Hacker Newsnew | past | comments | ask | show | jobs | submit | pwaivers's commentslogin

I am hesitant to install the plugin because Chrome warns that it can "Read and change all your data on the websites you visit." Is this a normal authorization to give to Chrome extensions?

Overall I think it's a really cool extension idea and I would love to use it.


This is a really cool website!


Not all of them. For example, the two masks pictures here do not: https://en.wikipedia.org/wiki/N95_mask.


Here's a pretty good explanation of how the figures are reported (and why it can be misleading): https://www.nytimes.com/2020/07/29/business/economy/us-gdp-r....


It's difficult to have both, no?


Please read the rest of the article. It discusses how China could force TSMC to stop producing chips. It even covers the US foundry, and why it won't make a difference.


> Please read the rest of the article. It discusses how China could force TSMC to stop producing chips. It even covers the US foundry, and why it won't make a difference.

But all of realistic options listed entail the fatal political defeat of Taiwan, mostly through military action. Existential political threats usually end up as military conflicts, one way or another.

The only option that doesn't involve military action is the disinformation campaign, but that's pretty far fetched.


Trade war looks most promising: Taiwan has 150B trade with China comparing to 100B trade with US.


How much of that 150B trade with China is for parts that are in turn sold in US? It’s very hard to winnow out real meaning from raw trade numbers.


What will happen to Taiwan, if China stops trading with Taiwan.

Like, a full on boycott.

Maybe Taiwan can sell their 150B in products to Americans?


Sure. Just like farms can switch over from supplying restaurants to supplying grocery stores.


What would happen to China? Foxconn alone employs hundreds of thousands of people


We share the same direction of thinking.

I think China's greatest threat comes from within. And the brinkmanship is at supply-side structural reform VS unemployment.

The risk has never been this high, given

1. belt and road initiative is already half-busted

2. drastic drop of foreign demand due to pandemic

3. cry for decoupling from all wealthy nations

And see this: https://www.youtube.com/watch?v=6Q8oQGBcFdo

My point is: the Chinese government has been trying to tell people that they don't need to work in the city. They could work in a rural farm and lead a simple life, albeit earning close to nothing.


In the end that's how dictatorships die.


The Foxconn employees are not doing anything really productive for China.

They’re in low value jobs that keeps themselves busy, but doesn’t bring in much profit for China, except for Foxconn and Apple.

Apple benefits from cheap Chinese labor, while selling their iWares at inflated prices.

What China can do instead, is to retarget these factory workers to work in software development, or AI grunt work.

Currently, AI requires a lot of manual human data labeling, in order to identify objects.

China can utilize this already cheap and organized factory labor, to build up an object inference database. That way, the AI systems of today, can cross reference against this inference database, to identify the object.

This inference database literally becomes, the new oil of the AI age.

And through a lot of sweat equity, China can possibly dominate this key area for the AI future.

Maybe this is why 5G is so important? The lower latency of 5G would usher in the AI age.

The western world is so focused on Foxconn, and its “supposed benefits” to China. But the benefits are meager. The biggest beneficiary is Apple itself.

China should let Foxconn and other low value industries go, in order to free themselves to work on higher value pursuits.


It doesn’t matter if you think they’re doing something productive for China Apple or anyone else, and the idea that they could suddenly repurpose them to do “AI” is incredibly far fetched. My point is hundreds of thousands if not millions of workers would be suddenly unemployed in such a trade war which would be an economic catastrophe, there’d be unavoidable civil unrest.


This is true. The key is to retarget these factory employees to do something else before that happens.

The other side of the coin is what will happen to Apple and the American economic system?

They will have to pick up and relocate to other third world countries, for their cheap labor. Vietnam, India, Mexico? Sure. But a lot of other mid level goods come from China.

If the supply chain is stopped, then everyone is impacted. The whole thing stops.

Apple’s stock price will crater. Supposedly. But the Fed is so busy juicing the market, and Apple can just repurchase their shares, that they can financially engineer themselves out of this stock meltdown scenario.


That number seems extremely low. Is it offset by purchases of American military equipment?


Options listed in article:

- direct military invasion

- disinformation campaign

- trade war with Taiwan

- nationalization of old fabs in China

- missile strikes on TSMC facilities in Taiwan


Those are not really 'options' because they have existential consequences.

'Missiles'? Really? Anyone can theoretically use 'missiles' to knock out the production capacity of some competitor's fabs.

It would be fatal to China's ambitions in everything because the world's reaction would be quite strong.

China has a lot of people upset around the world, but a lot of voices are muted because of perceived repercussions, but something 'over the top' would encourage all of those voices to come out at once.

Even Russia would have to 'think again' about their relationship.


Actually Iran used missiles strikes against Saudi Arabia oil facilities very recently as an argument in negotiation. No strong world reaction.


The US executed Iran’s top general three months later. While it wasn’t directly related, I would be stunned if it wasn’t at all related.


> Actually Iran used missiles strikes against Saudi Arabia oil facilities very recently as an argument in negotiation. No strong world reaction.

But didn't they do so through proxies, which gives them at least a little deniability?


They said it was proxies, but many signs say they executed attack themself.


You make a point but what I think the author is alluding to the fact that the US and allies have countered Chinese military actions (building bases in international territory) with words. If that track record holds, we may also use words to counter a missile strike on a TSMC fab.


If China attacks Taiwan it would lead to a war. If the USA backs down they are done as a power, the dollar would crash and the US debt would become unsubstantiated. Given that the war would escalate and China has an advantage of being close, the effort the USA would have to put in would cause a great deal of casualties, which would piss off then public in a Japan Pearl Harbor way so we could never back down. The logical course of action given the difference in ICBMs would be for the the USA just to nuke China after they fire the first missile. That’s the real world math. It’s highly likely the USA could destroy China without a great deal of damage. The Russians would sit it out hoping to become stronger after. Not making a moral judgement, just a military & economic one.

Feel free to disagree. I expect a storm of downvotes. Oh well.


Do you honestly think the USA would be willing to suffer even one city nuked for any reason at all, let alone to protect some small island in Asia most americans couldn't even find on a map? In my estimation there is absolutely no way they would; the question is laughable. China has over 50 ICBMs. The USA is not going to get into a nuclear exchange for Taiwan.

And it's pretty questionable whether the USA could even do much to stop a Chinese invasion of Taiwan. They could perhaps harass it a little, at great distance to avoid the otherwise inevitable loss of their carriers, but the success or failure of any invasion would largely depend on Taiwan's military and political/national will.


Is is possible the USA could wipe out the 50 ICBMs before they are launched. It would take around 15-20 minutes for an ICBM from the USA to reach China. A sub off the coast would take that down to less than 5 minutes. If the President gave the order to launch from the subs as soon as the first Chinese missiles hit Taiwan it would be seen as an acceptable risk. The real politic of this has nothing to do with Taiwan per se, but everything to do with the fact that everything is priced in dollars. Allowing China to invade Taiwan without doing anything places the dollar as the world currency in jeopardy. The ability of the USA to borrow money to fund its government and economy is key. Loosing that would pull everything down. So yes, I do believe the US would trade a city or 2 in order to ensure its place at the top of the world order.


Lol, no it is not possible. That is not how any of this works. You see to think the location of China's nuclear arsenal is on Google Maps or something. It is not.

In addition to its estimated 10-15 road-mobile DF-31 ICBMs, China has 6 active submarines with 12 missiles each. Even assuming a 50% reliability rate that's 30 cities. The USA would be utterly devastated.

Russia might then decide to simply finish the job.

Your ideas are just so ridiculous I don't need to argue further.


15 mobile launchers which the US spends a ton of satellite time tracking anything that looks like one. There are 1152 warheads just on the current US SSBN. They would target anything that even looks like a launcher. Assuming half are on Atlantic patrol, that is still 576 warheads. 15 real targets and 561 guesses. The odds are pretty good.

As to the subs the latest China sub is the type 094 which cannot reach the continental US with their current load of JL-2 (7200km) in their current patrol zone. Also they are pretty noise and easily tracked. JL-3 (9000KM) are not in wide service.

The US has a further 10600 or so warheads on ICBM. China’s total count is around 400.

The logical game theory response is for the US to go full nuclear day 0. That is what every general and every theorist in the pentagon will advise the president. I am not saying it’s a good idea, but the logic is sound if you except the fact that the US would be unwilling to allow the dollars position to be challenged.

The leadership of China is pretty damn smart and they know this. I applauded them in playing the long game, something which the last few American governments have forgotten how to do. Our inability to think long term will bring us down, not China.


There is no way, full stop, that the USA can track random trucks in China. Even if they could, of course they have nuke-hardened storage - we call them "mountains". And that's 15 we know about, who knows how many more there really are. My understanding is that the PLAN subs absolutely could strike the US. And it goes without saying that US forces in SK, Okinawa and Guam would be utterly obliterated.

But all of this is ridiculous. Yes of course the USA has more warheads. They could indeed cause catastrophic destruction in China. No-one said they couldn't. The point is that China can and would counter-strike, with devastating results. This is the whole point of nuclear deterrence. No-one believes they couldn't, except apparently you with your completely unrealistic fantasies about the USA's godlike superiority. The USA as you know it would be gone.

And you somehow believe the USA would trigger this frigging holocaust to protect its image, or the US Dollar? What Dollar? The Fed's a smoking crater. Just delusional.


Mountains and fixed silos do not protect much given the current CEP and over pressure capability of the US Mk5 Trident 2 SLBM that can reach China’s fixed ICBM in 10 minutes.

Here is a very detailed story on warhead guidance and accuracy and how harden targets are destroyed. It is quite interesting.

https://thebulletin.org/2017/03/how-us-nuclear-force-moderni...

Every fixed ICBM location in the world is known to all the major powers. You can find a pretty good list on fas.org. It will including things like year built, what was the state of the art at that time, yield needed to destroy, etc. Also there are tons of research papers on first strike imbalance and launch reaction time done by well respected scientist and think tanks available on the internet. Reaction time is more important because most silos today cannot survive an accurate strike so you have to launch before the misses hit. If they could then why build mobile launchers?

It’s not tracking a truck, it is tracking a very large thing that is easy to follow that has to exit a know location for servicing over its life time. The NRO has 60+ years of experience tracking way more items then the 30 DF31/31A mobile launchers that China has. The USSR had 1000s for them to practice on.

This has zero to do with god like powers or some unreasonably belief in US superiority and everything to do with math and technology. The logic is very simple, a protracted conventional war with China makes no sense as it will end up as a nuclear war, so if a war starts a massive first strike day 0 is the most logical path. And yes, I do believe the US would do so to protect its place in the power structure of the world and I am a rational person who would love to see all nuclear weapons eliminated along with the need of any or us to maintain large militaries.

The planners do not think in emotion. They do the math and odds and there is a calculation that says an overwhelming first strike has acceptable risk and Guam, et.al. that the smaller weapons could hit are an okay trade off for not allowing a hit on the continental US along with complete destruction of Chinese forces. There are 1000s of people in the US Defense Department who’s only job is to calculate all of the possible scenarios, including this one and wrap it all up in a nice summary of casualties (explosion + long term fallout given winds, etc.) and probabilities of success and they do so without the emotional reaction that you have. It’s quite scary if you really think about it.


If China is overly militaristic, then Japan will arm itself.

China may hate Japan historically for the atrocities in WW2, but they also fear them for the same reasons.


I remember a while back an Adbusters pastiche poster supposedly from Nestlé which read:

Go on then, boycott us! I bet you can't, we make EVERYTHING.

China is Nestlé on steroids. What is anyone really going to do to China? What country can continue to function without Chinese manufacturing?


The factories are in China, but they're making parts for American or multinational companies. And the CMs themselves may or may not be Chinese. Foxconn is Taiwanese, Flextronics is legally Singaporean but headquartered in the US. Those two alone might be close to a million jobs in China. How can Chinese manufacturing and the Chinese economy function without customers?


How long can China last without orders? Or oil? Or minerals from Australia?

If the Chinese political situation broke down to the point that they were bombing fabs in Taiwan, there will be plenty of ways to destabilize their efforts.


> 'Missiles'? Really? Anyone can theoretically use 'missiles' to knock out the production capacity of some competitor's fabs.

Yeah but PRC would totally do it. They already gave the go-ahead to their allies to attack U.S. partners with missiles, a first-party strike is really not hard to imagine.


Looking at history and the strategic positioning of the forces I’d say the US is more likely to strike first.


- Encouraging NK to start the missile rain towards SK (bye bye Samsung)


That does not really help China.


I wrote that in reply to - missile strikes on TSMC facilities in Taiwan

Wouldn't that be all out war? Then why stopping half way? If the goal is denial/disruption of chip production.


It provides cover. Between that, and racial tensions in the US, the US military will have their hands full.


Thanks for this article, dolftax! I followed all the examples on my machine with no problem, and I learned some new stuff.

I have a question: how difficult is it to implement the ast? It seems like that the bulk of the work for this static code analysis.


"Crafting Interpreters" by by Bob Nystrom (https://craftinginterpreters.com). Although the book falls short in covering static analysis (obviously), implementation of ast is covered in detail.


For this kind of (read: simple) static analysis, yes, the design and generation of the AST is the dominant factor. For more advanced techniques usually you'd start with an AST directly to not have to deal with parsing and then work on that.


What source do you have that Canada has disproportionate contributions, compared to the US? The Wikipedia page doesn't say that.


Do you have a link for this? I see free meal for 18 and under in SF, but not for adults also.


They don't ask for ID or proof of any kind.

https://www.sfusd.edu/services/health-wellness/nutrition-sch...


Herd Immunity (https://en.wikipedia.org/wiki/Herd_immunity) means that everyone doesn't need to get it. The threshold for COVID-19 is estimated to be between 29-75%.


On a different note in scientific paper what is considered too large an error margin to consider it a value to talk about. If the above is true that makes it what 52%+- 23


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: