Loon’s final flight

People often make fun of Google/Alphabet for their graveyard of failed initiatives/products. While some of it might be warranted, projects like this—that strive to solve a large, global issue through somewhat charitable actions—are generous and have only increased my respect for the Google/Alphabet/X team. I hope they don't get flamed for this, and hopefully some of their research and donation funds can live on and solve the issue that Loon originally intended to one day.

IMHO, this is NO different from google closing down any other product. If, however, this news came with a grant of all the patents into the public domain, then it would be another story.<p>Could a startup set up a commercial balloon based mesh network without running into patents now held by google? Probably not.<p>THIS is the problem with these &quot;skunkworks&quot; type projects from megacorps. They suck all the air out of the room and prevent others from innovating.<p>Loon didn&#x27;t fail, Google&#x2F;Alphabet just decided they were going to discontinue funding. Now they hold a pool of patents they don&#x27;t intend to do anything with. Now they will point that funding elsewhere and do the same.

Hey, 9 years is a pretty good run for any product, especially a ambitious one with no clear revenue model. And the shuttering is paired with a $10M donation to continue the support of the mission.<p>I know Google gets hate for shutting things down, but they have hundreds of products – I can&#x27;t think of any other tech company with such a wide range of products. They make bets, and with bets comes a certain amount of failure. I&#x27;d guess they have a better &quot;success&quot; rate than, say, YC.<p>(Ugh, I hate defending Google, but alas)

Loons are cute toys that got 100x the deserved media attention vs their actual impact. Actual remote internet connectivity is mostly being addressed through the grungy, mind-numbingly routine job of increasing the coverage of terrestrial networks. The folks doing that unrewarding work should get recognition for actually making a difference in people&#x27;s lives over the long run.

I’m going to say I called this one at least 8 months ago: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=23419270" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=23419270</a><p>Loon is not competitive in a world with Starlink (even without Starlink it seems).<p>The business didn’t make strategic sense.<p>I still really liked their original goal of loon for all and the first video they made for it: <a href="https:&#x2F;&#x2F;youtu.be&#x2F;m96tYpEk1Ao" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;m96tYpEk1Ao</a><p>It’s a little sad they had to downsize and give up. I guess Licklider was right to title the original internet paper an “intergalactic computer network”.<p>Musk started with interplanetary goals and he’ll at least end up with one that covers the earth.<p>Loon for all started too small.<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Intergalactic_Computer_Network" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Intergalactic_Computer_Network</a><p>—-<p>Related:<p><a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=23419880" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=23419880</a><p><a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=25601473" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=25601473</a>

I am currently reading &quot;The Idea Factory&quot; [1], the story of Bell Labs and its innovation streak for many decades. There were so many amazing things Bell Labs worked on - all directly related to solving business problems. At one point, they were given a task of developing the most perfect lubrication oil dispenser with a requirement that it dispenses exactly 15 drops of oil per squeeze of the trigger. They worked on Tractors that dug channels for laying telephone lines to materials that lead to the invention of the transistor to solve the problem of unreliability of vacuum tube based switch boards. Some worked on improving manufacturing and invented what we call Quality Control. Everything was deep and wide, but still tied to the Bell&#x27;s business.<p>When I look at Google X and bunch of modern corporate labs (Lab 126, Facebook probably has something, Intel Labs (drones!), Microsoft Labs), I see a whole lotta hoo haa about tech innovation, but nothing with a long term vision of integration, capitalization and sustenance. No sense of practicality and pragmatism. May be Loon is a way to make Google an internet company (ISP), I could be wrong but as an outsider, it feels like a PR stunt than anything else.<p>Highly recommend this book.<p>[1] <a href="https:&#x2F;&#x2F;www.amazon.com&#x2F;Idea-Factory-Great-American-Innovation&#x2F;dp&#x2F;0143122797" rel="nofollow">https:&#x2F;&#x2F;www.amazon.com&#x2F;Idea-Factory-Great-American-Innovatio...</a>

The successor, Project Taara, seems quite impressive (and more sustainable): <a href="https:&#x2F;&#x2F;x.company&#x2F;projects&#x2F;taara&#x2F;" rel="nofollow">https:&#x2F;&#x2F;x.company&#x2F;projects&#x2F;taara&#x2F;</a>

Given that Starlink will probably solve the same set of problems within a few years (that being access to internet worldwide in areas with minimal infrastructure) this isn&#x27;t too much of a surprise to me.

This is quite a surprise. Only recently, Loon collaborated with seasoned reinforcement learning researchers to build more efficient navigation controllers for the balloons, titled &quot;Autonomous navigation of stratospheric balloons using reinforcement learning&quot; [1]. Although, it would appear that much of the gains come from a simulator that they&#x27;ve been developing for a while.<p>[1]: <a href="https:&#x2F;&#x2F;www.nature.com&#x2F;articles&#x2F;s41586-020-2939-8" rel="nofollow">https:&#x2F;&#x2F;www.nature.com&#x2F;articles&#x2F;s41586-020-2939-8</a>

One thing that irks me a bit about Google X as a “moonshot factory” is that they seem to be claiming that they’re <i>good at moonshots</i> but I’m not sure that’s true. (I’m not saying they’re <i>bad</i> at it, just questioning that they have a special edge.)<p>I get the basic idea, that you try lots of big ambitious projects, and most of them will fail but some will hopefully stick; but how do you distinguish between a 5% success rate and a 1% rate, say?<p>Deep pockets and the willingness to place big long-term and medium-term bets is crucial, but it’s not everything you need. Does Google X have processes, people, notable successes they can point to to show that they really are good at moonshots?

Talking about X more generally, their Gimbal book is interesting:<p><a href="https:&#x2F;&#x2F;storage.googleapis.com&#x2F;x-prod.appspot.com&#x2F;files&#x2F;Gimbal_v2.0_X.pdf" rel="nofollow">https:&#x2F;&#x2F;storage.googleapis.com&#x2F;x-prod.appspot.com&#x2F;files&#x2F;Gimb...</a><p>I think their self-driving cars project is probably what X could be at its peak, incredibly challenging projects, high profit potential, well financed.<p>Projects like loon seem more like a project that a university would get grant funding to finance.

I really respect X and Astro Teller. We need this kind of optimism and radical exploration for social benefit.<p>Did you know that his grandfather was Edward Teller, inventor of the hydrogen bomb? In one of Edward&#x27;s final contributions [1], he mathematically modeled the cost and effect of pumping sulfur-based compounds into the upper atmosphere to reduce runaway feedback loops of global warming. On paper, it works. Solving global climate change is a real moonshot -- but too bad it is so taboo, because it could dramatically benefit all life on earth, if done right. Looks like balloons could be helpful for the implementation, too [2]<p>[1] <a href="https:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20160127185550&#x2F;https:&#x2F;&#x2F;e-reports-ext.llnl.gov&#x2F;pdf&#x2F;231636.pdf" rel="nofollow">https:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20160127185550&#x2F;https:&#x2F;&#x2F;e-reports...</a><p>[2] <a href="https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Solar_geoengineering" rel="nofollow">https:&#x2F;&#x2F;en.m.wikipedia.org&#x2F;wiki&#x2F;Solar_geoengineering</a>

&gt;...the road to commercial viability has proven much longer and riskier than hoped. So we’ve made the difficult decision to close down Loon.<p>I don&#x27;t really understand - wasn&#x27;t the whole premise of Google X and their &quot;Other Bets&quot; that &quot;Google has deep pockets. Let&#x27;s use this to explore radically big ideas, that are ahead of their time, and let the world catch up to them?&quot;<p>Bringing &quot;abundant, affordable Internet access, not just to the next billion, but to the last billion&quot; is clearly an investment. You bring them the internet. You spur them to achieve great things. Then you reap the financial rewards.<p>But that&#x27;s not a 10-year plan, that&#x27;s a 20-50-100 year plan. But, there&#x27;s no reason to think that Google couldn&#x27;t think like a 20-50-100 year company if they tried.<p>So what happened? Did they really think that they were going to make money from the poorest 1B people on the planet BEFORE they brought them ubiquitous internet and helped them use that internet to improve their quality of life?

This kinda tells you everything about the culture of Google:<p>&gt; Sadly, despite the team’s groundbreaking technical achievements over the last 9 years — ... — the road to commercial viability has proven much longer and riskier than hoped.<p>Focus is on the technical achievements, never mind if anyone ever used it.

This summer I had the luck of actually seeing a Loon balloon floating way up in the sky. Viewing it with a telescope was surreal, especially at night, you could see it blink!

I&#x27;m sure this type of tech can be rebooted at any time depending on the need. I think a sustainable commercial model would be difficult with the global slowdown. Many of these moonshots might sadly end earlier than anticipated. I don&#x27;t think COVID is the end of all momentum in startups but it sure feels like it may take a generation to return.

So what happens now to all the people who worked on that project? They shop themselves around the X building to see what other project(s) they could interview for and join? Or is this so big a change that this has a broader impact even to the job market around X?

Worth noting that Google invested close to $1B in SpaceX in 2015, and is estimated to own ~10% of the company. Starlink deployment going smoothly and being much closer to commercially viable may have something to do with the decision to finally kill Loon.

Loon always caught my eye, and I thought the reason it would catch on is due to cost. Isn’t launching a satellite much more expensive than a balloon?<p>For example, providing Loon antennas to a telecom provider, and decreasing the infrastructure cost of setting up networks.

I don&#x27;t think they will, but I really wish they would open source and release some technical data on exactly what components&#x2F;subsystems and design they were using for photovoltaic panels, charge control and monitoring systems, and batteries.

Google Nose. <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?app=desktop&amp;v=VFbYadm_mrw&amp;feature=youtube_gdata" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?app=desktop&amp;v=VFbYadm_mrw&amp;feat...</a>

Wondering how much SpaceX&#x27;s satellite internet has to do with their decision to shut down Loon.

The mesh network was informed by, and in turn informed LEO stuff. They did smart work

Starlink made this obsolete before it launched

out of curiosity, how many employees were working on this outdated idea?

When did it started?

Keep on the search and play ads on youtube

I thought this was parody

It&#x27;s surreal to think that a decade ago even corps like Google could get away with using ableist language for their projects.

and this is why we have quarterly goals<p>this project was dead in the water a long time ago, and for years

I don&#x27;t understand why google isn&#x27;t recognized as just about the largest failure on the planet. Other than search and gmail, everything they start, dies.

Google is cursed by its own success. Any product they release will always be on shaky ground when compared to their advertising and search money-printing monopolies.<p>At this point, they should give up and settle into their role as the 21st century IBM.

Cue <a href="https:&#x2F;&#x2F;killedbygoogle.com" rel="nofollow">https:&#x2F;&#x2F;killedbygoogle.com</a> comments...

Well if you want to attempt something big without having to work with company&#x27;s like Google then Elon Musk made an interesting offer today.<p>Musk is offering $100 million to the team providing the best carbon capture technology.<p><a href="https:&#x2F;&#x2F;www.reuters.com&#x2F;article&#x2F;us-usa-musk-carbon-capture&#x2F;elon-musk-to-offer-100-million-prize-for-best-carbon-capture-tech-idUSKBN29R024" rel="nofollow">https:&#x2F;&#x2F;www.reuters.com&#x2F;article&#x2F;us-usa-musk-carbon-capture&#x2F;e...</a>

Sorry to be this guy, but I am quite happy this will not happen.<p>The sky - our sky - is <i>not</i> a playground for tech companies.<p>Imposing a mesh-grid of thousands of balloons in the sky is an incredible intrusion for people around the world who have not asked for anything.<p>It&#x27;s the same for mega constellations of satellites (Starlink).<p>Would you be willing to have a company come and put unsolicited electronic equipment in your backyard? Probably not. It&#x27;s the same for our sky - I don&#x27;t want that Facebook or Google impose thousands of electronic devices in our sky. Or maybe ask &quot;the owner&quot; before you start, i.e. the citizens of the world.<p>What&#x27;s more beautiful than watching stars in the sky in a July&#x2F;August warm night?<p>Do you really want to lose this landscape forever and have 100 times more satellites than we already have? and have series of white dots one behind the other?<p>OK, companies and states have done it the past, but it&#x27;s not a valid argument to justify the acceleration with a factor 10 or 100 of this process with thousands of Starlink satellites for example.