According to Google’s official support page listing the current firmware versions of its speakers and smart displays, version 14.20230831.4.72 is now available to those enrolled in the Preview Program (which can be accessed via the Google Home app). These updates are often released in stages, meaning it may be a few weeks before your Nest Hub gets the latest build.
On the project’s website, Google offers a more in-depth look at what has changed in Fuchsia version 14 (as well as version 13, which the Nest Hubs skipped). Most of the changes will only be relevant to Fuchsia developers, but there are a handful of user-facing improvements.
Google is providing some very detailed release notes for each version of Fuchsia, which are quite interesting to peruse. The recent layoffs at Google hit the Fuchsia team hard, likely reducing its future prospects as Google’s unified consumer-facing operating system, but that clearly doesn’t mean it’s entirely dead in the water.
Thom Holwerda,
Fuchsia probably can’t grow as a unified platform under an anemic conditions, they’d really need to remove all stops for it to have a real chance.
You know, I would have like to see some new alternatives. They might have fixed some of android’s mistakes, or maybe they wouldn’t, who knows. But we need more options than the mobile duopoly we’re currently suffering under.
Edit: I would have liked to correct my mistakes, we need an edit button
Oh how I miss that buttton
Alfman,
Fuchsia has really interesting design choices for consumer hardware. It is one of the few “not Linux” designs out there. With a Rust based kernel, very heavy emphasis on security, and high level (Haiku like) API, it might have been a very good platform.
However, I am not sure about the future prospects. Can it realistically survive without Google’s support?
Fuschia still has funding. They just took a cut like many other divisions.
AndrewZ,
Of course they do. They have to support existing products in market after all.
My concern is for the long term. It is a wholly Google product after all. Repository is at googlesource, the governance is by Google, and I am not sure how many external contributors are out there.
Anyway, again, I do like the project, and hope it has more success outside of Google as well.
Alfman,
Out of topic.
It seems like we cannot comment on older articles.
This is re: SFP+ and “direct attach” cables discussion from earlier.
Decided to go back to USG (yes I know, they don’t have the best track record). Not only they finally offered a 10GBe gateway, they also had a very good black friday deal (this one: https://store.ui.com/us/en/products/udm-pro)
Anyway, using Mellanox cables, it can directly attach to:
https://www.amazon.com/gp/product/B08DJWZ6VJ
And of course it can also do fiber.
sukru,
That’s cool. I’m not very familiar with their products. I looked at them for Wifi but ended up going with a different solution. Are the LAN/WAN functions hard coded?
…screwed up the blockquotes, darn it!
Alfman,
It is a bit configurable. Any two of ports 8, 9, 10, 11 (GB, GB, SFP+, SFP+) can be used for Internet (one primary, one backup). So, it might be possible to have both SFP+ ports for LAN. Though as mentioned, I use another switch for for that purpose. (Essentially 3 more SFP+ ports)
One downside with remapping is, there seems to be no hardware switch between the two ports, and all has to go through the CPU. People mention ~5 gbits/s with IDS and ~9 gbits/s without it. So it should be possible.
Not sure about the warranty, but these machines are relatively new in their offerings (previously they could not even do a gigabit w/ IDS, something like 300mbit or so). Hence, I can’t be sure about reliability. But at this price and convenience, I could not say no.
They offer configuration backups. The entire system is like VyOS and can be deployed into new hardware quickly. At least in theory.
https://community.ui.com/questions/Relation-between-EdgeOS-and-Vyatta-VyOS/16e71fa1-cc29-4e89-a6e2-268264be25c4
Agree 100%. Wish I had more time, though. (I had actually built a mini supermicro server and was planning to do opnsense on it. But gave up due to lack of time. Now I need to figure out what to do with that one).
sukru,
A lot of routers technically support “backups”, but I’m not so sure it’s technically that useful across devices. For example my netgear router outputs a proprietary backup file that will do me absolutely no good in provisioning a new router unless it happens to be the same model.
With plain old x86 operating systems, we can often load up the same image on completely new hardware. This has little to do with x86, but historically x86 standards make x86 hardware much more generic. For example I’ve had great luck using mainline linux across a wide variety of x86 hardware in my custom distro whereas with ARM hardware it takes a lot more hacking.
Yeah, I can’t help you with that, haha.
Fuschia looks to be much more portable than Linux-based Android. So it could port to new architectures, when Google wants to release those. Fuschia runs on RISCV today, but not much activity to see. The development tools are very modern with a lot of software that could get ported easily. I thought Fuschia would have better security out of the box but this does not appear to be the case: https://swarm.ptsecurity.com/a-kernel-hacker-meets-fuchsia-os/
Fuschia team does not publish a detailed road map of future features. So it’s hard to tell what is coming. Sometimes you can read the source and make some good guesses. I haven’t done this. What I don’t see in Fuschia is a killer API. Something like what BeOS/Haiku has.
Architectures currently supported: X64, AMD64, ARM64, RISCV64.
Ok, well, the user interface capabilities in Fuschia look pretty ambitious:
https://fuchsia.dev/fuchsia-src/concepts/ui/scenic
It would be cool to see some demos and/or proof of concepts for these features.
I was convinced that what made more sense was to begin replacing Wear OS with this. I suppose I was wrong.