Huawei launches first laptops using home-grown HarmonyOS
reuters.com
reuters.com
The game is afoot, and China knew to de-risk and decouple. I don't think that it can be stopped at this point.
HarmonyOS, RISC-V, DeepSeek, domestic EUV, etc. China is standing up its own tech pillars.
So I suppose American lawmakers see this as a game of slowing down the competition rather than fully impeding it. China will eventually route around every road block, so the question is whether or not any of this will help America keep an edge, or if that edge will even matter.
In the meantime, we're holding up our own tech giants up to antirust scrutiny (and rightly so). But does that also hinder America's lead on China? And, if so, what will that mean for the tech/AI race?
Europe is also hell-bent on slowing down American tech. Again, rightly so - data sovereignty is important, and anti-competitive, monopolistic behaviors have long stifled domestic industry and talent. American giants shouldn't be allowed to behave that way as guests in other peoples' homes.
It was known and was accounted for.
The idea is to make them spend resources developing their own technology on our terms instead of their own.
They were always going to do this, they just had to do it faster than they otherwise wanted to, which has an opportunity cost.
It will pay itself and offset those costs once they reach breakeven and start selling their equal or better tech in the international market, displacing the incumbents.
Of course this doesn't automatically mean China wouldn't eventually pull ahead without the external pressure either. I'm just not as convinced it was so clearly a forced opportunity cost loss as much as something which provided a washed mix of both friction and acceleration despite assuredly preventing the US from making more money while its tech was farther ahead.
What was the opportunity cost in this equation? A substantially smaller bailout for their commercial real estate market?
> [The idea is to make them spend resources developing their own technology] on our terms (emphasis added)
What terms did we dictate? Timelines? Trade?
How does America or the West emerge ahead here?
> China will eventually route around every road block, so the question is whether or not any of this will help America keep an edge
I’d say the lead is so slim it’s basically already gone. At least in the practical sense. If you were to isolate both right now. Cut them both off from the outside. One would be able to produce a modern cellphone the other would not.
Any sort of residual technical lead in the pure IP/knowledge sense is good for 3 years max I reckon.
Still, I would never buy a Chinese tech device, you are buying a surveillance system to allow its government to spy on you.
There is no public evidence that Chinese consumer tech has ever been used to spy for the Chinese government. None. Meanwhile, the USA has been caught running mass surveillance programmes like PRISM and tapping the phones of its own allies. That is confirmed. And yet it is the USA making the most noise, spreading fear about Chinese tech. People only seem to worry when the device doesn’t have a US brand on it. You can be a patriot, but don’t be naïve. Believing unproven claims while ignoring confirmed facts is not critical thinking.
In contrast, the West has been caught with INTENTIONAL backdoors - many of which have been directly linked to government and intelligence services. Juniper firewalls had a secret access mechanism tied to compromised cryptography. Trustwave issuing subordinate certificates to facilitate MitM snooping on all TLS traffic. Netgear and Cisco devices including undocumented public-facing remote access features. These were not speculative or theoretical. They were discovered, documented, and in some cases quietly patched without disclosure.
None of these were revealed out of transparency. They were found by researchers or whistleblowers. If there is no public evidence against Chinese devices, there is no case. Assuming intent without proof is not analysis. It is projection.
And let’s not ignore the obvious: many of these devices are manufactured in the same Chinese factories. But once a US brand name is stamped on the box, the fear seems to vanish. Somehow, they stop being a threat...
> Still, I would never buy a Chinese tech device, you are buying a surveillance system to allow its government to spy on you.
And yet the audit of the Huawei routers fond no backdoors.
ftfy
HarmonyOS is a modern (post-Liedtke) microkernel, multi-server OS.
Meanwhile, the rest of the world is stuck with the likes of Linux (monolithic), Windows NT (ugly hybrid) and MacOS (pre-liedtke Mach, hybrid, ugly).
Good technology exists (e.g. seL4, genode, RISC-V) but we seem to be stuck investing into bad tech.
The game changer part is of course in terms of the broader tech war. What we have here might be a consumer operating system that is technologically better than what is on offer from Apple, Google, and Microsoft. Built by a vilified Chinese company.
There are reasons nobody uses true microkernels. IPCs are slow and the gains are limited compared to the strategies all broadly used kernels already use. They are no monolithic kernel anymore. Everyone has slowly but surely been shifting more and more things to user space in isolated processes including Linux and Windows.
Hongmeng might be an interesting kernel. It might also not be. Sadly its proprietary and there are very little benchmarks not published by Huawei. Personally I won’t hold my breath for this one.
The problem you are describing is a characteristic of 1st generation microkernels, and was solved by Jochen Liedtke in the mid 90s, introducing 2nd generation microkernels.
seL4 is a 3rd generation microkernel.
>I don’t know what "modern" microkernel means.
To get up to date, a good resource is Gernot Heiser's blog[0], read from oldest to newest.
Even SeL4 fast IPC which is not actually a full IPC but works well in the barebone context of SeL4 remains in fact slower than good old syscalls.
The fundamental question remains the same “Is this worse the costs (in terms of both efficiency and design complexity)?”
To me, the answer is muddy here. Sometimes yes, sometimes probably not. I think it’s why hybrid approaches are now generalised but no one is really shipping a microkernel outside of industrial applications.
Unless it does, it is unlikely there'll ever be a year of the hurd desktop.
Who in 1990 would have thought a Chinese telecom company would productionize it before Hurd even released 1.0
It makes me wonder why the Pentagon, with a US$1 trillion budget and being a critical piece of the US state apparatus, could not create a solution like that in recent years.
Much has already been said about why Huawei is not simply a state apparatus, so I won't repeat that. The point I rather want to make is this: having a factually wrong image of the counterparty is dangerous, especially if you view the counterparty as an enemy (justified or not).
If you care about advancing your material interests, then you might want to emulate what you believe makes the counterparty successful (in this case, the belief that they're a state apparatus). But when you find out that the emulation yields bad results because your image of the counterparty was wrong in the first place, you will have wasted a bunch of time and resources. It's in your interest to get your world view right the first time aroubd.
Calling Huawei a “counterparty” suggests it has real independence. But in China’s system, especially with big tech, that’s just not how it works. The CCP doesn’t need to own a company to control it. There are legal, political, and financial levers that ensure Huawei stays on track. That’s not comparable to how companies operate in the U.S. or EU, where they can push back on the state without fear of retaliation.
I get why my comment came off as simplistic, but it’s not baseless. If we want to understand what we’re dealing with, we have to be honest about the structure Huawei operates in. Misreading that is a bigger risk than calling it what it is.
Not laptops
*-unknown-linux-ohos:
> Tier: 2 (with Host Tools): aarch64-unknown-linux-ohos, armv7-unknown-linux-ohos, x86_64-unknown-linux-ohos
> Tier: 3: loongarch64-unknown-linux-ohos
OpenHarmony has no support for gcc. All the toolchains are LLVM. [2]
[1]: https://doc.rust-lang.org/rustc/platform-support/openharmony...