My initial thoughts on this came the from the opposite side in terms of Taiwans security.
One of the major bounties for invading Taiwan would be in the acquisition and control of their chip manufacturing. Which - as you mention, the entire tech world is dependant on.
By moving some of that capability offshore, that incentive is gone.
> By moving some of that capability offshore, that incentive is gone.
China's main motivation isn't about controlling TSMC, that's just a useful side-objective if it comes to pass. Their main motivations are:
1) break the first island chain barrier and gain a naval base with unhindered access to the Pacific, and
2) shut down a high-functioning Chinese democracy that is a constant reminder to the people of mainland China that democracy works for Chinese people and that they don't actually need the CCP.
These are also the reasons the US and Taiwan's other allies like Japan will continue to defend the country even if it moves some chip production to safer locales.
Perhaps it isn't about controlling TSMC, but control of TSMC would be a very strong leverage over the west.
Not that it is the only or even the most powerful leverage China has over the west, but it is still a massive one, and it has quite a lot of second-order effects.
> but control of TSMC would be a very strong leverage over the west.
They'll never control TSMC by invading Taiwan. TSMC will sabotage their equipment if it looks like a CCP invasion is about to succeed. CCP knows that, it's not their main reason for wanting Taiwan.
China wants to focus more of their economy on internal demand. Something like TSMC, which would allow them to fab CPUs and own more of the value chain in things like laptops and cellphones would help them reach that goal.
Everyone is aware that this would be scorched earth, right? If the fabs somehow survive the initial wave of strategic bombing both the US and Taiwan have an interest in preventing them from falling into enemy hands. In addition, the US will then place China under a trade embargo. And that's assuming that the US doesn't actively engage PRC forces.
To say this would make the world worse off is a drastic understatement.
China absolutely wants to be reunited with Taiwan. Taiwan is a constant reminder of the century of humiliation (Translated term that is basically China's term for Opium wars to WWII). Taiwan is an integral part of the Chinese nation and for nationalistic reasons, it would be an issue even if Taiwan had no other benefits.
> Taiwan is an integral part of the Chinese nation
No it isn't. China under the Communist Party has never ruled or controlled Taiwan (since 1949). Taiwan democratized, developed, and got wealthy first, completely independently of China. Taiwan has never in modern history been an integral part of China.
There's a difference between a nation and a state, although the two are nearly always synonymous in the modern world.
Taiwan and China being one nation is the policy of the governments of both China and Taiwan.
Of course Taiwanese nationalism is it's own thing now, but the Taiwanese people seeing themselves as not Chinese is a relatively recent phenomenon - it's been functionally independent for less than 100 years, and before that it was a Japanese colony like several others that have since been reabsorbed by China.
I'm not supporting Chineses irredentism, and I don't think the parent comment was either. Taiwan should remain independent. What the parent was explaining is why China would want Taiwan no matter what - it's a historical part of China that is relatively recently separated, so they want it for purely nationalistic reasons. It's no different from Serbia and Kosovo, or Russia and swaths of Ukraine. The economics don't matter if all you care about is your wounded national pride.
> Taiwan and China being one nation is the policy of the governments of both China and Taiwan.
No one in Taiwan and not even the current government consider China/Taiwan together. When KMT fled and created the constitution they claim China. Taiwan is now stuck in limbo because the people just want to live their lives in peace and already consider themselves Taiwanese and independent. But if they change the constitution then China will consider it a formal act of independence and use it as an excuse.
You're speaking to the "state" part of the above comment. Yes, clearly the PRC and ROC are not the same state. "Nation" is not necessarily the same thing.
I’m saying despite the constitution. People in Taiwan don’t claim China to belong to them. This is a relic of the past that is stuck in writing that people suffer from.
The pro-independence party just suffered a major loss in local elections. The prime-minister who invited Nancy Pelosi for a provocative visit had to resign. Looks like Taiwanese people are not very enthusiastic about becoming a new Ukraine.
Greens never do well in local elections. Local governments have no say at national level. It seems the only people who say they suffered a major blow are pro CCP bots.
For clarification: It was the President of Taiwan Tsai Ing-wen(as far as I can tell) who invited Nancy Pelosi for a visit. She (the President of Taiwan) did resign as head of her party, but not as president.
China has been a country for longer than the history of European civilization. They do not subscribe to the view that nothing that happened before 1945 matters the way Americans seem to.
China also makes the claim of 1000s of years of history but Taiwans inclusion in that history is about 200 years. And of that less than 10 was a province that China still didn’t govern or control. It was more or less just something they said to deter Japan. So historically China has never ruled over Taiwan. Japan has more claim to Taiwan than China as it actually ruled and controlled Taiwan.
No, China has not been a country for longer than European civilization. That CCP propaganda claim is the equivalent of saying that the European Union has been around for 8000 years because Plovdiv, Bulgaria was founded in the 6th Millennium BC.
Without the staff from TSMC, the Chinese Government probably can't run those facilities. I suspect there isn't much probability that china could mobilize and take those facilities as-is with no sabotage.
ATM PRC assaulting Taiwan would hobble western defense industry. By moving more manufacturing into NA & EU that alleviates it. It helps Taiwan as well who depends on western defense industry.
The Chinese government doesn't run those facilities now, and I agree that they won't be able to do so in the event of a Taiwan invasion either.
That's not the point though. Who runs those facilities now? TSMC aka Taiwan aka a western ally. Who stands to lose the most from TSMC facilities being burned to the ground? Taiwan (obviously) and the west.
As a layman with no background in international relations, to me it seems like TSMC is simply an extra bargaining chip for the Chinese government. Which is why I am all about the idea of building more TSMC facilities in places that are less susceptible to being invaded. And yes, the Arizona plant is just a drop in the bucket compared to their facilities in Taiwan, but you gotta start somewhere, and something is better than nothing in this case imo.
The Chinese government also wouldn't be able to run those fabs without ongoing support from ASML and other key foreign vendors. The production machinery is extremely complex with many specialized parts and a significant software component. Reverse engineering and duplicating everything would take years.
It seems like there are a few key parts that could be removed that would make those ASML machines utterly useless even if Chinese engineers spent time reverse engineering the rest. Surely there is no need to remove or destroy the whole thing as is suggested in other comments.
The equivalent of popping out the Intel/AMD/ARM CPU to disable a computer, and leaving the motherboard, RAM etc.
I've read that TSMC fabs are filled with explosives to set off if China invades so they wouldn't get their hands on tech and equipment. I'm pretty certain that's true, given how important TSMC tech is.
The difference is both sides want that gas to keep flowing (for the time being at least).
Both sides don't want China to take over the technology to produce state of the art chips and control the distribution of those chips. And in case of a war, no way is china continuing to sell those chips to the US defense dept. The profit is negligible compared to their strategic value. Different goals will produce a different result.
That meme makes zero sense considering TW don't want to be a third world economy dependant on exporting fruit even if PRC successfully invades. There's reason TW media was telling US think tanks to leave TSMC alone when US Army War College analysis suggested US should consider bombing TSMC... or exfiltrate TSMC engineers (before children no less) in event of war. As long as fabs and downstream supply chain supplies said fabs are intact, the island will have leverage to remain viable modern economy to support relatively affluent lifestyle. Note the point on downstream supply chain, there's sufficiently exclusive niche semi industries sustaining TSMC on TW that makes it as critical as ASML. Don't expect any Arizona TSMC fabs to operate smoothly without them. If anything, expect TSMC and TW + PRC to collude to threaten TSMC US fabs if try to sanction TSMC TW from making chips in event of successful PRC takeover. The people whose making bank off TSMC will want to so regardless of who rules the island. Ultimately short/medium term also in US interest to keep fabs going because not enough fab capacity will be reshored off island for long time, and the interest groups hurt most is US high tech industry who extracts disproportionate value add from TW fabs. Imaging every company that depends on leading edge chips turning into Huawei/ZTE because 95% of production goes kaput. Currently, cratered TSMC fabs actually works in sanctioned PRC's favour because it dramatically closes relative gap of who has access to high end semi. PRC vastly better off in balance where everyone is mostly stuck on 28nm+ instead of one where US has unfettered access to leading edge.
US military doesn't care. But the TWnese care, which US thinktankers/media, and I'm guessing US based commenters like you seem to forget. Hence the disconnect on why people seriously contemplate these TW will blow up TSMC memes. And why TW media reminding US, that if they're going to evacuate anyone off the island first, it's not going to be their semi engineers, it's going to be women and children. Or that more generally, they're not interested in blowing up their lively hood to stick it to the PRC. Like how in UKR war, it's RU whose blowing up UKR infra and industry when they decided it was better to scorch earth long term.
Yes, Taiwanese obviously care more about their children. But that doesn't change that the US military won't allow the Chinese to obtain TSMC or the knowledge of their engineers.
Sure, except original comment also highlights that the incentives of destroying TSMC is backwards. It's PRC who benefits most from denying US access to TW semi supply chains or engineers not vice versa. Denying TW to US closes relative semi gap for PRC, leveraging PRC control of TSMC in case of successfuly invasion to compel US to lift sanctions also closes relative semi gap for PRC. US has leverage via sanctions during peacetime, but PRC has leverage via threatening access or destruction of east asian semi supply chain during war. Utlimately it's in both US and TW interests for TSMC + co to survive because they extract most value / benefit, but not necessarily for PRC. And for TW, ensuring TSMC+co survival =/= paperclipping them to US. All the interest calculates points towards PRC/TW denying US access, and US wanting continued access since US fabs will still be dependant on TW inputs as much as current TW fabs or future PRC controlled TW fabs will be dependant on US/EU/JP inputs.
But this is missing the obvious. Yes the US would prefer being able to maintain its tech advantage over China by continuing production and receiving of state of the art chips. And this is clearly better for them than choosing that gap by destroying TSMC fab. But if China takes over TW, that's not an option. The options are either flatten the gap or flip the gap in China's favor as it now controls those chips while the US falls back to tech multiple generations old. There is no other strategic option for the US flatten the gap is the only choice, flipping the gap is the worst case scenario for the us.
This deal is in US and TW best interest. TW is still valuable as cutting edge is still made on island. So production and economics so benefit them. US still provides military protection / defense. They are still long term partners will aligned goals. And if China does invade TW, US can continue a long fight and win by falling back only a generation or two to is smaller but important domestic production to continue supporting its defense technology needs.
I can't speak for them, but I have to assume RW also sees its best interests if China invades to lose TSMC plants but US win the war and they maintain democracy, rather than keep TSMC plants but controlled by China and US lose the war and go under full control of mainland rule.
Plants can be rebuilt, just like Marshall plan or what will happen in Ukraine. TW wants long term freedom from China - strengthening US defense's tech position helps this the most.
The obvious option in event of PRC control is everyone deferring to mutual leverage, where US controls fab inputs, TW controls outputs, PRC negotiating access to % of high end production by controlling island. Meanwhile buys both US and PRC time to secure independant semi chain, TW gets to still profit off semi, much like pre techwar arrangment. That's the strategically rare win-win-win scenario where everyone gains / avoids loss vs having accessible leading edge setback 20 years. Of course PRC wins most in this arrangment, and US/TW relatively loses, but if PRC wins in TW scenario vs US, that's going to be least of worries / concessions. The least obvious option is US deciding to scorch earth tech that supports their most competitive high tech industries, responsible for huge % of competitive advantage, especially if that leaves PRC the biggest producer of mature nodes (where things are trending). Even less sensible considering long term timeline after which experts believes PRC can reach semblance of semi parity.
This is short/medium term balance / interest calculation irrespective of the deal - which I didn't comment on - substantial capacity won't shift off TW in the timelines US forecast PRC will make move. Leading edge isn't particularly relelvant to defense hardware ends which use mature nodes, nor does anyone project long fight with how irreplaceable modern platforms are. It's why PRC (and chip partners US is trying to coerce into export controls) views Oct chip curbs as attack on their commercial sector, not military since PRC also have small batch 7nm production capability.
> TW also sees its best interests
It's also obvious "best" TW interest is to hedge having continued leverage over being leading ledge supplier regardless of who wins. There is no scenario where giving away their silicon dominance makes sense, unless coerced to, which is why US had to unilaterally announce export controls - because CHIPS4 partners weren't biting and actively pushing back. Least of all because no one is sure TW can win local war against PRC. Wealthy TW industrialist/elites/chip talent would rather be wealthy under PRC control after war then be fruit farmers because their industry got glassed. Read between line of Morris Chang stating US efforts to reshore chips manufacturing doomed to fail - it's not commentary on US talent - but TW isn't ceding industry to be Americanized. IMO don't expect these plants to launch without hitch. Prolonged war is probably least in TW interests because island simply can't sustain past a few months on domestic resources. There's reason TW defense posture is not prepping for prolonged war like US wants, deep down leadership and industry knows best interest for TW is to survive short/sharp war as developed economy and work with whoever wins.
> strengthening US defense's tech position
Except, again, TW offshoring leading edge fabs doesn't substantively strengthen US defense indy that depend on mature nodes with respect to PRC war. As stated, US fabs will STILL depend on TW inputs for forseeable future - it's not just ASML that's bottle neck - niche suppliers and expertise will be stuck on TW long term. Also consider "partners" pushing back on unilateral US curbs or SKR, JP building out their own indigenous defense industry - very few are actually interested in giving/strengthining US further leverage especially at their cost. And let's be real, fabs aren't going to be rebuilt on TW if they get leveled when they can be reconstructed in more secure locations.
> By moving some of that capability offshore, that incentive is gone.
I'm not sure if that incentive was significant enough. In my eyes, PRC's interest in Taiwan is at best orthogonal with TSMC's manufacturing capabilities.
I thought the same thing. I believe it is more nuanced than that.
PRC wants Taiwan regardless of its production value (geopolitical & PRC narrative) though I'm sure they would like the control of chip manufacturing. However if the US did move all of its strategic production off island there would be less value accrued to defending outside that said keeping a close presence on China expansion is important to the US. So it would still have value maybe a bit less so.
Please excuse the human aspect of the population as we are separating that part of the discussion.
More bleakly, if TSMC is destroyed as a result of invasion, the rest of the world is denied access. This might be to their liking. With some of the capability in the US, they can no longer deny the rest of the world.
I actually really like it. I've been looking at IntelliJ for 8 years, the refresh is really very welcome XD. I also haven't noticed any lost functionality.
I agree that the additional value add over a video call might not be there, though the author did seem to speak of Workroom being somewhat compelling.
But remember that a single Telepresence room can cost $60k - $600k, plus an ongoing subscription for networking + management, without the capacity or flexibility of a virtual room. Every meeting room in my office has one.
This is touched on in the article, but remember theres a difference between how individuals and corporations spend money, especially if it can lead to [percieved?] gains in productivity/team cohesion.
See it through. I know sometimes at university it can feel like your having your wings clipped, or that your not getting the 'real' experience.
But it won't last forever, and what it does well - conveying often unintuitive OOP concepts - it does do well.
Your course conveners job its to teach the fundamentals, tooling is for you to shake out as you gain experience and scale up, and from their perspective - imagine trying to wrangle random IntelliJ/Eclipse issues with a cohort of 120+ students, for many whom this might be their first introduction to programming!
Scaling things back probably genuinely allows them to provide a better learning on ramp.
I'm still in high school so the scale of everything would probably not be an issue although I see your point of maintainability and showing the relations between classes graphically is probably also helpful when you're still wrapping your head around OOP concepts
The borrow checker makes a doublily linked list difficult, from my understanding.
They're almost always a bad data structure to pick though, as the speed difference between being able to iterate a contiguous list, and even completely rebuild it, vs having to follow arbitrary pointers mean that the 'complexity' improvements of a linked list often become irrelevant.
------
Open Question)
I have this intuition that the borrow checker might enforce a way of working that fundamentally stops you from doing things that aren't _really_ compatible with the way a CPU actually works on the lowest level.
Is there any credence to that, and if so, is it happenstance, or is there some more fundamental reason?
> I have this intuition that the borrow checker might enforce a way of working that fundamentally stops you from doing things that aren't _really_ compatible with the way a CPU actually works on the lowest level.
No, I think you are reading too much into it. One is free to Box-allocate/pointer-chase on every line in Rust without any trouble. Sure, as other low level languages, rust also prefers/makes stack-allocation more ergonomic/the default, which is a quite cache-friendly way of operation but not even the stack is “native” in and of itself - it is just a frequently and extensively used part of the “heap”.
I think it's safe to assume this class of vehicles (and likely any new class of vehicles from here on out) is not designed to be piloted by a human past its prototype stage.
In business management, financial markets, hiring, investment, real estate patterns etc etc.
There have been some lasting effects, but it was always clearly a temporary, if not prolonged situation.