The Google Nest Hub is first commercial Fuchsia device

Disclaimer: I used to work and knew a lot Fuchsia developers very well.<p>No, Fuchsia is not a pet project. Instead, it&#x27;s solving a very realistic problem: building a clean driver interface, while keep it open and gain hardware vendor support.<p>Anyone have shipped a Linux-based embedded system (including Android smartphone) knows the pain: there are tons of ad hoc driver source files need to be manually merged&#x2F;rebased against the Linux Kernel. Want to upgrade the Linux kernel from 4.x to 5.x? Good luck redoing lots of merges and rebasing. There are tons and tons of git branches floating around, crazy examples like: &quot;Linux-4.11-some_soc_vendor-some_oem-random_device-random_project-rev-1.3&quot;. It soon becomes unmanaged and device manufactures just gave up on keeping up with OS upgrades.<p>Overall, the monolithic natural of Linux Kernel and its strict licensing terms made it hard to get code upstreamed back or to write properly modularized driver extensions. Even when people managed to get it work, you end up something like AMD&#x27;s GPU driver takes 10% of Linux kernel [1].<p>The term of OS is heavily overloaded. But Fuchsia&#x27;s goal is NOT to build other Android, but mainly to replace the Linux kernel.<p>Is it worth it? Can Google pull it off? I don&#x27;t know. But it probably worth a shot.<p>[1] <a href="https:&#x2F;&#x2F;www.phoronix.com&#x2F;scan.php?page=news_item&amp;px=Linux-5.9-AMDGPU-Stats" rel="nofollow">https:&#x2F;&#x2F;www.phoronix.com&#x2F;scan.php?page=news_item&amp;px=Linux-5....</a>

I&#x27;m suprised to find people surprised at the development of a new OS. Linux is around 30 years old; Unix and C are around <i>50 years old</i>.<p>Eventually the parameters of the world and of user needs shift far enough from the original design that it&#x27;s less expensive to develop something new than to adapt the old thing. Sometimes it&#x27;s impossible to adapt the old thing: for practical purposes, no amount of work will make C secure.<p>EDIT: Apparently Fuschia uses C and C++, and uses them exclusively in the kernel (if I understand correctly), and Rust and Dart are permitted for some uses, if this document is up-to-date: <a href="https:&#x2F;&#x2F;fuchsia.googlesource.com&#x2F;fuchsia&#x2F;+&#x2F;refs&#x2F;heads&#x2F;main&#x2F;docs&#x2F;contribute&#x2F;governance&#x2F;policy&#x2F;programming_languages.md" rel="nofollow">https:&#x2F;&#x2F;fuchsia.googlesource.com&#x2F;fuchsia&#x2F;+&#x2F;refs&#x2F;heads&#x2F;main&#x2F;d...</a><p>It&#x27;s an impressive run, but the ground has shifted, slowly, under our feet, and now Unix&#x2F;Linux are no longer a good fit. Let&#x27;s not be conservative; let&#x27;s do what our smarter predecessors did and embrace change.<p>Fuschia is free and open source, per other comments in this thread. If that&#x27;s true, let&#x27;s be very thankful that the successor to Unix, if that&#x27;s what Fuschia becomes, is FOSS. If Google put a proprietary OS on its phones, etc., even if they didn&#x27;t charge for it, they&#x27;d have a lot of market power and they&#x27;d be competing with a 50 year old OS. It could be a terrible blow to FOSS.

It&#x27;s worth noting that Nest is using (likely heavily customized) Linux, pretty hard to keep sync with the upstream due to its monolithic kernel architecture while doesn&#x27;t get aggressive investments like Android. But lifespan expectation for those products is much longer (~10 yrs) than typical smartphones (3~5 yrs) which leads to more maintenance headaches as Google accumulates their product portfolio, even compared to Android<p>And... Even notorious Google couldn&#x27;t completely pull their hands from Nest Secure although they decided to discontinue its production and sales. This is probably the problem they want to solve with Fuchsia. In this context, using Fuchsia is a pretty natural engineering decision. Fuchsia team can onboard its first customer while Nest team can outsource lots of its maintenance issues to a more appropriate team. I guess they already consider using Android before, but probably that&#x27;s a no-go option since its support usually ends before 5 years.<p>Of course, Android and Chrome OS have completely different needs and environments, so I don&#x27;t expect them to converge to Fuchsia in any foreseeable future. More likely scenario would be a shared ecosystem based on Play Store which enables more future strategic movements.

Capability Based Security is a proven solution to computer security problems that arose in the Viet Nam conflict. Those problems keep cropping up here almost daily.<p>The simple truth is that access control lists aren&#x27;t up to the task of protecting systems with persistent internet connections and mobile code. A Windows or Linux machine can not ever be made secure.<p>A microkernel, with the smallest possible attack surface, that never trusts driver code is the way forward. I don&#x27;t care if the eventual winner is Fuchsia or Genode, all I want is momentum forward out of the morass that is the current Linux&#x2F;Windows world.<p>I look forward to trying out Fuchsia and Genode in the next few months, and porting a Forth interpreter to each.

Many years ago, I interned at Google and a very wise mentor of mine explained Google to me in a way that has stuck with me.<p>&quot;Google discovered a hose that money poured out of. It&#x27;s called &#x27;online advertising&#x27;. All that we do now is find ways to make that hose go faster, and desperately search for another hose.&quot;<p>Why does Google develop 3+ different OS&#x27;s? In case one of them is a money hose. Why has Google created thousands of products, only to cancel most of them once people fell in love? Because they decided that they weren&#x27;t money hoses.

I think this is great! Obviously Google has been working on porting Fuchsia to the Nest Hub for a while and I think they made the right call by deploying it to this kind of device first. They will get more users to test it in real life scenarios. At the same time, it&#x27;s not a device that people use in mission critical situation, so if there&#x27;s issues, it won&#x27;t cause an uproar.<p>I could see the rest of their portfolio migrating to Fuchsia OS eventually, including both Chrome OS and Android.

I hope that performance is improved. The experience of using the Google Home Hub UI is quite bad. It looks slick and modern until you touch it, and then it feels like a car nav system display from 10 years ago. The frame rate is in the single digits. All the modern swoopy animations are wasted because the device can barely manage to display two or three frames before the animation is over. And the touch latency is off the charts.<p>It&#x27;s inexcusably bad. I mean, the hardware was purpose built for this software. They should have designed the software to meet the constraints of the hardware they knew they were shipping on. And it&#x27;s plugged into the wall with a big form factor so performance should be better than phones where power consumption and heat dissipation are much more constraining.

Possibly unpopular opinion:<p>This is a tragedy for Free Software. The Linux kernel (and specifically its GPL nature) is the best thing that has ever happened to Free Software drivers and embedded systems. The only reason that companies release open source drivers for their products is because they are legally required to in order integrate with the Linux kernel. Clearly Google wants to move away from this world.<p>I will mourn the day that a non-copyleft kernel supplants Linux and our only reason for free software hardware support comes to and end.

The comments on this article are disappointing.<p>The first major open source kernel since Linux, and somewhat arguably OS X, has been released.<p>I was looking forward to commentary on that, not blinkered commentary complaining this exists when Android&#x2F;ChromeOS do already (the error there is Android is a window manager + runtime on top of a kernel, ChromeOS is a window manager) or how performance reviews should be run.

This makes no sense to me. Why is google building it? They want a more stable experience by using a microkernel and pushing everything to the userland? Maybe this will help with updates and vendors letting their devices rot? Or is it something to do with licensing?

Linux was, AFAIK, the last GPL component of Android.

This is huge and exciting! One of the first open-source microkernels to get some serious commercial rollout, if I’m not missing anything. (I believe BlackBerry used a microkernel which was proprietary). Should be interesting!

Huh, genuinely thought this was a research kernel type situation. But they have a display manager and everything built in to it. Wonder why they&#x27;re using a different kernel instead of using Linux.

Aside: Fuschia* is such an odd choice of name unless you are on a dual mission to not only make an MIT licensed Linux, but also to improve the ESL** world’s spelling and vocabulary.<p>*Fuchsia, sorry.<p>**German speakers will be fine.

Worth noting that golang was evaluated and then discarded for fuschia: <a href="https:&#x2F;&#x2F;fuchsia.googlesource.com&#x2F;fuchsia&#x2F;+&#x2F;refs&#x2F;heads&#x2F;main&#x2F;docs&#x2F;contribute&#x2F;governance&#x2F;policy&#x2F;programming_languages.md#Go" rel="nofollow">https:&#x2F;&#x2F;fuchsia.googlesource.com&#x2F;fuchsia&#x2F;+&#x2F;refs&#x2F;heads&#x2F;main&#x2F;d...</a>

I understand that Fuchsia is currently targeted at IoT or consumer devices, but can it run &quot;containerized apps&quot; in server environments with its current design (compared to Linux namespaces)?<p>Maybe Kubernetes can be extended later to support Fuchsia nodes. It sounds really interesting to have a new target OS, with a different network and virtualization stacks.

ChromeOS, Android, WearOS, Fuchsia, Fitbit OS...<p>Doesn&#x27;t that make 5? (Potentially dumb question, I know...)

It&#x27;s a damn shame that you have to use flutter, that&#x27;s all I&#x27;d say.

With this update, will the Nest Hub finally be able to tell time without an internet connection?<p>I returned mine as I had this happen to me, which completely invalidates its use as a clock.

Damn, are microkernels finally back?

Well, if Fuschsia doesn&#x27;t do too well initially, at least they picked a hardware device that will probably be as forgetful as the Verizon Hub from 2009.<p><a href="https:&#x2F;&#x2F;youtu.be&#x2F;1z1nsuQi40w" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;1z1nsuQi40w</a>

Are there any privacy aspects to this? Does Google track everything that is going on?<p>Do you have control over your device or can Google release updates and do whatever they want with it?

Just another bot net from Google. How much personal information will these devices running this new OS capture&#x2F;steal? No thanks....

Probably the biggest Google vaporware ever

Who wants to start a betting pool for when they cancel the project now that it&#x27;s officially released into the world? I&#x27;m going with 2025. Just enough time for it to build up some momentum before they pull the rug out from under the user base!

It&#x27;s notable that this release has nearly ZERO attempt to market to or engage with 3rd party developers. It is solely for 1st party apps, solely under their control. Google picked a consumer device with extremely limited, contained functionality to target Fuchsia to.<p>This is an extremely crazy twist for a device that is supposed to play a star, core role in the home, and a strong indicator to me of where we are in the War Against General Purpose Computing.<p>There are definitely ways for 3rd parties to add to the experience, with the chatbot platform, but it&#x27;s all expressed in Google&#x27;s existing mold, via Actions with Google Assistant &amp; other pre-baked systems of manipulation.

This has been open knowledge for a few years now. I was fully expecting it to be cancelled before launch. In that sense am surprised it saw the light of day.<p>However, I have never seen an explanation of why they need an OS written from scratch. The whole kernel and driver ecosystem, apps, development model, every single thing written from scratch. It took 6 years too for it to be productized. Why would anyone make that kind of an investment? It has to be extremely compelling. I would like to hear it from them

I think it&#x27;s telling that they are &#x27;launching&#x27; this with a product that barely matters in their product portfolio. That indicates to me that they don&#x27;t see this as an Android killer just yet. A bold move would have been to put this on the next Pixel. Clearly, that&#x27;s not happening. If that was at all feasible, they would have waited for that and that would have been the grand launch. So, clearly it&#x27;s nowhere near ready for such a launch.<p>Their main issue with this is that OEMs like Samsung probably are not in any hurry whatsoever to jump on board and be even more dependent on Google. Without OEMs, app developers will drag their heels as well and Google is forced to maintain Android and not do a half-assed job of it. At this point Google has Android in cars, tvs, on phones, etc. Killing that would be suicidal; Google needs users interacting with their ads.<p>If they botch the Fuchsia launch somehow, they&#x27;d have more OEMs taking things in their own hands and cutting loose from Google and forking Android (like Amazon and Huawei already did). They clearly are not ready to pull the plug on Fuchsia just yet but at this point it&#x27;s a more than likely outcome. IMHO, they should just rip off the band-aid and move on. Most of what is wrong with Android is fixable.