is.


Launch capacity scaling isn't only possible through useless comsats.

And astronomy is still an extremely important component for becoming space-faring. No use in going somewhere blindly when you can have a look first. But if there is no-one looking because nothing to see, funding dried up, scientists demotivated,...


But why would you want that we become a space-faring species, at all costs? There is not much interesting to find in space, compared to what we have on Earth (especially not close to us, and I do not see us getting out of the solar system for a long time). Perhaps you could argue that we might do asteroid mining at some point, but at the moment that's very far out of reach.


Because we are limited in every conceivable way down here on Earth. Transportation is expensive, resource extraction is expensive, resources are limited, space is limited, pollution is a problem, energy generation runs into geopolitical, geographical and efficiency issues. And so on.

Never mind the problem of us being one asteroid strike or supervolcano away from a cataclysm, and that's not an if but a when.

I have a feeling that for humans to transcend the proverbial great filter, we have to tap into the vast quantities of resources and energy in the solar system, but more importantly rekindle the pioneering era that last ended with the industrial revolution.

The expensive part of the space industry is lifting infrastructure from ground to space. Moving within the solar system is comparatively cheap if we avoid descending into the gravity well of other planets. Luckily, this is unnecessary for most asteroid mining.

Humankind experienced incredible advances with the toppling of every transportational frontier. The wheel, seafaring, motorized transport and flight all resulted in expansions lasting a hundred years each.

The next frontier is the solar system. We don't know if we'll be able to ever leave it, but that's irrelevant because it can be our home for the next billion years. Our best shot at actually preserving the habitability of Earth is exploiting resources out in space.


I have no idea who downvotes such a wonderfully worded and inspiring comment, but surely, those willing to follow the space path and those willing to remain here on Earth should neither deny the other side of their dream nor enact "rules" that forbid it.

In fact, I'm pretty sure both "sides" would benefit greatly from cooperating. It might even be that these aren't "sides" but just different cousins of the family with different outlooks on their own life, and we need a little bit of everything, and everyone, to make a world.


Exactly. We need a backup. Right now, the Earth is a single point of failure.


It's all always going to remain out of reach for a long time if we don't try.


There is also the possibility that we miss an open window of opportunity. This might be a key resource running out, a plague or global war or even an asteroid impact.

That might close the window of opportunity and not give us a second chance. So better not risk it. :-)


I'd bet you a coffee that, barring a collapse or decline of our present global society, we will have launched a probe to another star before 2100 with a planned arrival before 2150.


Exactly, eventually you should be able to build much bigger telescopes than what can be reasonably built on Earth thanks to the lack of gravity, no horizon and no atmosphere to trash your giant scope with a storm.

Many kilometers in diameter in circumference should be possible and likely much more. The whole thing would at the same time be likely really really light, just thin stabilized foil, as it does not need to fight gravity or survive launch loads. Could be quite a sight. :)


>eventually you should be able to build much bigger telescopes than what can be reasonably built on Earth thanks to the lack of gravity

>Many kilometers in diameter in circumference should be possible and likely much more.

This is a misconception. There is plenty of "gravity". The mirror has to keep a very precise shape and attitude, which severely limits the possible size, considering it has to be light and is a subject to gravitational perturbations. Large and thin constructions in space (solar panels, antennas etc) are mechanically non-trivial on their own, and for telescope-quality mirrors it seems downright impossible.


I'd say it may envisioned theoretically (better materials, new photo-sensor processus, you-tell-me, physics say we have a much higher bound).

But in the short-medium term, the cheapest course that delivers is to use normal telescopes and interferometry (say on some orbit between Venus and Mars). I'm pretty sure it's also a domain where narrow AI may help because finding "anomalies" in space is a lot like finding anomalies on X-rays to find malignant tumors — something AI apparently can do well. Both problems fit incredibly large datasets + ultra low resolution of said anomalous blobs, and discrepancy with normal ones barely statistically significant (well below what human eyes may spot).

This is how I see the immediate future of space-based observation: lots of small things that cooperate extremely (increasingly) well with each other, "networks" more than "giants", much like down here on the ground.

It's just easier, cheaper, and lets you grab a lot of low-hanging fruits. Meanwhile, space-based fabrication can kick off and take the time to reach 'self-sustaining' velocity.


Do you know anything about development work that's been made regarding manufacturing mirrors in space? I'm not thinking assembling them from pieces, I'm thinking more about melting some raw material and using some physical process to create in space something larger.


Gravity is an issue for mirror production. But it is far less of an issue than tension and deformation during the months- to years-long cooling of the substrate. The temperature and environment control to do that is challenging on earth, fairly impossible in space. Except with maybe a really really huge spacestation as a thermal sink, which we won't get for the next few hundred years I'd guess...


Aren't the terrestrial mirrors so thick due to the need to survive all the tilting under the strong terrestrial gravity ?

I would assume a micro gravity only mirror could be much thinner & thus easier to cool down. Or possible alternative techniques could be used to get the needed reflective surface geometry if it does not need to take gravity and atmosphere into account.


Yes, but stability against vibration and thermally induced warping is also important. So you could make the mirror thinner than on the ground, but not really thin.


You also need a minimum thickness and stability if you ever want to turn the mirror to point elsewhere.


I've read a novel from Daniel Suarez titled Delta-v. Chapter 31, Alchemist beginning on page 175 and ending on 177 changed my assumptions about that. With not that much suspension of disbelief. It's about chemical vapor deposition at large scale, the chapter, not the whole book. That is about near term asteroid mining.

edit: book -> novel


The manufacturing tolerances for large mirrors are very small, and it seems unlikely that we would be able to do this - note that barely any manufacturing has been done in space.

Assembling from pieces on the other hand, is more doable.


A potentially more interesting solution is liquid metal mirrors, either mercury-based or some sort of ferro-silver suspension managed electromagnetically.

A hollow toroid could be spun to provide force and the liquid would naturally form a parabola.

There've been a few experiments on earth and it could have huge cost saving potential in space. Instead of having to launch a 8m+ solid mirror machined to perfection, you could launch a lightweight toroidal substructure with a vat of reflective liquid at a fraction of the cost.


I wonder if something akin to glassblowing would work in zero-G.


Aren't the newest and biggest ground based telescopes made up of many mirrors that have millions of actuators on them to correct for atmospheric distortion? I don't see why you couldn't make similar multi-mirror telescopes in space. Here is a link to an article[1] that shows three of the biggest telescopes that have 7, 798 and 492 individual mirrors respectively. The biggest one has the most mirrors.

[1]https://www.space.com/22505-worlds-largest-telescopes-explai...


Physics does not exclude such an instrument. Budget does.

Shipping complicated physical hardware to space requires that it absolutely must work on the first try, and never require maintenance for the lifetime of the instrument. In contrast, the large earth-based telescopes can be regularly updated, maintained, and debugged. Access for such activities costs as much as a plane ticket to Chile, instead of a dedicated mission to space.


Yes. You still have to fit it in.


That's one of the reasons I'm excited about SpaceX's Starship. It has a large diameter, so space telescope mirrors can be much bigger with less folding.


And it will (hopefully) be able to fly to the moon "for cheap" and put large, land-based telescopes in the best location there is for telescopes.


Not an astronomer, but ease of maintenance comes to mind. If a mirror is defective on a ground-based telescope, it's easier to build a replacement and install it. Hubble had a defective mirror which required an essentially dedicated Shuttle mission in the mid 90s.


Mounting new detectors, fixing broken things, swapping out (optical) filters and shear mirror size are all a big problem for satellites. No atmosphere is nice for IR, but current optical telescopes are 10 meters and up, much larger than even a Falcon 9 heavy can fit.


10 meters should be doable. You'd need a custom fairing.

Probably you'd want a fairing that holds the mirror on edge, so that it travels edge first as if looking to the side during launch. This would save on air resistance during launch.


Spoken like someone who's spent a bit too much time in KSP :)

There are more issues than just sheer mirror size in this case.

- Resistance to vibrations is a _major_ issue if the mirror is monolithic (as with HST) as opposed to segmented (as with JWST).

- The mirror also has to either be very resistant to extreme temperature changes (and the contraction that comes with it), or must be cooled before launch. For extremely large mirrors, the cooling assembly would add _tons_ of extra weight, since this would most likely be active/liquefied gas cooling, and the weight of the cooling medium alone would be significant in this case.

- Using a shape that's close to being a cylinder is great for cooling too, because cylinders have a very good surface area-to-volume ratio, which matters when you have to account for heat exchange. If memory serves, only ellipsoids are better (with spheroids being a better than ellipsoids, and spheres being ideal).

- The support structure for the mirror (the satellite fuselage proper) would probably have to fit in there too, at least for the first launch. That's not small, even though it can be made collapsible.

- All of the optical elements have to stay outstandingly precisely positioned within insanely tight tolerances. Any shifting could result in a lack of clarity, chromatic aberrations, or other issues, even if the mirror(s) aren't directly warped or damaged.

Even accounting for that, a fairing that's irregular around the long axis of the rocket is often undesirable. While it reduces (compared to the alternative), it can cause issues with rolling after launch, and introduces additional complexity.


If only there was a company that could launch telescopes into space and get above all these satellites and atmospheric interference.


If they could put relatively cheap cameras on the "back" of these satellites (without them being destroyed when facing the sun) and publish the data they collect, that might be enough to stem the flow of negative publicity.

Maybe even become a net positive for astronomy? Since the volume and density of the data collected would increase as the fleet grows, the value presumably increases as well, hopefully faster than the problems that they create.


That's not really how things work. Even with a billion little cheap cameras, you can't do anything because they still can't see nearly enough as one proper telescope, which they blinded. As someone other in this discussion said, if you can see it with your eye, it is way too bright.


Let's think about this rationally...is there a cost-benefit analysis for global internet access vs higher fidelity astronomy? Can astronomers use software and API's to minimize the downsides? Are the upsides for human inter-connectivity as great as claimed?


My initial reaction was NO, but then saner minds prevailed and you might have a point - to be able to get unrestricted internet in China or Iran or India (didn't realize how bad censorship was even here. And satellite internet is a criminal offense) would be amazing. We might still not get it though, since it's not just receiving but also transmitting, so surreptitiously using SpaceX internet might not be possible in those places.

If you can't solve that problem, then I don't think it's worth it just so some first world folks can YouTube when camping.


Accessing those satellites from China or Iran or North Korea and many other countries (though, why did you mention India?) will probably still be illegal.

But there are many poor countries with rural areas that have almost no Internet access and would greatly profit from it. Starlink seems to be especially useful for people in those areas.


Are there that many poor rural areas that could use and pay for internet? It seems necessary that they have both electricity and digital devices, which isn't really something you see in what I would consider poor rural area on a global scale. You might be thinking of a poor urban area instead, the poor rural areas in sub-sahara africa rarely have built up stoves, so requiring electricity is pretty far fetched.


Starlink is important not simply because it gives faster internet, but because it creates steady demand for space launches. The only way we can get cheap and reliable rockets is using them for something, and at this point rockets do not have many other uses. So if spacex gets "first world folks" spend their money on improving rockets, it's good for everyone!


I would think Space X isn’t even going to transmit to the ground in China.


It will be very interesting to see how China will react to this. Will they use the Tesla investment in the country to "convince" Musk to disable the signal when the satellites are over China? Or will they simply make the satellite receivers illegal and not care about the few who use them anyways?


Pretty sure SpaceX has to get licensing everywhere they broadcast (not sure how international waters works).


The will undoubtedly apply financial pressure, regardless of what other actions they take.


Rationally the company that was founded by a guy who thinks we need to start seriously considering space travel probably shouldn't be fucking up our primary way of studying space.


NRO already does this. "Stealth" satellites are a big area of interest right now.

https://en.wikipedia.org/wiki/Misty_(satellite)


NRO is willing to spend much more than SpaceX is on these and they're (probably) much larger giving them more volume to deal with the additional heat accumulation that comes with a darker satellite.


How much of the leading edge of astronomy is ground based, vs space based? Are they complementary or is space based going to eventually assume any and all roles ground based could?


That depends on the wave length regime.

Radio is firmly ground based, because you need huge dishes and potentially many (thousands) of them, at very precisely known distances.

IR is mostly space based (with SOFIA and ALMA the notable exception) because of atmospheric absorption.

Optical is firmly ground based, due to much lower cost for large telescopes. (See https://doi.org/10.1117/1.2031216 for the factors that affect cost). The notable exceptions are Hubble and satellites monitoring the sun such as Stereo and SDO.

X-Ray is space based again due to atmosphere.

Gamma-ray telescopes are an interesting mix between ground based air cherenkov telescopes (IACTs such as Hess, Magic and Veritas) and water cherenkov detektors such as HAWK and space based Fermi (with relatively poor sensitivity and low upper energy cut off, but very wide field of view).

Neutrino detectors are firmly ground based because the need huge detectors (the cubic kilometer of icecube is basically the lower limit).

So they are very much complementary. And some things will probably never moved to space, even if launch was free.


For radio telescopes, some Very Long Baseline Interferometry experiments have a space based component:

https://en.wikipedia.org/wiki/Very-long-baseline_interferome...



Slightly off topic but I love the fact that (over simplifying) some of our most advanced technology - for detecting neutrino - is a cubic km of ice.


It is very clean ice, with is shielded by another 1.5 km of ice on top and that is filled with more than 5000 photomultipliers.


If we are talking about visual light astronomy (the wavelengths where black satellites could help), a majority of astronomy is ground based. There exists just a single capable space observatory in visual light, the Hubble telescope, which will be going out of service in a few years. There's some metrics here[1] which show that about 10% of the most cited papers use Hubble data, which is still quite impressive. This is dated though, it could very well be that Hubble is becoming less relevant with recent advances in technology.

[1] http://www.stsci.edu/%7Ewebdocs/STScINewsletter/2003/spring_...


I see a cube sat and I want it painted black

No sparkles any more I want them to turn black.


I find this interesting considering that when the threat to astronomy from these swarms of thousands of satellites from dozens, maybe eventually hundreds, of different companies was first brought up on HN, there was a massive outcry from the technosphere saying it was a non-issue and saying that satellite internet is far more important than being able to see a natural sky.

At least SpaceX seems to be taking these concerns a little seriously.

Also, wasn't there a sci-fi TV show not that long ago that was popular on HN that had a theme song along the lines of "They can't take the sky from me?" I guess the lyricist was wrong.


Yeah it is interesting that people are more willing to sacrifice the night sky then exert political pressure to get the horrible internet situation in the US fixed.

Also interesting how people who could not find the big dipper think that removing the streaks is "just basic image processing" without knowing anything how modern astronomy is done. Never mind that professional astronomers are complaining. Oh and of course you get suggestions like "you can just fill a 30 km bubble with gas in space and use that". As another commenter put it nicely "everything is trivial when it is somebody else problem".


A lot of arguments in this HN discussion boil down to "It's not that bad and it doesn't matter because this will earn Saint Elon enough money will fix it!"

That's like saying "Sure, let the giant industrial conglomerate dump toxins in the drinking water. That way it can earn enough money to build a machine to clean the water up and sell it back to us and everyone will be happy!"

Why not just not pollute the water in the first place?


> A lot of arguments in this HN discussion boil down to "It's not that bad and it doesn't matter because this will earn Saint Elon enough money will fix it!"

This is a forum for engineers (for a broad sense of engineer). We like to talk about technology. We like to speculate about the future and about politics. Starlink is a cool idea, so it's not surprising there's enthusiasm for it.

We should not be put in control of anything, ever. If that weren't already common sense, you'd just need to put half a dozen policy threads from Hacker News in front of a congressional committee to have them warning of the dire effects of engineer influence. Stuff we create should be heavily regulated when it attempts to "disrupt" society, like Uber or a lot of Silicon Valley startups.

Uncharitably, you might say HN has a ton of Dunning-Kruger about anything not directly technology related. I wouldn't put it that way: it's everyone's right to speculate about politics, the future, and values, but most people here don't actually think they should be put in charge of anything.


> Uncharitably, you might say HN has a ton of Dunning-Kruger about anything not directly technology related.

It's even worse. It is people thinking that because they are brilliant in some technical field (JS frameworks, or compilers, or machine learning or whatever) they are also brilliant in every other technical field (be it astronomy, high performance computing or medicine).


The path to fixing ground-based broadband in America doesn't seem obvious given how deeply entrenched the big ISPs are. It's probably actually much, much cheaper to build, launch, and operate a thousands-strong satellite constellation than it would be to dig out Comcast and its ilk, and quicker too. That's an extremely depressing fact (assuming it's correct, being a semi-educated guess), but it is what it is.

But there's another dimension, which is that Starlink is supposed to be funding the only company that's materially doing anything new in the domain of spaceflight technology. For a bunch of nerds who grew up drenched in science fiction and promises of humanity's bright future in space, who've instead seen decades of regression in capabilities, and who live in the same world where my first paragraph is true ... that's very meaningful, far beyond the "fanboy" slur.

Not saying I come down on either side of the Starlink should/should not exist fence, but I think the motives and attitudes of its proponents are often misrepresented and that's no way to have a productive conversation about it.


Firefly. And ouch


I don't think that opinion has significantly changed.


How did we ever get here? Why is someone allowed to put this many satellites in the sky, and in the name of money and not science.

These lunatic leaders are taking it too far. I will have to put on my superhero coat soon, and create some balance


If SpaceX doesn't roll out a constellation like what SpaceX is proposing, someone else will, and that other entity (China, Russia, India, some up-and-coming African nation) will likely not be as receptive to the complaints of astronomers when they think their economic future rests on a quick, cheap Internet roll-out.


Can we stop with the whole "You've got to let an American company completely screw things up, because if you don't then a Chinese company might! And if a Chinese company does we won't be able to stop them even though when the American company screws things up we don't stop them anyway!" Routine. It's old and unconvincing and I don't think the people making this argument even really believe it. It was played out when Zuck said it to congress, let alone now.


What's stopping other entities from doing it...especially being able to say "Well SpaceX did it...".


Mainly electromagnetic spectrum allocations.


Why can't they just use modern AI tech to remove the satellites like we do with people? https://www.theverge.com/2019/6/25/18715676/camera-app-bye-a...


Astronomy telescopes are pretty different from the sorts of optical sensors used to take pictures of people. Through amplification they can detect very small numbers of photons. The error margin of trying to exclude something like a bright satellite in the field of view is going to be significantly larger than the detection threshold.


I imagine that the bright objects also make it difficult to see less bright objects so it's not just a matter of removing the brighter objects.


I see that, but we'll know exactly where the satellites are in the part of the sky we're watching. It should be possible to remove them automatically.


You have to give credit to SpaceX for tackling this thing head on. If someone else was in question, I don't think they would be as responsive.

We are also coming to a point where we have to acknowledge that sky is changing and astronomy or where we place telescopes should change. Depending on what is being observed.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: