Hacker News new | ask | show | jobs
Spy Balloon Simulator (spyballoonsim.hornetsnestguild.com)
442 points by benryon 1217 days ago
32 comments

Hi I am the creator of the site

to answer everyone's questions: 1. it is supposed to be a for fun sandbox, a 'toy' if you may 2. it can in no way become a credible source for any level of military intel 3. the altitude is at 100hPa 4. I really should add a disclaimer about how this is for fun purposes only 5. adding othe levels will give me even more data chunks, but it is certainly doable 6. I want to add (programmable) steering too! will do soon

I’m having an absolutely awful day. My Dad had to call 911 to get my stepmother the care she needed and everything just went downhill from there.

As rough as my day has been, I have spent a lot of time with your app today. It’s not perfect (of course) but I have had a lot of fun with it today. As dark as today has been, that really means a lot.

You have gotten some really solid feedback today and I can’t disagree with it. But speaking as a user, this product has made my day significantly better. Thanks for putting this out into the world.

thank you for your support!

I hope your stepmom and dad all the best

A little bit of UX feedback: It would be great if the balloon were a draggable object, not just click once & then again, I tried to do this about 10 times before I realized: click once on the balloon and then again on the map! Or maybe edit the instructions (click balloon and then map).

Either way, I enjoyed this greatly! Thank you for building this.

I see, thank you for your feedback!
I absolutely need a cloud to blow on it from time to time!1 jk, great sim!
I love the disclaimer "not a credible source for military intel". I know you are probably putting that to disclaim knowing where a particular balloon came from but it sounds a bit like "If you want to spy on another country, build your own simulator" - lol.
well I'm putting that to stop people from screaming about my site's 'credibility' and actual military offices have way more advanced tools anyways

but yea some people just won't stop so

Thank you for making my Antarctica balloon simulator dream a reality.
haha antarctica is a fun region to simulate

at around 60 deg south there is no land at all, causing the wind and sea current to be exteremly strong, meaning the balloon can 'circumnavigate' the globe in time as little as 5 days, it is also known as the screaming sixties

Isn't the actual distance to cover much shorter also at those latitudes, in order to 'circumnavigate' the globe? Because an equilatitude circle is much smaller.
Correct. Look at the specifics for Burt Rutan's Voyager world-record non-stop heavier-than-air circumnavigation. They specified the need to cross the equator twice in order to avoid such shenanigans, perhaps with a few more constraints.
They try to launch at least one balloon a year from Antarctica https://www.csbf.nasa.gov/mission.html
Reminds me of one of my favorite games from my youth: https://youtu.be/LgWo0xjQvBM?t=702 - with real atmospheric data, but with less control :)
That game was built with the same engine as maniac mansion, aptly called Script Creation Utility for Maniac Mansion (SCUMM)
This was really fun to mess about with. Interesting how some balloons take very similar paths for a bit, but then go on an entirely separate second stage. UK and Norway can follow a very similar path to get to the US, but then Norway's loops around Canada for a bit while the UK's head south. It's fascinating.

If you're bored and want to run mass simulations with your data I'd love to know which country is the hardest to spy on via balloon.

I wonder if there is a way to simulate a balloon in reverse, i.e. given its destination figure out which sources it could come from. Maybe it would just be a matter of reversing the winds and time? Would be interesting if it were doable.
You’d need to know it’s previous altitudes.
Thanks! I now know that I need to release my 20m Winnie the Pooh balloon from the Aleutians islands in order for it to sail over China 22 days later.
Big thanks for creating this. A great simulation to show anyone that their knowledge of how the atmosphere work actually have some applications.
NOAA† also has their own Balloon Flight simulator. The interface isn’t as slick but it really dives into the details: https://www.ready.noaa.gov/hypub-bin/trajsrc.pl?trjtype=4

It has a feature where you specify a target point and the origin point and it plots the changes in altitude your ballon will need to make to reach the target (if possible). Was surprising to me that if launched far enough away, a balloon can end up nearly anywhere in the US it wants just by controlling altitude.

update: They also make it easy to share results! Here's one of my simulations: https://www.ready.noaa.gov/hypub-bin/trajresults.pl?jobidno=...

They also have a Windows desktop package that does the same thing: https://gml.noaa.gov/ozwv/wvap/sw.html

---

†NOAA = National Oceanic and Atmospheric Administration, a US government federal agency

Without any context on the page, my first impression is this is presented as a tool for people to validate certain theories such as "could a balloon released from china at some recent time and date have actually blown into the US?"

As other comments have pointed out, the lack of ability to test with different altitudes makes the tool unfit for that purpose.

That leaves an open question of what the creator is hoping people will see in this. Is it a game or toy? Is it a technical experiment? Is it art? Does the maker not care what it is to us? (but, still, I'm curious what it is to the maker)

You don't need a simulation. This has been done successfully from Japan with bombs hitched to it.

People were killed.

https://www.damninteresting.com/curio/ww2-japans-balloon-bom...

The key is to use high altitude balloons to catch the Jetstream and a system of timed weights for release.

From the article:

> The news media cooperated with the military

Same as it ever was.

Certainly one can think of obvious improvements - which I generally think is a sign that a tool has a lot of potential. I would put it somewhere between art and educational tool: I hadn't really thought that much about how far a balloon could drift in a given time period and this did kind of make me realize that they move pretty "fast" - around the world in 80 days? How about 80 hours.

Sometimes you just have time to make a think that has the potential to provoke further discussion and interest - which seems like what happened here.

Also, they said the balloon had "limited steering capability" which I assume means it could nudge itself in a direction or at least adjust it's altitude.
Look at https://www.aviationweather.gov/windtemp with the levels on the side for different altitude.

It is possible for the balloon to change its altitude and then pick up different wind directions and velocities.

https://x.company/projects/loon/

> To identify helpful wind patterns, Loon used advanced predictive models to create interactive maps of the skies. These maps allowed the team to determine the wind speed and direction at specific altitudes, times, and locations. The team then developed smart algorithms to help determine the most effective flight paths through the varying wind layers. With the aid of these algorithms, the balloons could accurately sail the winds over thousands of kilometers to reach a desired location and remain clustered around those destinations in order to deliver consistent connectivity below.

And https://x.company/blog/posts/drifting-efficiently-through-th... gets into it more.

> It is possible for the balloon to change its altitude and then pick up different wind directions and velocities.

I've tried multiple different levels from 090 to 520, and it seems the deviation within level is extremely minimal.

i.e. maybe a difference between 70deg and 75deg, but if it's going in the general direction of east, there's not an altitude to turn around or even sidetrack.

That's today with today's weather patterns, and you're only up to 52,000 feet. The spy balloon was working in the altitudes of 60,000 to 120,000.

Look at +6 hours at GPI (Glacier Park in Montana). At 34,000 its easterly at 24,000 its westerly.

Yeah, there's no way to adjust altitude with this simulation. It apparently assumes that it reaches some (unspecified ) altitude and just stays there?
Isn't it more like a zeppelin / blimp if it has any steering capability at all? I always thought that was the main difference to a balloon.
Airships like a Zeppelin or blimp have some propulsive component to them that let them navigate without the wind.

A balloon can only navigate by catching different winds at different altitudes. When you can control the altitude from 60,000 feet to 120,000 feet there is a significant amount of variation in the direction of winds.

https://www.govtech.com/question-of-the-day/what-is-now-stee...

https://blog.x.company/drifting-efficiently-through-the-stra...

It's fun and stimulates the imagination
Never ask a geek, "why?", just nod your head and back away slowly.
"Your scientists were so preoccupied with whether they could, they didn't stop to consider if they _should_ ~ Dr. Ian Malcom, 1993.
Can’t give an answer but my thought here is that this thing probably would like to be swept up before the news cycle turns and balloons aren’t a thing anymore. Looks and feels like something put together in a bit of a hurry (which is perfectly fine)
It is what you believe it to be (most of the time politically). People use it prove the theory. Other people use it to disprove. I can only see the fun with technicality on using the real data to simulate this.
definately modern art.

The question is the purpose

This is hilarious. I love it.

Release from San Jose, CA arrive in London, do loop around France including a stop at the Eiffel.

Makes me think we should almost ban motorized flight for climate reasons and resort to long distance steampunk trips with the trade winds.

Q: How would we differentiate between a spy balloon and a weather balloon?

A: Well, it's not really possible - the NRO (the highly secretive US satellite agency) makes a lot of geospatial data available to scientists for use in studying everything from forest fires to fault geology to ocean ecosystem productivity and beyond.

Here's a fun challenge: try to draw something nice on the map. My first submission is a not too perfect heart:

Deploy time: 2022-11-10 03:00

Deploy point: 23.795° S , 133.047° E

Simulation length: 10 days

And a fish:

Deploy time: 2022-11-10 06:00

Deploy point: 5.31° S , 18.138° W

Simulation length: 9 days

Ha ha, everything I launched from the U.S. passed over China.

What is that faint whiff I detected? Is it web-site-as-performance-art?

Your balloons flew west..?
No, they went East ... and kept on going until they were over China.
Thanks, using this to launch my next balloon. If its not accurate I'll be sending you a bill for my services.
This has to be one of the coolest things I have seen this year on HN; last year it was the "endless acid banger".

It's tempting to let feature-creep make this bigger, but other than a few UX tweaks, it's perfect the way it is.

Then again, is there is anything I could think of for version 2.0 it would be satellite imagery from the virtual balloon (depending if Google maps allows that in their API)

update: I have now added a disclaimer that this site is aimed at for fun usage only, in case we would scream at each other over the topic of fidelity and application

thank y'all for your support

This is amazing; really puts in perspective how hard it would be to say you know someone sent a balloon exactly to you, or that they didn't mean to... its something you could plan for (like this shows) and something that probably is very fickle
Spawning at Singapore goes to the east, most other places go to the west. Interesting.
Weird, the one I launched in SG flew over to Africa and then flew back, didn't really traverse much distance compared to other spawn points.
Only fun thing I found to do with this is race two cities' balloons around the world and back to original longitude. DC comes from behind and beats NY (at least at precise locations I selected).
Deploy time: 2022-11-13 06:00

Deploy point: 22.106° N , 110.918° E

Makes it all the way back.

Another one:

Deploy time: 2022-11-17 17:00

Deploy point: 47.548° N , 121.537° W

Floats all the way around the world before passing within a few hundred feet of the start point. I was quite surprised when the second launch I tried turned up an (almost) eigenfloater of the matrix.

Deploy time: 2022-11-11 20:00 Deploy point: 32.695° N , 167.695° E

Comes back to exactly where it started, at first I thought the sim was broken.

Feature Request:

Missile command mini game where one county launches a barrage of them at your country and you have to shoot them down over ocean.

your feedback is heard!

I will see what I can do. Thank you!

This site can be used to demonstrate how small perturbations in initial conditions led to big differences over the long run, aka: butterfly effect. Try to place to balloons near each other and you'll see how their trajectory end up differing by a lot.
Great success, not one has been shot down so far. Would like a medal screen.
Now we are getting towards a real game: Deploy a ballon and do not let it run over the US. And I also want to see the jet launching
It's interesting because climate modeling/forecasting with a very powerful computer could tell you where and when to drop your balloon to pass certain targets.
Uruguay has to be the least interesting country to be a Spy Balloon operator.

Anything you launch come back to you only traveling across the sea and some farms in New Zeland xD

Atmospheric data won't help if you don't specify altitude. Different altitudes = different wind speeds and sometimes even directions.
I use 100hPa level but I can definitely add more levels for a better simulation
You should be aware that it is possible to direct your balloon pretty much anywhere by just changing the altitude.

If I was China and I wanted to make super advanced balloon to spy on US I would definitely make it so that it can use air currents at different altitudes to direct it where I want.

For actual HAB missions, check out sondehub.org
Yes indeed. This is a fun toy, but there is software out there that can be used to plan real balloon missions, such as:

https://amateur.sondehub.org/

of course. that and the NOAA planner provide real professional perspectives as opposed to mine which is for fun only
Darn, I was hoping this would actually be one of those "XYZ Simulator" games on, say, the Steam games store.
anyone else disappointed when you spawn one over the US that little jets don't come and shoot it down?
interesting - what is the confidence interval on this? if you start a balloon in the same spot at the same time does it always end up in the same place? or is there a wide range of where it could be
My (uninformed!) suspicion is that the confidence interval has to be very wide.

I’m worried about things like: the size and weight of the balloon, the altitude of the balloon, and very difficult to predict future changes in wind speed and direction during the course of the balloon’s flight.

Both funny and interesting at the same time. Very nice work!
Starting in Kenya makes for interesting shapes.
Any plan to add future predicted weather patterns?
amazed this isn't causing my macbook to take off. good job!
Pretty cool. Good job!
Rewrite it in Rust
Why does the map (appear to) use exclusively endonyms?

Is this a new style requirement for socially correct behavior?

it was from openstreetmaps and they do so by default