Hacker News Reader

This is what makes the Albuquerque ballon festival kind of unique (they call it the box). Balloons take off fly away, change altitude and fly back to land in almost the same place.

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

https://www.krqe.com/news/balloon-fiesta/local-pilot-explain...

Except the day I went, which is something.. Still quite the sight.

well, you almost got it. you forgot the part where the fly past for a distance so that when they start to lower altitude, it brings them back to the start. what you describe doesn't really make a box

What is "the fly past for a distance"?

IIRC it’s related to the nearby mountains. Fly up, pushed towards mountains, go lower, pushed back towards the river/valley were launched. I feel for other people who didn’t get to grow up seeing those balloons every year on their way to school or random times.

NM also has the “original burning man” https://en.m.wikipedia.org/wiki/Zozobra.

Just to put numbers on it for example…Say they travel 1 mile down range where they then rise in altitude. Now, they float back 2 miles before lowering altitude to travel back 1 mile to where they started.

They fly past for a 1 mile distance in this example

I think I saw in a museum once that that's how Japan sent over bombs during second world war. They were able to calculate that even back then. It's crazy impressive.

Edit: https://en.m.wikipedia.org/wiki/Fu-Go_balloon_bomb

The problem with Loon was that there weren't enough of them, and weather patterns would inevitably cause a clumping of them in certain areas of the world where they would be useless. Tethered balloons at 100k' / 30km would be more useful for comms.

There's no tether material nearly strong enough to withstand the forces from wind and the tether's own weight and drag at 30km altitude - plus 15km additional length for lateral drift.

There's a concept in engineering called "Reißlänge" (≈ breaking length), which is how long a cable of a certain material can be before it would rip due to its own weight if suspended by one end [0].

The values for some materials are actually much higher than I remembered, with Kevlar reaching ~250 km. From a purely hold-its-own-weight perspective it seems like it might actually be possible!

Obviously there's a ton of other problems that wouldn't make it feasible, the first I can think of is the weight of a 50 km tether wouldn't allow the balloon to even take off. I'm sure you can think of another million reasons it wouldn't work.

[0]: https://de.wikipedia.org/wiki/Rei%C3%9Fl%C3%A4nge

That's interesting and surprisingly long for Kevlar.

The highest altitude recorded by a single kite is less than 5km. I understand the biggest constraint for higher flights is tether strength-to-weight ratio. Nylon is typically used - fishing line.

Tether diameter is another factor as the load from wind, on the tether alone, is considerable over such lengths.

Not to mention the incredible hazzard it would present to aircraft

It's a materials manufacturing to applications and lifecycle problem, not a technical one.

Graphene does 50 GPa in bulk. A single atom sheet of it does 10 MPa on its own.

Take 75 metric tonnes of cable / space elevator ribbon.

That expressed as tensile pressure at an attachment point(s) is conservatively 7.5 GPa over an area about 10 cm^2. With sensible engineering, the entire craft (a very large one) would possess a web of attachment points across its perimeter to distribute forces more-or-less evenly.

Dynema might work, the more tension it receives the stronger it gets

Project loon was super cool.

It's a shame that Richard DeVaul got caught up in a political moment and a pearl clutching media that was eager to misrepresent burning man in the most salacious way possible.

Richard co-founded loon and is one of the most brilliant people I've ever met.

Awful what happened to him. Google's collective IQ dropped precipitously when he left.

1. The guy sexually propositioned a woman in an interview.

2. Seriously, do we need more? He's not stupid, he took the same management harassment training as everyone else in the company, he should have known better, he did know better, and he fessed up to it.

When you're a director, your reporting chain is not your personal Tinder. Please inform yourself on the subject of illegal sexual harassment in the United States, you'll save yourself and your reports and HR a lot of trouble.

The offending interaction occurred after the interview at burning man. He did not sexually proposition anyone in the interview. Furthermore the allegations were made two years after the supposed incidents occurred, the more distant an event is in our memory the more I'm inclined to give the benefit of the doubt to the possibility of people's accounts drifting from the truth.

The issue is whether Simpson was denied before or after burning man. If it was before, which is apparently what Devaul was insisting, it really was inappropriate to take any action in my opinion. If you match on tinder with someone you once interviewed, after the hire/no hurry decision is made, then your company shouldn't be treating this relationship any different from a total stranger. Or are we supposed to keep spreadsheets of every person we ever interviewed? Because that's not in my sexual harassment training, I dunno about yours.

Think about how you would defend yourself if someone accused you of making a sexual advance in an interview after you rejected them for a job? If testimony alone is enough to get you fired, then that's a significant source of pressure to give female candidates positive outcomes even if their performance isn't up to par.

This aligns with how the situation was described to me, with the exception that he was actually trying to help her get a job at Google, but she was turned down by the hiring manager (not Richard) and was unhappy about it.

It was terrible timing given the other stuff that was going on with Google at the time, and he sort of got swept up in all that.

I do agree with Richard's statement that he had a lapse in judgement.

I think it's critically important to keep personal and professional life strictly separate, especially as an executive.

Telling someone you're poly and that they should come to burning man with you during an interview is the offending interaction.

It's a fucking interview, not a date. Unless you are hiring an escort, it is not the place to discuss your sex life.

You've been given more information multiple times on this thread.

There's another side to the story that's very different than your belligerent assertions.

Are you interested in learning, or just cursing and throwing unsupported accusations?

And no, he didn't "fess up to it" as you've said elsewhere. He admitted to an error in judgement, without going into details.

I don't know specifically what he meant by that, but it's possible that he meant his error in judgement was trying to help the wrong person, in the wrong context.

Do you have evidence for this?

Because I was told a very different story. There's a lot more to it than what you may have read.

Perhaps you should consider reserving judgement until you have all the facts.

Consider for a moment that there may be important details and nuance that's missed by newspapers that are most interested in clicks.

Please inform yourself that accusations are not evidence, and maybe consider what you would feel if you were being judged by accusations alone.

I'm not at liberty to discuss the details, but I highly recommend you slow your rush to judgement. There's a lot about the whole situation that you are not aware of.

> 2. Seriously, do we need more?

Yes! Evidence.

He fessed up to it.

Richard??? Is that you??!!

No, I've met Richard and was impressed by him. I don't know him well, and haven't talked to him in years.

I did briefly discuss the whole situation with him after it happened, and came to understand that there was a lot more to the story than what was reported.

I do think he made an error in judgement.

I also believe he's utterly brilliant.

Both of those things can be true.

Even if the allegations were baseless, it would be hard to mourn someone at Google being set upon by the monster they helped to feed.

The balloon controls its own movement using only itself (and GPS) as a data source, right? Like, it's not receiving radio instructions from the ground?

Reminds me of this absolute stunning film about the japanese war balloons from WWII: https://archive.org/details/gov.archives.arc.13084

This allows "one button steering" as in the game rust

This is such an uncanny representation of an episode of Space Force it's hilarious.

Do unallied countries respect patents for military hardware?

Probably not, but also my personal experience says that you can't really replicate things just always from the patent. There's always some secret sauce to get things to work. The question is what exactly did Loon sell when they shut down and where did most of the people working on it end up? I know some folks at Loon and they went back to either Google Brain or some other moonshots in Google or Google Research but a lot of people didn't do that.

Lawyers who worked in chemical and medical patents confirmed this for me when I was in Google Legal. Don't even get me started about software patents.

The legal fiction is that a "reasonable amount of experimentation" may be required, beyond what's in the patent. The notion is that a lab tech determines the exact concentrations, temps, and pressure to get the thing to work.

Of course this is abused.

If you need secret sauce to get things to work, you can have the patent invalidated.

Tell that to the pharmaceutical industry. Many pharmaceutical patents can't be reproduced without insider knowledge as discovered companies that entered the "generic" business.

https://dlj.law.duke.edu/article/patent-laws-reproducibility...

You're very wrong about that.

What you're talking about is called a 112 rejection. Section 112 has to do with "written description," "enablement," and "best mode."

Getting a patent invalidated after it's issued can't be done on 112 grounds. It's done based on anticipation and obviousness (prior art, basically).

I was referring to how things work in my non-US locale, unclear patent specifications are grounds for having a patent revoked.

Which country is that?

Singapore

OK. I don't know how common it is to have a Singaporean patent.

Why would anyone respect patents for military hardware? It's for war, where's the respect in that?

That was my point, he seemed to be suggesting something almost like espionage if the patents had been sold to someone in contact with China or something, but the patents are already public and I don't think would be respected for this application.

He could have argued they maybe should have originally been covered by https://www.wikipedia.org/wiki/Invention_Secrecy_Act to begin with maybe.

Many Chinese equipment does look like a copy of Western equipment. I don't think China respect patents at all. And I think even allied countries wouldn't respect patents with military applications if in their interest.

e.g. "How Turkey Defied the U.S. and Became a Killer Drone Power" https://theintercept.com/2019/05/14/turkey-second-drone-age/

No-one respects patents for military purposes.

That's why important stuff is kept secret.

>some of their tech and patents were sold to SoftBank

Loon was shut down because Google stole the tech.

"In February of 2008, Google founders Sergey Brin and Larry Page visited Space Data’s headquarters with as many as 10 other Google executives in tow" https://www.wired.com/story/the-lawsuit-that-could-pop-alpha...

https://9to5google.com/2017/07/10/project-loon-patents/

https://www.law.com/therecorder/2019/07/29/google-settles-ip...

I know people who worked at Loon. It certainly wasn't shut down because of the lawsuit. It was pretty much because of X's move fast approach and they didn't think Loon would make money anytime soon. They killed another one of my favorite projects for same reasons: https://x.company/projects/makani/

From the point of view of somebody who works in residential last mile internet service loon was always a pipe dream, because you'd need a truly absurd number of them to provide persistent packet-loss-free 24x7x365 coverage over a USA county sized area. Paying customers won't tolerate situations like "oh sorry the balloon has drifted away with the upper atmosphere winds, wait a few hours for maybe another one to be back in our region".

With various ground based data relays for aggregating traffic coming from the PtMP balloon-to-cpe radio links.

With advances in 600/700 MHz (and 2500 MHz band) based 3GPP rev-whatever tech for LTE-based last mile that can punch through trees, and starlink (and hopefully someday some viable non-spacex competitors for starlink), and other terrestrial WISP stuff, the need for something like loon is greatly decreased.

Yeah it didn't make a lot of sense in most of continental United States and maybe that's why most of Loon's initial deployment testing was in Africa with sparser terrestrial network. Starlink definitely changed the rules and made LEO based systems a viable alternative for sure.

I had funding lined up for an experiment with Loon to deliver bandwidth to hospital ships, and now work with another guy who was at X during that time. Their perspectives are similar: the logistics to make this work with any reliability is truly heroic (I'm not asking for 5 9s, one 7 would be a good start). The experiments they did were incredible. But as a reliable asset ... good luck.