Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Starlink Signal Leakage Threatens Radio Astronomy's Most Critical Frequencies (gadgetreview.com)
135 points by upofadown 22 hours ago | hide | past | favorite | 83 comments
 help



I realize this is off topic, but why do people use Claude of all models to write these, when its style is so obvious and so grating? I just immediately hit Back on the first Claudism.

Not to mention 13 em-dashes in a 500 word post. At this point you may as well ask Claude yourself about news and cut out the middleman. That'll also cut into their attempts to manipulate your opinion.

Did you care about the writing style of Rex Edison before and now you are disappointed?

That page is not the New Yorker.

And don't get me wrong, but the writing of normal newspaper became shit through the internet not due to claude. You have the clickbait header, than you have the subtext which repeats the clickbait header, then you had the first paragraph which repeats the subheader, than you had the second paragraph giving you some random 'background infos' and then you got the gist of the story in the 3 or 4th paragraph.


Clickbait headlines have a long history. Go look up old newspapermen’s reminiscences about them. The most famous are classics like “HEADLESS BODY IN TOPLESS BAR” (1983) from the NY Post or “STICKS NIX HICK PIX” (1935) from Variety.

Once starship is running, spacex should launch a lunar crater telescope https://www.nasa.gov/general/lunar-crater-radio-telescope-lc... to permanently shut up this kind of grievances.

While I pine for a dark sky, it makes sense that these operations should move in the same way century old optical telescopes in cities are no longer used for cutting edge science.

The far side of the moon seems like an increasingly viable option.


People need light to see in cities. This light serves a function.

Starlink satellites are emitting unintended and useless EMR that serves no function. It is not necessary for the function of these satellites.

This is a money issue. The question you should ask is whether it will cost more for SpaceX to fix their satellites than it will to build radio telescopes (which are not small devices) in space. Personally, I suspect launching radio telescopes into space will not be the cheaper solution.


> People need light to see in cities. This light serves a function.

Street lights with (full) cutoff will still provide light for people:

* https://www.blancocountynightsky.org/guidelines.php

* https://highways.dot.gov/safety/other/visibility/fhwa-lighti...

* https://www.zgsm-china.com/blog/light-distribution-of-outdoo...

In fact it will improve the situations as all the light will be directed down, and so you'll need fewer lumens/lux/candela since half won't be waste going up (saving electricity), less glare on the roads after rain, and folks won't have street light blasting through their windows.


So because 10 Million people want to have internet, 8 Billion people no longer get acccess to that resource?

That sounds like a nice attitude from you.


There's a lot fewer than 8 billion people who have access to the resource of "radio astronomy". I suspect it is in fact less than the 10 million figure of people who might want to use satellite-provided internet.

Exponentially more people benefit from Starlink than from radio astronomy. It doesn’t need to be a one or the other conversation though.

Its not just radio astronomy, its also optical.

And no i don't think so. Radio astronomies research results are fundamental for all of us, our understanding and for modern technology.

This makes it nescsaary that you have to send something away out of earth. Is Spacex sponsoring this type of research? No. So who has the researcher go to? To someone with money.


You do realize that the far side of the moon is in direct sunlight for 2 weeks, right?

It’s also in darkness for 2 weeks. Terrestrial radio telescopes experience sunlight too. That would seem to be a solved problem.

and the moon is over 200°F in that sunlight.

There is no atmosphere to speak of, so only if directly exposed. Hence that's a solvable issue.

The whole thing is incredibly expensive though, while it may be doable with current technology, the amount of money they'd have to set aside may be a hard sell for the miniscule potential of this research - at least from the lens of 2026... So the original position of the article makes sense, it may become functionally impossible, even if technically possible


So, less than the boiling temp of water. That is not outside the realm of solvable problems

Instruments on the far side of the Moon will need some communication with Earth. While you can fantasize about running very long surface cables, the common solution consists of orbiting communications satellites.

There are currently at least 8 satellites orbiting the Moon. We can expect that number to increase. Also to add to the excitement, when a satellite deorbits there, it will never slow down, burn, fall into water, or disintegrate. Just one big hunk of equipment impacting at full speed. Enjoy your craters.


And ? You are just describing how the lunar environment works.

There are more peculiarities, like most orbits being long term unstable, abrasive dust, etc.

But nothing that can't be habdled with good engineering.


Just wait until wireless networks on the moon ruin that too.

Or Space Karen can fix his space junk...

Maybe even start paying for the toxic waste he is landing when his junk continuously fails to stay up and re-enters the atmosphere.


There is some evidence that little engineering effort has been put into mitigating this problem. I think if someone actually tried really hard we could make these birds a lot quieter.

> We compared the detected power flux densities of this UEMR to that emitted by the first generation v1.0 and v1.5 Starlink satellites. When correcting for the observed satellite distances, we find that the second-generation satellites emit UEMR that is up to a factor of 32 stronger compared to the first generation.

https://www.aanda.org/articles/aa/abs/2024/09/aa51856-24/aa5...


There's probably a reason for that.

> Current ITU regulations ignore unintended satellite hardware leakage, leaving astronomers without legal protection.

Oh, what a coincidence.


Another reason is these emissions are from power supplies

> Algorithmic mitigation is being explored, but scientists describe it as “embryonic” — potentially demanding computing power that rivals the science processing itself.

It sounds like this would work? Just be more costly, at least in the short term.

> Engineering fixes are the real answer

What is the fix and tradeoff? The article doesn't seem to cover it at all.


> The accidental static bleeding from onboard hardware, coupling through satellite structures and radiating across frequencies nobody authorized.

> The numbers tell the story bluntly:

I wouldn't waste too much compute trying to parse it, lots of LLM tells in the article. Pangram puts it at 100% AI.


maybe Internet for people on earth matters more than astronomy

We had that before Starlink flooded the radio quiet zones with leakage.

Were that the case, there would have been nobody buying Starlink.

As they are, I think you are wrong.


My neighborhood has both cable and fiber, both >= 1Gbps, fairly cheap

My neighbor has starlink. I assume because they are MAGA.


okay for real, i think you have elon derangement syndrome.

I'm Canadian, I disapprove strongly of Elon and his antics, and there are cable and fibre options in my neighbourhood, but Starlink is cheaper and I need to pinch pennies everywhere I can. And the service is excellent.

Not everyone lives within a metro that has good infrastructure. This is even more true in developing countries.

More tech people need to go places.

I was living in the eastern Caribbean ten years ago (a place that is not particularly poor), and the number of web services that could not handle a slow or unreliable connection was kind of shocking. Netflix was actually pretty impressive, they were fine to about 500kb/s with a lot of dropped packets. Below that, they were slow, but by no means unreliable. Most other sites would shit the bed on a similar connection.

Starlink has been an absolute godsend for my friends that are still down there.


You mean privileged people in developed countries had it. Billions of people did not.

I’ve seen the first hand impact of starlink in communities that before had dialup as the fastest option from Australia to Tunisia to remote arctic Canada.

Medical stuff, school, talking to loved ones.


These are either incidental or unauthorized transmissions. Not the Ku band downlink to terminals.

They are leakage from switching mode power supplies.

I bet the 137 isn't though since that's a common satellite downlink band. Swarm who they acquired used that

Maybe but people already had plenty of internet before Starlink.

Everyone i know who is using starlink doesn't use starlink because htey need starlink for internet.


Is that because everyone you know lives in a high-wealth country/area?

Yepp. But if you look were Starlink is available, thats true for sure for a few more from the current 10 Million customers.

So is it worth it for a handfull of people on our planet vs. destroying our atmoshpere and removing access to our space?

I don't think so.


> maybe Internet for people on earth matters more than astronomy

Maybe Starlink can design proper filters and shielding so both Internet and astronomy can happen at the same time.


As if most people use the Internet for proper utility and not just creating and consuming outrage-bait and slop. Suddenly the cost-benefit of "internet for people on earth" seems a lot less genuine.

You need surprisingly little bandwidth to send and receive messages, make calls, and read the news.


I have such a love hate relationship with the tech. On one hand, I live on the edge of the mountains and can get away while on-call, set up my starlink and still be within SLAs if I got paged while out of service (weekend pages are fairly rare). On the other hand, I'm an amateur astronomer, for visual astro, it's not a huge deal, but I see how it could be for astrophotography (streaks) and obviously it's a problem for radio.

The two aren't mutually exclusive, as long as aluminum foil is within a trillionaire's budget.

Google "Ukraine war Starlink". Neolibs who hate Musk's companies unconditionally are going to go quiet real fucking fast, I bet.

[flagged]


“Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith.” --https://news.ycombinator.com/newsguidelines.html

[flagged]


I’m a human that just so happens to disagree with you. I don’t like Elon either, but what I dislike more is that it’s impossible to talk about any of his companies without people like you assuming my motives.

What precisely are your motives, then?

The umpteenth example of Musk companies behaving poorly and here you come to suggest that we can’t let pesky things like science stand in the way of “the internet”. As if Starlink services is synonymous with the internet itself. The implication is clear, to me at least.


Yeah Starlink is clearly so much better than the satellite internet that came before it that I think it's reasonable to want to expand it, even at the cost of making radio astronomy on Earth's surface harder. Fast, reliable internet access in places where it's infeasible to string high-bandwidth physical connections is a pretty valuable human good, and radio astronomy can be done in space (perhaps using technology similar to Starlink satellites themselves!).

Precise motives? I don’t know, I don’t think I have any. I don’t use starlink myself or have a Tesla or invest in any of his companies. Feels silly to disclaim that, but there you go.

Having used starlink before on a plane and on land, it’s fast and seems far superior to previous generations of satellite internet. I’m earnestly asking if creating problems for astronomers is outweighed by giving more people access to the internet. Not just giving them “access”, but giving them very fast and low latency access.


That might be dispositive if we didn't already have almost everybody on Earth wired up without using satellites.

The fact that so many people are paying for starlink suggests that it is adding something important that was missing before.

Starlink has 12 million subscribers, or about 0.14 percent of the population of Earth. In addition to paying for Starlink, those subscribers are externalizing costs to everybody else on Earth... for instance by screwing up radio astronomy.

It is unreasonable to talk about allowing 0.14 percent of the people on Earth to have Internet as if "Internet for people on earth" were at stake. I mean, yes, they are people, and they are on Earth. You could say the same if it were two people... but that's not the slant intended by phrasing that way.

It's especially unreasonable when many of those 0.14 percent want Starlink because of their personal lifestyle choice to live in the middle of nowhere. Yes, I know, somebody's grandpappy built his farm with his own two hands and the help of his trusty plow goat, Irving. That doesn't mean that person is entitled to infinite subsidies in the form of satellite access, fiber, roads, electricity, hospitals, or anything else. Economic activity can justify some of those things (and a lot of that can be priced in by markets). If you're not doing something useful, and you can't pay for all that stuff including traditionally-externalized costs, then move.


Streaming video on airplane flights?

Most of Starlink’s customers are simply rural people who want Internet access at decent speeds. There are wide swathes of the US where Starlink’s best competition is measured in kilobytes per second.

Seems like a case for the US having better fibre deployment than it does for leaking interference from unshielded poorly made tin cans.

Fibre rolls in a dense city have different economics from somewhere with just a few people per square mile.

and yet Western Australia, 3x size of Texas with less than a million people in 1980 had a telephone network that covered the population.

The "sums are different for rural V urban" doesn't make leaky tin cans a better solution over fibre to population, particularly given the downsides.


Pretty sure we achieved that level of coverage here in the US a decade or more earlier than that. A lot of those same rural customers still rely on the telephone system for Internet access. They might be able to get DSL now, but the speeds are going to be low; often just a few hundred kilobits per second.

So pull modern fibre; as should be done.

And who pays for those miles of fiber? That one customer, because he chose to live there?

I assume you understand that it that's the case, Starlink is a much better option as it won't cost them 5 figures upfront


This is beside the point. Starlink exists and provides excellent service to customers who have no alternatives.

Don't even need fiber. Starlink is only on par with 5G.

~80% of Starlink customers are not in the US.

That'd be a point to make to the GP db48x

No, I mean globally most of their customers are rural. Starlink is limited primarily by customer density rather than by absolute numbers. They can provide service to just so many people per square mile before the speed drops below the customer’s tolerance. Most of the square miles in the world are rural, therefore most of the customers are as well.

Plus cities have Starlink’s best competition. Starlink is amazing but FTTH is even better. If you can get FTTH at all then you probably won’t even consider Starlink. Nor should you.


Suburban+Urban is more than rural.

Rural more than Urban


Got a source for that?

Because there's entire continents of people who think it's the most ridiculous statement they've ever read on this site (which they can only access because of Starlink).


That’s a mighty nice field of study you have there. It would be a shame if something happened to it

[flagged]


An endowment isn't free money and launching a telescope into space remains a costlier venture than plopping a few arrays on mountains like Mauna Kea or atop the Chilean Andes. Contrary to your imagination, radio telescopes aren't that cheap either. The Smithsonian and NASA fund most of the construction, operation, and maintenance of Harvard's telescopes rather than the university. Suffice it to say, it's unlikely an academic institution will find an economically viable route to launching space telescopes any earlier than twenty years from now.

Sure, sure. The endowment generates revenue which is then spent on stuff. Harvard spends billions every year on stuff. A small slice of that spending over the last decade could have gotten them a huge scientific and prestige boon: the first major private telescope in space. (Cubesats don’t count.)

Government subsidies might explain it though. Why should Harvard spend their own money when they can spend mine?


What you call "the endowment " doesn't exist in any monolithic sense. "The endowment" is a disparate collection of donations from alumni such as scholarships, artwork, buildings, etc., governed by their respective, specifically-worded contracts. Many of these gifts aren't easily convertible to cash. In fact, a large cash gift is rarely offered at all and rarer still would any such gift come unconditionally.

> Government subsidies might explain it though. Why should Harvard spend their own money when they can spend mine?

The government's justification has historically been that the research value gained by putting the brightest minds to work at the frontier of academic science is worth the allocation of necessary monies. Whether that is (or has been) true is obviously debatable.


Cubesats can’t take very good pictures. You need a big mirror to get really clear imagery and you can’t cram one of those in a cubesat. Space telescopes are by nature massive, heavy satellites unless they aren’t intended to look very far or see very much. The other option is to cheaply build something absolutely massive on Earth.

I know, that’s why I said that cubesats don’t count. There are already lots of cubesats with telescopes in orbit, but they don’t count because you couldn’t do good astronomy with them.

Ground constellations (eg: SKA in radio astronomy) are still vastly cheaper than orbiting arrays, also easier wrt incremental upgrade via physical visits, etc.

Upgrades are overrated. Stick your telescope in sun–synchronous orbit and leave it there. Launch a better one every year or three. After a few launches your undergrads will still be excited to use that first telescope, your PhD students will use the next several, and your faculty will all be drooling over the latest.

Upgrades don’t increase capacity.

Oh, and we’ve got universities spending billions of taxpayer money to carve up mountains in Peru to build telescopes on top. Those are not cheaper than space telescopes would be.


Again: Radio astronomy.

> universities spending billions of taxpayer money to carve up mountains in Peru to build telescopes on top

Which university and telescope are you talking about?

> Those are not cheaper than space telescopes would be.

In general, they are - take a ground based telescopes costs, now harden it for space (extra cost), now "do the right stuff" to ensure it will work in isolation (more cost), and now package it it and launch it (more cost again).


Another example of billionaire wait trillionaire power capture. It’s really tough for democracy when one man has all that power. Hah Elon is lex luthor.

I just have to say: I _hate_ with a white-hot fury the growing usage of html-load and error-report dot com bullshit on these sites. I block those domains for gratuitous ad-serving, and sites just cycle a modal error now.

(I know, I could disable JavaScript everywhere and enable it on a site-by-site basis. Maybe that time has come.)


I’ve got JavaScript disabled and no websites work.

And nothing was lost

No problem, Starlink will sell us sattelite telescopes to solve that. /s



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

Search: