frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

Open in hackernews

Sorcerer (YC S24) raises $3.9M to launch more weather balloons

https://www.axios.com/pro/climate-deals/2025/06/12/sorcerer-seed-weather-balloons
50•tndl•1d ago

Comments

ChrisArchitect•1d ago
Related from 9 months ago:

Launch HN: Sorcerer (YC S24) – Weather balloons that collect more data

https://news.ycombinator.com/item?id=41291219

xnx•1d ago
Good timing

"At least 10 sites have suspended or limited weather balloon launches because of the Trump administration’s cuts to the National Weather Service."

https://www.nbcnews.com/science/science-news/weather-balloon...

wturner•1d ago
Their agenda is to intentionally make government dysfunctional as a catalyst to privatize everything. It's working.
jmcgough•1d ago
Next up is the postal service.
mlinhares•1d ago
Problem is there is no money in this for the private market, so the government will have to pay 10x more to get a patchwork of private entities to do the work the government was already doing.

If we don't get these people out and revert all they did there won't be much of a future for this country that isn't becoming another failed economy with a few places being a little bit better than the others. The US will be a third world country soon.

fastball•23h ago
Yes, the US Federal Government, famous for doing things efficiently and cheaply when no private contractors are involved.
Retric•23h ago
The quiet parks of government really are, privatizing the VA was looked into several times but the numbers were prohibitive.

It took me a while to realize there’s just less money to be made when the system actually works.

fastball•7m ago
Realistically this is only because our private sector in that area is incredibly broken, not because the government is doing an incredible job.
Wololooo•21h ago
Some things cannot be done cheaply or extremely efficiently but needs to be done.

One of the reason for governments to exist is to deliver services to the communities that, by essence, should not be seen as primarily a source of revenue, but enabling communities to thrive in order to generate an environment where the generation of wealth for the community is facilitated.

And tell me how actually efficient and cost saving having many entities that have as sole mission to generate profit really helped (take a look at privatisation of any entities that were of public interest and the good that it did long term... Spoilers It didn't...)

mlinhares•20h ago
No natural monopolies should ever be privatized and there's no reason to have more than one entity collecting weather data. Also, this has been one of the best and most reliable services provided by the government, no one ever bothers thinking about the weather because its always there and it is also very right.

Maybe they were just too good and now everyone thinks this is easy and every single other country works like this with weather service. I'm in florida and I know i'm fucked this hurricane season.

fastball•6m ago
When Japan privatized their train system, the system improved massively.
guywithahat•4h ago
I feel like there's important context here, such as this company was founded before Trump entered office, and the government weather balloons were so inefficient that a private company thought they could compete with free data.

Government weather data is exactly the kind of thing that should be private.

artembugara•1d ago
Congrats, @tndl

You guys rock! Big fan

khazhoux•1d ago
Congrats to the team.

I gotta say, though, these days I'm used these announcements all being "Two-month old FooBar startup announces $850M angel round at a $38B valuation"

huslage•1d ago
This is an anti-pattern. Doing the work that the government should be (and was) doing and then selling back the data to them or others when the data should be (and was) public domain is absolutely terrible for society.

No one should fund this.

jaccola•1d ago
By this logic anything could become government run but never transition from government run to privately run, creating a ratcheting mechanism that would eventually lead to ~everything being government run!

The pro case for privatisation (that I happen to believe in) is: you were paying for it anyway, via your tax dollars, having it private leads to competition and stronger incentives to improve/cut costs meaning it will net cost you less.

huslage•1d ago
You are oversimplifying here. I ALREADY paid for the weather balloons and they are no longer being launched. This is not privatization in the way that you seem to think it is. This is explicitly against the will of the people.

I'm fine if they want to make new weather balloons and sell them to people to launch for whatever reason they want. Selling what by law should be public data is anathema.

krisoft•1d ago
Weather balloons are a recurring cost. It is not like you launch a weather balloon once and it provides data forever. You need to launch new balloons once the ones previously launched land. (This is typically a very short amount of time. Days not weeks.)

It is not like this company is going to take over the management of weather balloons you have already paid for. Or I don't know how you imagine this is going to work.

> This is not privatization in the way that you seem to think it is.

Can you tell me more about how you think it is?

huslage•1d ago
I imagine the government is going to start launching weather balloons again after they get sued for illegally firing the staff that's supposed to do it.
tartoran•1d ago
> Weather balloons are a recurring cost.

What do you think taxes are? Do we pay taxes once and that's it?

Alupis•1d ago
It's extremely unlikely any of your tax dollars were allocated to projects like what is being discussed here. It's much more likely (given the Federal Government's total budget and allocations) that this money was being borrowed and/or printed.

So, put another way, is it better for the government to continue going into debt to operate projects like this with potentially dubious returns - or better to allow the private industry to find a way to operate it instead?

skyyler•1d ago
>So, put another way, is it better for the government to continue going into debt to operate projects like this with potentially dubious returns

Yes. The government is not a business. It does not need to turn a profit. Government services are not products for the generation of profit.

Alupis•1d ago
Nobody said anything about profit. We don't need to move the goal posts here.

There is a difference between the government operating programs with tax dollars and operating programs with fantasy money that ultimately hurts every single citizen.

fragmede•1d ago
This framing is off. Weather data isn’t a fucking mars habitat, it’s core infrastructure. Airplane travel, agriculture, emergency response, and private weather services all depend on it. If anything, handing it off to profit-driven firms creates more risk with things like black-box pricing, gaps in data coverage, or national security issues.

And, “dubious returns” ignores that some of the highest-leverage investments in history looked like this. Government-funded satellite weather programs, GPS, and early internet tech weren’t obviously profitable, but I'm so glad we wasted tax dollars on that.

Alupis•1d ago
The government obtains weather data in other ways. This isn't an all-or-nothing thing based on balloons...
krisoft•1d ago
Huslage said “I ALREADY paid for the weather balloons and they are no longer being launched.”

Past tense. You could say that you have already paid for something where the cost is largely up-front. Like for example you could say it for the aircraft carriers. Imagine that (ad absurdum) the administration would want to sink all aircraft carriers. Then you, or Huslage, could rightfully say “I have already paid for the aircraft carriers…”. You could complain that your tax dollars are being wasted by sinking them.

But with a recurring cost like weather balloons the same sentence doesn’t make sense. There you could say “I have been paying for those balloons” (for which presumably you got the data you wanted from the balloons). Once they no longer are launching them, you are no longer paying for them. (Modulo some stock remaining on the warehouse shelves I guess. But that is basically a rounding error in a government budget.)

What Huslage said makes sense if they think of the weather balloons as a large up-front cost, like an aircraft carrier. Huslage already paid for them and now they won’t be used anymore! What a waste! But in reality it is more like a recurring cost. Like for example if the pentagon had a Netflix account and now they are canceling it. You wouldn’t say “I ALREADY paid for the netflix account”. You haven’t “already paid” for it. You were paying for it up until now, and you won’t be paying for it from now on.

There are many great reasons for why it is a good idea for the government to keep launching weather balloons. Huslage “already paid for it” is not one of those great reasons. It demonstrates a misunderstanding of how weather balloons work.

But do change my mind. Why do you think it matters that taxes too are recurring? How does that make the weather balloons “already paid”?

CGMthrowaway•1d ago
You are also simplifying. You didn't pay for anything. You were taxed, and representatives selected in accordance with a social contract between government and the people (the Constitution), apportioned and spent (or didn't spend) the money.
huslage•1d ago
That explicitly has not happened in this instance. Don't tell me how democracy works.
CGMthrowaway•1d ago
Which part? And I wasn't describing democracy
throwawaymaths•1d ago
can you explain by what criteria you decide what data should be public or not?
thrance•1d ago
How's that working out for public health in the states?

Weather reporting is a common good. It worked very well for pennies and benefitted the economy greatly. Why privatize it?

Workaccount2•1d ago
There is a key difference, privatization means a flat cost, whereas public means an income based cost.

About 30% of Americans get (NWS) weather data for free. They pay no income tax yet receive the same level of public benefits. On the other hand, a handful of Americans pay millions for weather data, and receive the same thing as those who paid nothing.

For a private service though, it would just be $20/mo or whatever for everyone.

gusgus01•1d ago
Where did you get 30% from? I'm just curious since NWS data is widely used as a source for creating weather forecasts, which if I had to guess near 100% of people use in one way, shape, or form. I think Google uses it, so anyone with an Android phone is one click away from a forecast using the data.

On the matter of taxes being proportional to income, I'm not going to argue about progressive taxation or any moralistic standpoint of proportional taxation. From purely a utility standpoint, those handful of people probably reap way more value from that NWS data being available. The richest people (those paying the millions for NWS) usually are that rich from the labor of others, and those labor forces all get value from the data to help plan their days, including getting to the workplace safely. Another even more direct use for the economy would be routing of trucks through snowy passes, or planning for large construction companies.

Workaccount2•1d ago
~30% of Americans do not pay income taxes, i.e. they get public services like NWS data for free.

Nothing else you said is wrong, I'm just saying that government services are effectively progressively priced based on income.

gusgus01•1d ago
For the first part, I totally misread that several times. Sorry about that.

Looking around, the exact number is quite hard to pin down because of the definition of it, but ~30% is probably a very fair estimate based on https://taxpolicycenter.org/taxvox/tpc-number-those-who-dont...

shaism•1d ago
Do you mind elaborating?

I agree that weather data should be public but I don’t see why we should restrict innovation in the private market if there is demand for it.

Also more generally, I see no issue in the government outsourcing work to a competitive private market wherever possible.

huslage•1d ago
I have no issue with it if it is part of a legislative/regulatory framework. This is not inside of any framework. There has been no conversation about privatization of NOAA or any of its functions. These things need to be explicit as part of a democracy.

The current regime has upended that process and has created a situation where the government has no choice but to outsource data gathering to third parties. This is corruption and not in the spirit or the letter of the law.

This startup is attempting to take advantage of an illegal situation which is just ridiculous.

I'm happy if they want to sell fancy weather balloons to anyone that they want, even the government, but selling data back to the government that should be already collecting the data in the first place BY LAW is just corrupt.

frankdenbow•1d ago
The reality is that the government is not doing it, so the choices are to sit back and watch things crumble or have private companies work to try to fix things. I agree with you in principle that its a sad state of affairs though.
huslage•1d ago
This is an excuse for normalization of a state of affairs that cannot continue to exist outside of a proper legal framework.
mcdow•1d ago
I can see why that might be frustrating. What about this problem makes it the best fit for the government to handle? Is it prone to natural monopoly? There are lots of things that the government can handle and shouldn’t. Just because the government handled something in the past isn’t a reason in-and-of-itself for it to resume handling it in the future. I’m genuinely curious as I am ignorant of the space.
jmcgough•1d ago
What private industry has been pushing for for decades is privatizing weather data so they can sell it at a profit. But weather information is a huge public good and has been provided by the federal government for decades. Privatizing it adds more costs to public research, and means that people who don't have money to spend on weather forecasting - those living in poverty and most at risk when it comes to life-threatening storms - will likely die in higher numbers from severe weather.
fastball•23h ago
"People will likely die" is a pretty bold claim. Do you have a citation that deaths were avoided because poor people saw a weather report?
CGMthrowaway•1d ago
Why can't private enterprise do weather?
huslage•1d ago
There are many private weather companies.
fastball•23h ago
Right, so why should "no one fund" another private weather company?
bigveech•1d ago
Just to put it in perspective: it costs $300–500 to produce a single atmospheric profile with current balloon infrastructure. The U.S. launches ~180 a day—that’s at least $54K daily. Not exactly “pennies.” :)

And the government already buys the helium, radiosondes, and ground systems from private vendors—so the money’s going to private industry anyway. It’s just inefficient.

With 50 of our systems doing 4 profiles a day (which is no where close to max scale), you get the same volume of data for way less. And on top of that, because we reach remote and oceanic areas that aren’t being measured today, the data is also more valuable!

Also, the data you’re referring to isn’t inherently public domain. It becomes public when the government buys it and redistributes it. That’s true whether they pay for the infrastructure themselves or buy the data directly from a company.

huslage•1d ago
I'm glad you have come up with something more efficient. The problem has nothing to do with efficiency. You are welcome to make a government contract to sell them equipment or data as you wish.

My problem is with baseline services that have already been stopped that you claim to want to replace. This data feeds all of our weather models and should be done with existing infrastructure until congress changes things. The data must be freely available.

The fact is that the data is available for anyone anywhere and is a valuable resource for scientists everywhere. Your current goals might be laudable right now, but that is not going to be the case when you have to pay back an investor 100x in 5 years. You will do everything you can to lock that data up and make it as expensive as you can. You will have no choice.

Daviey•1d ago
It largely depends if it is bought by a public service under a license that allows public release, surely?
khazhoux•1d ago
Flawed premise. Weather modeling is not an essential service of a government
huslage•1d ago
Tell me that after your house gets leveled by a tornado and you don't get a warning.
msgodel•1d ago
Better to put all the PE money to work on it than borrowing at 5%.
counters•1d ago
It's not - it actually the core mechanism through which the "Weather Enterprise" works. Over 20 years, an important report from the National Academies [1] laid out how an enterprise comprised of public, private, and academic sector interests could work cooperatively to bolster the public good that is weather and climate information and services. It has always been the domain of the federal government to provide core, foundational data products (including forecasts and raw weather observations of many modalities) for both bolstering academic research as well as private sector innovation. The government's mission in the enterprise leaves plenty of room for private sector players to extend, complement, and supplement the foundational services provided by the public sector.

Sorcerer fits perfectly into the existing framework of the weather, water, and climate enterprise (WWCE). They produce complementary data and ensure that the government has access to it - even if the government must procure it (which they're happy to do - no one expects that these companies should give away all their data, gratis). But they could potentially greatly extend the core global synoptic observation system that powers conventional numerical weather prediction, especially for organizations which are more flexible and can work with broader data sources.

This is the WWCE working well. The real concern is on ensuring continuity - making sure innovative companies like Sorcerer can persist, in perpetuity if necessary (or at least the data products they collect and produce).

[1]: https://nap.nationalacademies.org/catalog/10610/fair-weather...

guywithahat•4h ago
This is exactly the kind of thing that should be done by private companies, what are you talking about? This data should be something companies compete at to get better at
mandeepj•1d ago
Google had something similar as well. I think they phased it out.
rokobobo•1d ago
I believe theirs was to deliver internet
thrance•1d ago
Welp. Good for them I guess, too bad weather reporting was gutted and is no longer a public good.

At this rate, we'll soon see YC startups raising money to maintain roads.

hodgesrm•21h ago
> At this rate, we'll soon see YC startups raising money to maintain roads.

That model has existed for hundreds of years. It's called a turnpike. There were hundreds of them in the US and UK. [0] Many still exist today.

[0] https://eh.net/encyclopedia/turnpikes-and-toll-roads-in-nine...

selimthegrim•1d ago
Will they have a sick electronic music track as they launch?
jzig•1d ago
Isn’t this the point of NOAA?
pfych•23h ago
NOAA might not be around much longer! ... which is grim.
andrewstuart•1d ago
What happens to weather balloons do they just drop out of the sky? Do they ever hit anyone? Are they just garbage the gets left on the ground?
Daviey•1d ago
Yes, I found one in a car park many years ago. Disposed of the balloon, but gave the radar reflector to the local school geography department.
tartoran•1d ago
We should be expecting more government services now going through YCombinator. Lots of profit to be made in the name of efficiency.
neonate•1d ago
https://archive.md/gTnOw
pyman•15h ago
I can't access this site... it's behind a paywall

Implementing Logic Programming

https://btmc.substack.com/p/implementing-logic-programming
75•sirwhinesalot•5h ago•17 comments

AI Isn't Magic, It's Maths

https://zerofluff.substack.com/p/ai-isnt-magic-its-maths
7•squircle•2d ago•0 comments

Self-Adapting Language Models

https://arxiv.org/abs/2506.10943
102•archon1410•7h ago•33 comments

The International Standard for Identifying Postal Items

https://www.akpain.net/blog/s10-upu/
24•surprisetalk•1d ago•0 comments

OxCaml - a set of extensions to the OCaml programming language.

https://oxcaml.org/
263•lairv•12h ago•77 comments

RISC-V in AI and HPC Part 1: Per Aspera Ad Astra?

https://www.eetimes.com/risc-v-in-ai-and-hpc-part-1-per-aspera-ad-astra/
14•fork-bomber•3d ago•1 comments

Whatever Happened to Sandboxfs?

https://blogsystem5.substack.com/p/whatever-happened-to-sandboxfs
24•zdw•2d ago•5 comments

I convinced HP's board to buy Palm and watched them kill it

https://philmckinney.substack.com/p/i-convinced-hps-board-to-buy-palm
462•AndrewDucker•8h ago•408 comments

Meta invests $14.3B in Scale AI to kick-start superintelligence lab

https://www.nytimes.com/2025/06/12/technology/meta-scale-ai.html
402•RyanShook•13h ago•419 comments

Using computers more freely and safely (2023)

https://akkartik.name/freewheeling/
64•surprisetalk•9h ago•6 comments

The Hat, the Spectre and SAT Solvers (2024)

https://www.nhatcher.com/post/on-hats-and-sats/
78•todsacerdoti•11h ago•5 comments

Show HN: Tattoy – a text-based terminal compositor

https://tattoy.sh
152•tombh•12h ago•43 comments

100 years of Zermelo's axiom of choice: What was the problem with it? (2006)

https://research.mietek.io/mi.MartinLof2006.html
97•Bogdanp•11h ago•96 comments

Endometriosis is an incredibly interesting disease

https://www.owlposting.com/p/endometriosis-is-an-incredibly-interesting
11•crescit_eundo•3h ago•0 comments

When random people give money to random other people (2017)

https://quomodocumque.wordpress.com/2017/06/27/when-random-people-give-money-to-random-other-people/
73•munificent•9h ago•53 comments

If the moon were only 1 pixel: A tediously accurate solar system model (2014)

https://joshworth.com/dev/pixelspace/pixelspace_solarsystem.html
692•sdoering•17h ago•212 comments

Apple's Liquid Glass is prep work for AR interfaces, not just a design refresh

https://omc345.substack.com/p/from-skeuomorphic-to-liquid-glass
168•lightningcable•6h ago•194 comments

Student discovers fungus predicted by Albert Hoffman

https://wvutoday.wvu.edu/stories/2025/06/02/wvu-student-makes-long-awaited-discovery-of-mystery-fungus-sought-by-lsd-s-inventor
65•zafka•3d ago•49 comments

A Study of the Winston Red: The Smithsonian's New Fancy Red Diamond

https://www.gia.edu/gems-gemology/spring-2025-winston-red-diamond
15•bookofjoe•4h ago•1 comments

High-speed fluorescence light field tomography of whole freely moving organisms

https://opg.optica.org/optica/fulltext.cfm?uri=optica-12-5-674&id=570897
35•PaulHoule•3d ago•1 comments

The concurrency trap: How an atomic counter stalled a pipeline

https://www.conviva.com/platform/the-concurrency-trap-how-an-atomic-counter-stalled-a-pipeline/
88•delifue•6d ago•43 comments

How the Alzheimer's Research Scandal Set Back Treatment 16 Years (2022)

https://www.discovermagazine.com/the-sciences/false-alzheimers-study-could-set-research-back-16-years
64•walterbell•4h ago•35 comments

EDAN: Towards Understanding Memory Parallelism and Latency Sensitivity in HPC [pdf]

https://spcl.inf.ethz.ch/Publications/.pdf/shen-ics-2025-edan.pdf
15•matt_d•3d ago•1 comments

MUMPS

https://en.wikipedia.org/wiki/MUMPS
73•surprisetalk•5h ago•64 comments

Ask HN: How do I give back to people helped me when I was young and had nothing?

303•jupiterglimpse•12h ago•169 comments

Simulink (Matlab) Copilot

https://github.com/Kaamuli/Bloxi
29•kaamuli•7h ago•5 comments

Jemalloc Postmortem

https://jasone.github.io/2025/06/12/jemalloc-postmortem/
709•jasone•1d ago•214 comments

Frequent reauth doesn't make you more secure

https://tailscale.com/blog/frequent-reath-security
1138•ingve•1d ago•480 comments

Thiings

https://www.thiings.co/things
159•pentagrama•3d ago•72 comments

Geometry from Quantum Temporal Correlations

https://arxiv.org/abs/2502.13293
47•ljosifov•13h ago•21 comments