Hobbling Huawei: Inside the U.S. war on China’s tech giant
reuters.com
reuters.com
The United States and its allies were derelict in not developing a 5G supplier, former Australian Prime Minister Malcolm Turnbull said in a speech in London in March. “With the benefit of hindsight it beggars belief that the countries which pioneered wireless technology – the United States, the UK, Germany, Japan and with wifi, Australia have got to the point where none of them are able to present to one of their own telcos a national, or a Five Eyes, champion in 5G,” he said.
I would have thought Qualcomm, Cisco, Intel and the like would have been all over 5G. Why weren't they?
Qualcomm and Intel focus and specialize on chips. Cisco focuses on layer 2, layer 3 and partially layer 7 equipment (think: routers & switches). Cisco has some products that compete against Huawei to a certain extent, alongside a wider product suite by Ericsson and Nokia, but all three vendors are more expensive than Huawei.
From the article:
"But the options are limited. Huawei is one of only three major global companies that analysts say can supply a broad range of advanced mobile network equipment at scale. The other two are Ericsson and Nokia. And Huawei has a reputation among telecom operators for supplying cost-effective equipment promptly."
As per the article, Huwaei is flat out cheaper. On top of this, they provide good-enough products (sometimes better, sometimes not) across performance, feature, stability and scale dimensions. The telecom vendor game is often one of references and precedent, especially when dealing with large telecom groups operating in multiple countries (think: Orange Group, Deutsche Telekom Group, Telefonica Group, Vodafone Group, etc.).
[1] https://www.3gpp.org/technologies/keywords-acronyms/100-the-...
[2] https://www.ericsson.com/en/blog/2018/5/core-network-evoluti...
The two types of network are not heterogenous because their layer 1 characters are completely different (electromagnetic spectrum vs. wires), which has led to vendor specialization over the years as few companies are/were able to compete across the full suite of different workload characteristics demanded by data-plane and control-plane needs.
Huawei's strength comes in their ability to provide an end-to-end turnkey solution for both types of networks at competitive prices with hardware & software that works.
https://blogs.cisco.com/news/huawei-and-ciscos-source-code-c...
And I wouldn't call it "documented", unless it provides references to the official reports. The link points to a Cisco site, and what would you expect Cisco to say?
here's a legacy archive of the relevant articles from a Chinese third-party site: http://tech.sina.com.cn/focus/cisco_huawei/
I'm a bit surprised it is still kept intact after some 20 years. You could see trails of history -- mind the "Google it" header at the top.
You can pull down tons of news reports and even the court filings if you are motivated, but why? Huawei themselves publicly admitted to stealing Cisco's code, the only disagreement was about the scope. I suppose we should just accept at face value their claims, when caught, that it was just a little bit, and only by accident.
Cisco is the only source for any part of the actual expert opinion. Huawei naturally declined Cisco's invitation to permit publishing the entire report; one can imagine how unfavorable it would be.
Those published excerpts are smoking guns - concrete demonstration of wholesale cutting&pasting of Cisco source code, down to whitespace and comments, outside the scope of what Huawei was willing to admit to.
By that time, though, Huawei had already achieved commercial success, and could afford to start cutting fewer corners & better comply with international law. So their plan worked pretty well.
Huawei is right up there with semiconductor mfg as a poster child for government-condoned (even government-assisted) industrial espionage as a tool to bootstrap strategic industries.
I'm not convinced that Huawei will backdoor their 5G gear, but I'm definitely on board with the plan to bar them from US commerce. Why play fair with opponents who cheat?
I do use a Huawei fiber modem+router 2-in-1 day to day. The router firmware is crap and I have to do some modifications, and effectively disable the router part. The UX is more aligned with the other cheapos in the Chinese market, I think. It’s unlikely Cisco would ship something like that.
I’m not sure if I’d be convinced by what Cisco says, though. The author was direcltly involved in the case two decades ago btw.
Huawei admitted the usage of Cisco code, and promised to remove that part. Case closed. However the article is referring to other parts (strcmp and the like). We have limited neutral information, and in this case, I think, should remain in doubt.
There are plenty of accusations that this was the case -- I'm not going to comment on the veracity of those accusations. However, if it were true, then that would certainly be a huge factor why Huawei was able to catch up and overtake other domain players within a relatively short period of time.
In defense of Huawei, I would like to point out that Bunnie Huang[1] noticed a certain type of generically Chinese innovation leading him to coin the term "Gongkai"[2][3], the effects of which led to him cataloguing and writing a book[4]. I think the combination of this laissez-faire "flexibility" around WTO-mandated IP rights and a grueling and debilitating 996[5] work ethic helped Huawei, among other Chinese companies, to catch up with Western firms pretty quickly.
[2] https://www.bunniestudios.com/blog/?page_id=3107
[3] https://www.bunniestudios.com/blog/?p=4297
This is essentially copied from the Japanese. Japanese mega-corps far as I understand very often unofficially pooled patents and technology. That became an issue as dominance was asserted in the consumer electronics market as outsiders weren't in the club. Entry into an established market meant going up against a thicket of patents owned by multiple entities.
We know for (nearly) certain that Huawei stole folding screen tech from Samsung (via a supplier), and now that Samsung has reneged on it's Galaxy Fold launch the Huawei folding phone will be first to market.
Now that they have a reputation for stealing it is hard to view their other advancements as purely the result of honest work.
https://www.lightreading.com/mobile/5g/bts-mcrae-huawei-is-t...
The reality is that Huawei has been outspending its competitors in R&D by a wide margin for over two decades, because the latter are managed by bean counters in thrall to the stock market's short-term horizon. Eventually, the difference would show.
One further wrinkle is that Chinese 5G uses frequencies like 3-4 GHz that are hogged by the military in the US. US carriers have to resort to millimeter wave frequencies in the 30+ GHz range that can't penetrate a door, let alone a building, and are thus mostly useless for 5G penetration outside a very few congested open-air locations, and thus their 5G technology is at a severe disadvantage over the Chinese one:
https://spectrum.ieee.org/tech-talk/telecom/standards/how-am...
Most countries not aligned with China or the West will prefer the Huawei technology: superior technology yet cheaper, and far less expensive to deploy and operate. Any country that doesn't puts itself at a competitive disadvantage over others that do. The US may yet convince its European allies to shun Huawei, but it will come at a severe cost.
I am not expert on networking equipment by any means. But I would assume that 5g equipment is built upon previous technology.
It would thus follow that stealing the preceding, underlying technology and then focusing resources on building out the 5g part of the tech would be the fastest/cheapest way to become a leader in 5g.
The idea that "the stock market" prevents companies from doing any long-term development is one of the most delusional, easy to disprove memes going. Public companies in America have been churning out game-changing technology forever. The Chinese are ahead on 5G, so suddenly the system is a failure?
Going public is a capital raising mechanism, nothing more. And it's probably driven a few orders of magnitude more innovation than it's hindered. The disdain for public markets (or finance in general) on these boards is bizarre.
The current situation was completely predictable, but industrial policy has a bad name in rentier capitalist circles that dominate our policy apparatus.
Plus an army of support engineers to help you turn-up and operate your new network.
https://www.lightreading.com/mobile/5g/bts-mcrae-huawei-is-t...
For Europeans, they have to balance the possibility of Chinese espionage against the certainty that the US and UK are already doing it:
* https://www.theguardian.com/media/2015/jul/02/wikileaks-us-s...
* https://theintercept.com/2014/12/13/belgacom-hack-gchq-insid...
In that light, Huawei equipment that the Chinese government tries to make as immune to the 5 Eyes as possible may actually be preferable.
Granted, neither Nokia nor Ericsson are US companies or subject to direct US pressure the way, say, Lucent would have been, or Cisco is, but the US has suborned equipment makers in neutral countries before, as with Crypto AG:
Unless you're contributing to or benefitting from that 5 eyes intel.
Our governments aren't as averse to these things as you and I are.
> agents of the Australian Signals Directorate, the nation’s top-secret eavesdropping agency
Straight from the geniuses that outlawed mathematics.
I'm sure not all of the "cyber security" people are complete idiots, but how many can resist an offer of unlimited funding, and all you have to do say things are terrible and very scary without ever being caught out, since it's all about national security, and your boss doesn't really care because more bad things mean infinite budget.
"Cyber" really has stopped being a useful word in as far as it's ever been one, and has been relegated to politics, scaring the population, and fleecing people and institutions in the name of "security".
1.) 5G is a mobile network that lets people watch really high- resolution cat videos really fast.
2.) Your country is not under "cyber attack" or "cyber threat".
History tells us most of the threats and developments in Cold War I were posturing and positioning. Of course Blackbird planes are really cool, but do we really have to go through the rest of all that bullshit again? Can't we all just, you know, grow the f* up?
https://www.reuters.com/article/us-netherlands-huawei-tech/d...
An option would be to confidentially share with UK or German intelligence agencies and verify, but they chose instead to publish a vague article. Seems doubtful this will result in anything concrete.
They don’t get a free pass as a “trustworthy news organisation” just because what they report isn’t wrong, if they also aren’t reporting anything useful.
https://www.reuters.com/article/us-kpn-huawei-5g/dutch-telec...
It was arguably beneficial to not worry about China's IP issues, foreign ownership, capital controls etc. in the 1990's because a growing China was good for everyone.
Now it's not, so there's a realignment.
There's cynicism here, but both issues are perfectly credible: trade/technology and security.
There are a ton of externalizations that enable China to provide, especially 'services' so cheaply. Those externalizations matter in trade, and probably should be capture in some form of tariff.
i.e. it's not actually 'cheaper' for Sweden to buy stuff from China, if all things otherwise equal the only difference is the manufacturer in China doesn't have to pay for carbon taxes or something like that. It's just moving deck chairs around.
The only thing that's unique with this case is that it's rare for the US government to be so openly hostile toward a major trade partner.
So the USG can tell a company "you cannot transfer this controlled material to these people, 'cause NatSec", but (probably) can't compel a company "build a tool for us that does X".
The fact that Google and affiliates are being used in a similar manner that we’re accusing Huawei probably makes the people in charge laugh even harder.
The real question should be that if Chinese government policy creates a security risk for the US, why are they only going after Huawei? Why not a blanket ban on all sophisticated Chinese electronics? Lenovo has been caught before putting spyware on computers, but somehow the US government still buys Lenovo laptops...
I feel like if China were to backdoor 5G equipment deployed to the US it'd be relatively trivial to detect by operators, and absolutely crippling to the market share of Huawei if they were found to compromise their equipment.
That's one of those empty, non-falsifiable statements that don't do anything to further a discussion. Spying and illegal activities are by definition supposed to be secret.
But it also doesn't matter. Because...
> Hypocrisy and arrogance at its finest
... as a matter of general principle: it's in the best interest of a country to know what other countries are doing and to protect its own assets. There is no contradiction in doing both at the same time.
If a country suspects that one of the cornerstones of its communication infrastructure is at risk of being infiltrated by a foreign power, it makes sense to take steps to avoid that.
I'll leave it to others to judge whether or not that principle applies to the Huawei case.
The US does not view China as an ally, it's a competitor that the US sometimes cooperates with economically.
The US and its allies are saying they prefer to keep the spying inside their own allied domain. That's not only entirely rational, it obviously makes perfect sense from a national self-interest standpoint.
As an American, if I get to choose, I'm buying my 5G tech from European or Asian allies. I'll happily pay more for it.
The US isn't saying China can't spy, it's universally understood that China is spying and will keep trying to, just as the US will. The US is basically openly saying it's going to attempt to counter or limit that spying, because China is a global rival in all regards. Just as China will always attempt to limit the ability of the US to spy on China and its interests. The same goes for Russia and its spying, going all the way back to the early days of the USSR.
There will soon be a heap of 5G phone hardware going cheap. Software support will be lacking. The hardware manufacturer may well be open to publicly releasing the information independent third party developers will require to write software for their otherwise useless phones. If you're reading this Huawei, why not dump all the required information on your web site right now?
Plus I do like the two-facedness - nobody should spy unless it's us.
This is all still very speculative, but there isn't a telecom out there I'd trust to run my servers.
I agree we could use nice ways to serve content more locally, and there's already some solutions here, but I'm struggling to see how it works as part of a cell network protocol without redefining a lot of how we expect the internet to work.
- Almost any existent device will likely be connected to the Internet through 5G almost as fast as it became available
- Access probably will be a LOT cheaper than 4G, will get there a lot faster (months not years), so everybody will be onboard ASAP
- 5G network could easily replace many fiber optic deployments currently used by telecommunication companies big clients (so if you breach the 5G infrastructure, you're reaching internal networks in banks, trust funds, government sensitive information offices, whatever fancy dangerous stuff not suitable for public access)
- 5G is REALLY FAST and RELIABLE, (very hight bandwith plus very, very low latency), so if you hack into something, or if you deploy malware, it will spread faster than the Thanos snap. The data exfiltration - taking your stolen data out to your infrastructure - could take a LOT less than now, this probably means the bad guys could hack out a lot more data than now, also probably using a "raw" approach: if you get in, just transfer everything you can access, right now
- The massive (freaking massive) traffic you're going to see into the network monitoring infrastructure will probably be very similar to what Five Eyes currently sees for entire countries, but in city per city base (NY, Boston, DC, Berlin, Barcelona, etc.), so initially very few players will be able to get out something meaningful from the monitoring stuff, but what devices failed: more bad guys oportunities.
Finally.
- Whoever makes the hardware has the upper hand, you - whoever you are - go behind them, in almost any security scenario.
So yeah, we need to have this stuff under some serious control.
Trump is almost certainly right in his attemp to make what he's doing.
In the past, faster almost always mean a denser network of costly towers. Is that different for 5G?
1) In order to stimulate (or meet, depending on which side of the supply-demand equation you represent) demand for 5G, wireless operators will most likely need to increase data caps (e.g. Canada) and keep the same prices, because 5G speeds are much faster than 4G. This means that "wireless access" denominated in "$/MB" will likely drop.
2) 5G equipment is not necessarily cheaper than 4G. Depending on the spectrum bands used, 5G may in fact be more expensive. A recent April 2019 Wapo article[1] kind of speaks to some of the issues with the US settling on 24-300Ghz vs. rest of the world settling on sub-6Ghz bands. But the answer from a wireless operator point of view is that 5G is probably much more expensive than 4G, not less.
[1] https://www.washingtonpost.com/opinions/global-opinions/on-5...
Also I want to say that wireless communication will never be technically superior to something shielded in some way, and if you can afford it, and you don't need to move around, it will always be a better choice. The fact some ISPs can compete with cable using wireless modems is more a statement about wire ownership laws than anything else.
It should be self evident that turning over domestic communications infrastructure to a global rival with the largest military on the planet, nuclear first strike capability, a history of violence, newly minted dictator, and global ambitions is a terrible idea.
Also, the US government isn't stealing US IP and giving it to Chinese businesses. The PLA is, and their ties to Huawei run deep.
When you say it like that it's not quite clear whether you are referring to China or the US.
I don't know if there's a name for this but I'd call it "top dog syndrome", this is possibly the first time the US witnesses another player rising up to status and they don't seem to take it well.
No, it's stealing "allies" IP and giving it to US businesses [0] while subsidizing Intel to the tune of billions [1].
Still, nobody writes headlines along the lines of "State-sponsored Intel processors come with ME backdoor for government spying." instead we get, super trustworthy, articles how the NSA totally helped with disabling the backdoor [2], because a sucker is born every minute.
The absurdity of it all is immense, yet still seems to elude most US Americans.
[0] https://en.wikipedia.org/wiki/ECHELON#Examples_of_industrial...
[1] https://subsidytracker.goodjobsfirst.org/parent/intel
[2] https://www.csoonline.com/article/3220476/researchers-say-no...
According to your source the US government is subsidizing Intel to the tune of $228,154,638, $141mm of which are loans. The lions share of those billions are incentives from Oregon and New Mexico to keep Intel manufacturing in their states.
>The absurdity of it all is immense, yet still seems to elude most US Americans.
What seems absurd in my opinion, is the idea it's a bad idea to prevent an unfriendly foreign power from gaining more access to your communication infrastructure. That's not hypocritical, it's pragmatic national security policy...
Is this British slang or a typo ?
Huawei maintains that it is not a hardware issue, while Google says that it is a hardware problem. A class action lawsuit was just settled because of this.
My suspicion: Huawei surreptitiously collected data, using up CPU threads, and core Android software did not detect this hardware usage.
I know many many owners of the Nexus 6p that had this problem, and it was only this Nexus which had it, the only one made by Huawei.
I exchanged it for a Pixel 1 because of extended warranty.