No dog in this fight, and I agree with the premise, however there was never a time Apple made a ton of noise about MacBooks being manufactured in the US.
There was a ton of noise about Mac Pros being manufactured in the US, but sadly, I am not nearly as familiar with Apple after, say 2018*. Not even sure if they have a Mac Pro anymore. :X and if they do, I assume it's not the same model (the black trashcan), so it makes me wonder if they bothered retooling here, or quietly moved it somewhere else
* TL;Dr at some point it became clear to me Cook is Sculley 2.0. I date it to around walking around NYC and seeing an absurd amount of Apple News bus-stop ads. Services! (TM)
Apple won the phone wars because it did not allow the Microsoft Windows playbook to repeat itself. The only way to stop that was to hire an expert from a PC company and deploy significant American capital to buy up all the available offshore capacity to efficiently produce that new class of product before the potential competition could even take it seriously. I blame Apple for a lot, but going to bed with the CCP is not one of them. They were forced to play the game that was created by the Taiwanese PC clone manufacturers that used cheap Chinese labor to outcompete American based manufacturers, including Apple. Blame AST and IBM (who, after opening the hardware to create the PC clone business, ironically sold its own to Lenovo in a final spit on America.)
If Apple didn't take such an investment risk in China, then we'd all have Windows Phones now and Android would have been canceled long ago.
Bringing it back? The DNC pivoting to expand entitlements beyond sick, elderly and legacy civil rights victims to leverage economic downturns greatly expanded their political base in California and also played a major role in locking the door behind Apple, CISCO and others, leaving all the supporting manufacturing tech and processes to evolve elsewhere. Consider what happened when a billionaire tried to open up Tesla down the road from Apple's original Fremont manufacturing facility. Texas has grown rich building up as a California alternative, but doing more than moving people out of California, actually bringing back overseas operations that would have gone to California had it not devolved into an economic wasteland, required a powerful financial incentive and Trump has finally given one.
> Texas has grown rich building up as a California alternative, but doing more than moving people out of California, actually bringing back overseas operations that would have gone to California had it not devolved into an economic wasteland, required a powerful financial incentive and Trump has finally given one.
Wasn't Texas a tech hub dating back to TI?
I doubt anyone will seriously bet on any of Trump's changes since his mercurial temperament and tendency to chase the shiny makes it all so unreliable.
You really think Apple's success with the iPhone is because of them manufacturing them in China? That has nothing to do with it, as hardware pricing is not why Windows Phone failed or even why Android is declining.
If anything, Apple won by playing exactly the Windows playbook with software: Embrace, extend, extinguish. I'll add a new one: Restrict. (App Store)
Much as it might be pandering to Trump's nationalist (America first / American only) policies, or simply an action to avoid some of the effects of tariffs that might be imposed, this time around, I see no such connection for the 2021 announcement. Unless they are connected as they are a more generic "pleasing the incoming administration" to try curry favour for when decisions that might affect the company are being made.
Setting specific parties and people aside, that’s not what “the base” means. The base is the diehard supporters. Many an election has been won by taking a minority base and adding in people who are not supporters but chose that candidate for whatever specific reason, e.g. “the price of gas.”
He won with fewer votes than Biden got to win in 2020. And not all of those are his 'base'. Many probably didn't even know who Biden is, they just voted for the guy promising to lower prices.
"Apple’s most recent announcement on US investment was a 2021 promise to spend $430 billion over the following five years, including a 3,000-employee campus in North Carolina, though development on that project has since paused."
- a larger investment number in a previously announced Austin campus
- new factory in Houston "which will create thousands of jobs"
- "doubling its $5 billion US Advanced Manufacturing Fund to $10 billion"
- "It will also open an Apple Manufacturing Academy in Detroit in which Apple engineers and other experts will offer consultations to local businesses on “implementing AI and smart manufacturing techniques,” along with free classes for workers."
Some useful context: this is almost certainly being driven by Apple’s Private Cloud Compute architecture and not tariffs, as an investment of this magnitude is not planned overnight.
Why is PCC driving Apple to spend billions to build servers in the states? Because it is insane from a security standpoint (insanely awesome).
PCC is an order of magnitude more secure server platform than has ever been deployed for consumer use at planet scale. Secure and private enough to literally send your data and have it processed server side instead of on device without having to trust the host (Apple).[1] Until now the only way to do that was on device. If you sent your data for cloud processing, outside of something exotic like homomorphic encryption[2], you’d still have to trust that the host did a good job protecting your data, using it responsibly, and wasn’t compromised. Not the case with PCC.
To accomplish this Apple uses its own custom chips with Secure Enclaves that provide a trust foundation for the whole system, ultimately cryptographically guaranteeing that the binaries processing your data have been publicly audited by independent security auditors. This is the so called hardware root of trust.
It is essential then that the hardware deployed in data centers has not been physically tampered with. Without that the whole thing falls apart. So Apple has a whole section in their security white paper detailing an audited process for deploying data center hardware and ensuring supply chain integrity.[3]
You can imagine how that is the weak point in the system made more robust by managing it in the US. Tighter supply chain control.
Trusting Secure Enclaves custom chips over processing locally is going to be a hard to impossible sell for those who truly care about privacy.
Thankfully for Apple that's a very low number in a world where people demand tiktok remain legal when shown how their data is being used by foreign actors. People only care about privacy when it's local (don't want mother to find out, neighbours to talk, friend to think a certain way about you or classmate stalking) and that's why ai fakes are much more concern then a company knowing everything you do.
But this product is great for fortune 500 businesses.
I think this is a level of security Apple is providing at additional cost to themselves that only a tiny fraction of consumers would even pay an extra cent for.
From that perspective I really appreciate this effort by Apple.
Yup. Apple knows that they don't have to ship anything more than a whitepaper to justify their stance to current customers. They could announce an internet-connected bidet with a webcam and there would still be people arguing that it's safe until someone exploits it.
The fact that Apple is comfortable shipping a whitelabel ChatGPT is proof that the whole Private Cloud Compute thing is just for show. They're perfectly happy partnering with the Worldcoin guy to sell you something popular if there's money in it for them. Apple knows people expect them to release some haughty whitepaper, so they cook up PCC and claim you can audit it if they think you're worthy of seeing the insides. Now all the privacy nuts can pipe down while Apple plans a longer-term strategy to make their hardware compete in the datacenter.
There is a world where Apple takes their own privacy commitment to the next level through radical transparency. But that's not what PCC is, it's another puppet for the Punch-and-Judy security theater that sells their iCloud subscriptions.
PCC is completely different from the ChatGPT integration. ChatGPT is indeed not a privacy-hardened system, but Apple devices only use it for so called “world knowledge” queries and make you confirm when calling out to it, typically involving limited personal data.
PCC is designed to handle extensive personal data, and the auditing is attested by cryptographic proofs provided to software clients, not just white papers read by humans. It is significantly different from what we’ve seen before in the industry, and highly worth the effort to understand it if you are at all involved in server engineering.
The problem with any remote arrangement is that you have to trust Apple that the server side is running all that stuff. Their answer to that is "you can audit us", but I don't see how that would prevent them from switching things in between audits.
As far as local processing goes, though, you're also still fundamentally trusting Apple that the OS binaries you get from them do what they say they do. Since they have all the signing keys, they could easily push an iOS update that extracts all the local data and pushes it to some server somewhere.
Now, I don't think that either of these scenarios is likely to happen if it's down to Apple by itself - they don't really gain anything from doing so. But they could be compelled by a government large and important enough that they can't just pull out. For example, if US demanded such a thing (like it already did in the past), and the executive made a concerted push to force it.
> When a user’s device sends an inference request to Private Cloud Compute, the request is sent end-to-end encrypted to the specific PCC nodes needed for the request. The PCC nodes share a public key and an attestation — cryptographic proof of key ownership and measurements of the software running on the PCC node — with the user’s device, and the user’s device compares these measurements against a public, append-only ledger of PCC software releases.
> compelled by a government
Sadly, the bar is much lower than "compel". Devices are routinely compromised by zero-day vulnerabilities sold by exploit brokers to multiple parties on the open market, including governments. Especially any device with cellular, wifi or bluetooth radios. Hopefully the Apple C1 modem starts a new trend in radio baseband hardening, including PAC, ASLR and iBoot, https://www.reuters.com/technology/apple-reveals-first-custo...
> Their answer to that is "you can audit us", but I don't see how that would prevent them from switching things in between audits.
PCC does actually prevent Apple from switching things in between audits to a high degree. It’s not like a food safety inspection. The auditor signs the hardware in a multi party key ceremony and they employ other countermeasure like chassis tamper switches. PCC clients use a protocol that ensures whatever they are connecting to has a valid signature. This is detailed in Apple’s documentation.[1]
See, this is why I think privacy engineering is low key the most cutting edge aspect of server development. Previously held axioms are made obsolete by architectural advancements. I think we’re looking at a once in 15 year leap - the previous ones being microservices and web based architecture.
At some point having trained and certified Apple engineers overseeing this sort of thing gives far more confidence than random startup #1345134 who promises they hired the best college drop outs that they could find.
<<<security is not my domain, asking genuine questions!!>>>
At the end of the day, it ultimately still boils down to trust though, yes? Trust that they are running the data centers the way they say they are, trust that their supply chain is what they say it is, and so on? At the same time, using some open source piece of software also entails a great amount of trust: I’m not going through the source code of Signal myself, and I’m also not checking that an open source locally served model isn’t sending traffic/telemetry etc back to some remote server via whatever software is running the model… rather, I’m placing my trust in the open source community that others have inspected and tested these things. I’m sure all sorts of shady PRs into important open source code bases are made on the reg after all. So that’s not to say that trusting Apple is necessarily more or less wise than trusting open source software from a security standpoint… my point is just that it seems like they are aspiring to a zero trust architecture, but at the end of the day, it does still require trust that they are operating in good faith vis-a-vis what they are representing in the white papers right? To me, it seems like a relatively safe assumption that they are for a variety of reasons, but nonetheless, it is an assumption right?
> I’m placing my trust in the open source community
You’re right, security is a matter of degrees not absolutes, but open source software requires considerably less trust than closed source. Right?
PCC applies this principle by making the binaries it runs public and auditable by you or anyone in the security community. (In some cases the source code as well.) The craziness is in the architecture that provides cryptographic proof to clients that the server they’re connecting to is running an audited binary and running on secure hardware. It even does TLS termination at the shard level so you can have high confidence that if the binary isn’t connecting to anything your data will be unreadable by any other server in the org.
So it goes way beyond trusting what the whitepaper says. Data center hardware deployments are audited by a third party that signs the servers in a key ceremony. That ultimately undergirds the cryptographic attestation that servers provide to clients that everything has been audited. And it’s also the element that tighter supply chain control helps shore up.
If you’re new to security the architecture documentation I linked to is a very friendly read and a good intro to some of these threats, countermeasures and rationales.
Thank you for the really great response! It answered my main question:
> The craziness is in the architecture that provides cryptographic proof to clients that the server they’re connecting to is running an audited binary and running on secure hardware.
I definitely missed this concept when skimming the links before posting my comment - very very cool!
> open source software requires considerably less trust than closed source. Right?
Of course… but at the same time, I think the difference in the degree of trust I am placing in say, Signal’s end to end encryption and Apple’s (claims of) end-to-end encryption is not as large as it might cursorily seem. Would I be more surprised to read in the news that Apple had secretly embedded some back door than I would be reading in the news that malicious actor managed to push some hidden exploit through to Signal in an otherwise innocent PR? I’m genuinely not sure which would surprise me more, or which event would be more probable, so can I really make any claim as to which is more secure, given the current knowledge I have? Obviously I could think more deeply about this, but superficially, both are requiring pretty large amounts of trust from me - which I don’t think is misplaced in either… though I do personally trust something like signal more at the end of the day based on… what, intuition? A gut feeling?
That’s good food for thought! I would just add that the kinds of threats PCC is primarily targeting, I think, are attacks by malicious third parties (including state actors), rogue internal employees, and privacy-leaking software bugs. These are sort of bread and butter real world threats.
I would go out on a limb and say Apple would love to also prove beyond a reasonable doubt that they too as an organization cannot get away with planting a secret back door — not because they have pure angelic hearts, but because this is good for their privacy-differentiated business model. And PCC certainly makes a huge leap in that direction. But it’s not the problem it’s primarily targeting nor an easy one to solve completely.
As another example, Apple has an implementation of OHTTP onion routing[1] called iCloud Private Relay. It’s really cool and easy to use. The point is to make it so nobody but you can tell what website your IP address is connecting to, not even Apple, the operator of the relay. But bottom line, Apple picks who they collaborate with for the gateways and there’s nothing stopping them from colluding out of band to de-anonymize you if that’s what they wanted to do.
Does this defeat the purpose of iCloud Private Relay? No. Its purpose is to better protect you from common privacy attacks, better than a traditional VPN would. It happens to also narrow the trust you need to place in Apple, namely that they would need to collude with another company to defeat the system as opposed to some rogue lone wolf SRE deciding to access your logs. But it wasn’t put in place to make people who fundamentally distrust Apple as a company start trusting them.
> At the end of the day, it ultimately still boils down to trust though, yes?
Isn't that pretty much the story for most every thing though? It comes down to discernment, which is mostly subjective itself.
Same here. Personally, do I trust Apple? I don't have a leaning one way or another about that. What I trust is that Capitalism is gonna capitalize. And Apple doing what it says here, is its Brand. If down the road it comes out later it was all a lie. That Brand has no more standing. No more standing, no more sales. And Apple is in the Brand/product selling business. I trust they won't throw away their trillions because they would rather sell their Brand on white papers over an actual product that the papers describe.
Yes, I think along similar lines there… but on the other hand, brands need not reflect underlying truths about reality, and in fact often do not. Suppose two years from now, it is revealed by a whistleblower that they were part of a special skunkworks team responsible for creating various backdoors in PCC in order to enable Apple to access the data, train new models on queries, or maybe respond to government requests etc etc, all of which which were subtle, complicated exploits. Maybe Apple denies and discredits, or minimizes, or issues some sort of limited mea culpa. To what extent would it affect Apple’s brand? How long would it stay in the public consciousness? Would people (writ large, not those on HN) care? Perhaps it impacts sales and the stock price, but for how long and to what extent? Obviously there would be some sort of cost to such an event occurring, but would it outweigh whatever benefits that Apple might gain in the meantime? Maybe those benefits have to do with avoiding the wrath of the federal government… who knows. There’s definitely a world where the amoral calculus suggests lying might be better, right? Maybe not ours, but it is plausible. Like you said, discernment is the only tool we have, and it’s difficult to really know what’s going on at the end of the day.
Moscow rules and George Smiley’s tradecraft are probably the only real security… ha!
Absolutely right. My comment was strictly about “for consumer use at planet scale.” It’s the aggressive adoption and rollout of confidential computing architecture in an easy to use consumer platform that I’m celebrating here. (Including a 12 figure financial commitment!) Prior to PCC, smartphones generally had to process data on device to ensure privacy.
PCC is a kludge for mitigating battery life on smartphones doing Personal Assistant work, for knowing what their chances of getting nVidia chip allocations are, for knowing how unreliable nVidia hardware is --basically for having been caught with their pants down when genAI took off. That said, it's a good kludge.
The easy fix is to add more vector cores and RAM to the chips and shrink them to use less power, but it takes time and initially these
go to power cord systems (first in the kludge, then maybe MacPro and some kind of AI-hub that sits in your living room and vehicle), then..well you wonder why the small form factor iPhone just was dc'ed?
PCC is an awesome solution for Apple to ensure that no one other than Apple can execute code in that environment.
That is however not most users' concern (in fact, I'd guess less than 0.001% of Apple users are concerned with supply chain attacks on Apple's servers); what we're concerned with is Apple itself misusing our data in some way (for example, to feed into their growing advertising business, or to redirect to authorities). PCC does NOT solve any of this and it's in fact an unsolvable solution as long as their server side code is closed source (or otherwise unavailable for self-hosting as binaries). For me, Apple Intelligence stays off on my devices (and when that is no longer an option, I'm jumping ship - I just wish there was something at least passable to jump to).
> what we're concerned with is Apple itself misusing our data in some way… and it’s in fact an unsolvable solution as long as their server side code is closed source (or otherwise unavailable for self-hosting as binaries)
It is in fact a solvable problem. The binaries are indeed available for self hosting in a virtualized PCC node for research purposes.[1] Auditors can confirm that the binaries do not transmit data outside of the environment. There are several other aspects of the architecture that are designed to prevent use data from leaking outside of the node’s trust boundary, for example TLS terminates at the node level and nodes use encrypted local storage so user data is unreadable to any other node / part of the organization.
That is a lot of mumbo-jumbo but what it boils down to is that you cannot run the PCC on your own hardware; you can download some "components" whose hash matches the supposed "transparency log" they publish (and some demo models) but since I can't go into my iPhone to say "set PCC server ip: 192.168.1.42" and see it work, I don't trust it (and it cannot be trusted).
> PCC is an awesome solution for Apple to ensure that no one other than Apple can execute code in that environment.
Doesn't PCC guarantee even more than that? From my reading, Apple can't exfiltrate any data to other servers (even ones that Apple owns) nor can they inject any processing other than what is outlined into that server. Otherwise, what's the point of such a stringent hardware integrity requirement?
There is no way to verify that. It's just something they "pinky swear they won't do". The stringent hardware integrity is to protect against supply chain attacks (Apple making sure they fully control the stack down to the hardware and can run any software they want that connects to any external service they desire - such as the CCP, NSA, 3rdPartyAds, etc.)
Nitro is good! And showcases a great many of the foundational architectural concepts in PCC.
But there is a major difference that is germane to the topic of Apple’s investment in US server manufacturing: The hardware root of trust. Hardware tampering is the weak point and afaik AWS doesn’t describe any process to certify their supply chain integrity. I think the most they’ve done is commission a review of their architecture document.[1] PCC actually has an auditor sign each server node in the datacenter.
Thank you for mentioning them though. It’s an important advancement in generally available confidential computing infrastructure.
> this is almost certainly being driven by Apple’s Private Cloud Compute architecture and not tariffs, as an investment of this magnitude is not planned overnight.
The tarriffs haven't happened overnight. They've been discussed for going on 2 full years now. Anyone who wasn't blinded by their own political preferences saw this coming.
I think this is conjecture, there is no indication anywhere that it’s driven by the PCC data centers. If anything I would guess they are trying to build hardware in the US. That has to be the only reason to invest that much.
Coincidentally, construction isn’t set to start until late November 2028—convenient timing. If this mess blows over, they can quietly backpedal and carry on like nothing happened.
Can they realistically cancel their construction contract right before the construction starts? Sounds implausible to me, at least not without huge compensatory fees
Good timing because Trump should be significantly weaker, and it'll be clear where Trumpism is headed in the culture, but also it will be more clear where AI will end up.
Even if the move forward with investment, they will be a bit of a 'late' mover, but will have had a chance to see what is working and what isn't working for everyone else.
1. JD Vance independent of Trump will have the same policies.
2. JD Vance will have enough popularity for a serious 2028 run. He might fall out favour with Trump as Trump tries to mount a bid for a third term, Trumpism might just generally lose popularity if policies lead to bad outcomes.
3. Dems don't figure their shit out. They should be able to take back some control in mid-terms, and then start to push their own policies, or at least credibly show that the most extreme policies from the exec branch don't have teeth anymore.
And you're assuming that the 2028 election (and 2026 for that matter) will be business as usual elections, against all evidence to the contrary staring us directly in the face.
The federal government doesn't control elections in the US so they don't have much power here. Also, firing all the FBI agents is a bad first step to using them for interference. They have no clue how to be authoritarians; to do that, you need to be popular and have the security forces like you.
Parties in the US aren't real, though the brand value is good at keeping the right kind of people in each one. They can't actually kick anyone out and people in the federal government have no particular control over same party people in any state.
From my view, the two parties are quite a bit more real than the 3 branches of government. The people funding state and federal elections are the same.
If you mean campaign funding, that's mostly Actblue (upper middle class Dem donors). Republican donors are like, five people, yes - but voters have their own opinions and they definitely don't vote for whoever has the most money. Because the Democrats have more of it.
The playbook is obvious - if Trump loses, Vance refuses to certify the election results due to "fraud", Republican states will produce alternate lists of electors that vote for Trump, and he will claim that in reality he just won reelection with the biggest margin in history.
How do we know this isn't just crazy conspiracy theory? Because they already did attempt the same thing in 2020, and this time they had the chance to vet the VP candidate for this scenario.
Then we can have a color revolution. He's not doing the right steps to prevent that, because he's annoying the security forces instead of supporting them.
I just checked. US Constitution has amendment #22 that says: "No person shall be elected to the office of the President more than twice". Is there a workaround? I doubt he can change that amendment. None have been ratified in the US for a squillion years.
Going to be watching closely - but cynically, a promise of investment (for avoidance of tariffs) only needs to last one news cycle until tariffs are no longer top of mind. Then it can be walked back without tariffs being imposed.
Maybe instead of saying the t-word tariff, US gov can charge Apple a special fee on each iPhone. They can call it something catchy, like say, a Core Technology Fee.
This should be the top comment. Apple are doing business the way business is done, just like last time. Results don’t matter, it’s economic policy via press release. Form over substance.
Results matter, it's not hard to imagine that Apple considers the real risk of its promise and market position of being the privacy option being undermined by their supply chain risks, and leverage being used against them by privacy unfriendly actors.
What's the added risk here? It's fine to "risk" almost the entire iPhone itself to be manufactured in China but the servers for some random AI features need to be pure?
Sounds more like technical marketing and the company will treat any decisions around it as a marketing exercise.
Apple's commented previously on why they build in China, and it's beyond just the pricing - the supply chain for every single part they use is in China and mostly in the same geographic region, so there's a level of flexibility there they couldn't get in the US. It wouldn't surprise me if it was genuinely a goal for Apple to manufacture more in the US - they're a notoriously privacy-focused (corporate, not end-user) company, and China's known for IP wandering its way off campus. They're not going to sacrifice the iPhone economics until the US option is actually viable, but I'm not surprised they keep kicking the tires on US manufacturing.
> "the supply chain for every single part they use is in China"
Not entirely true. Some of the highest value components in an iPhone, including the CPU/SoC, baseband, and the majority of OLED displays, are sourced from countries that are not mainland China.
> They're not going to sacrifice the iPhone economics until the US option is actually viable, but I'm not surprised they keep kicking the tires on US manufacturing.
Apple could, with its immense cash hoard and cash flow, _make_ the US viable, but it chooses not to because it'd rather take the easy way out and have China or India or $COUNTRY fund it and return money to shareholders. They've returned money to shareholders rather than invest it in US operations, by design.
This is a classic feint to protect Tim Cook's entire raison detre. He built his career on super high efficiency operations by outsourcing to cheap labor countries. It relies on the low-to-no tariff access to US consumer money.
And I don't care that it's better for their stock price; that's Apple's problem not mine as a US citizen. And even as an Apple investor I would rather the money be spent on US on-shore operations.
There is no way Apple as a public company could just burn cash getting everything made in the US, shareholders would revolt long before the money ran out.
> And I don't care that it's better for their stock price; that's Apple's problem not mine as a US citizen.
That is the shareholder’s problems. People like to think that their investments won’t go batsh*t insane overnight.
> And even as an Apple investor I would rather the money be spent on US on-shore operations.
Apple doesn’t even make all, or even most of its money in the states. Not all of its shareholders are American, if they went this route they could lose half of their revenue overnight (as other countries note the protectionism and tariff or simply forbid Apple products from being sold).
This is correct if you only care about Apple stock price. But consider that there are people who simply do not care about Apple's stock price. Like me, an Apple investor. I care more about the US's industrial base than I do about whether it goes up 5% or down 5% (or whatever, it's besides the point)
>Apple doesn’t even make all, or even most of its money in the states. Not all of its shareholders are American, if they went this route they could lose half of their revenue overnight (as other countries note the protectionism and tariff or simply forbid Apple products from being sold).
So? I don't care about their revenue, I care about the future of American industry. Having a bunch of cash hoarded by old people is irrelevant if it isn't reinvested in something I care about. And I don't care about your supercar or Nobu reservation, or if some fund returns an extra 2%. This is despite being a direct beneficiary.
Live by shareholder return, die by shareholder return; the US is not and shouldn't ever be geared to shareholder return over everything else. Apple and other companies have freeloaded off the US for far too long.
Again, Apple ceases to be much of a company at all if they go your route of being isolationist. It isn't even about the stock price: their revenue tanks, their ability to produce tanks, everything about the company is basically just decimated. You might as well say "I don't care if Apple exists or not". So blow Apple away and how does that help the future of American industry?
Juche doesn't work in North Korea, it isn't going to work in the USA.
It's mostly about cost and market access to China.
Most smartphone supply-chain for Samsung and Apple exist outside China -- primarily in Japan (camera, sensors), South Korea (DRAM/NAND, OLED), and the US (various ICs fabbed at TSMC in Taiwan). There are quite a few reliable estimates/teardowns showing that these three countries account for close to about 90% of iPhone BOM (bill of materials). That's one reason why Samsung's smartphone unit was able to pull out of China without much disruption back in 2019 -- ie, low dependence on China.
I feel that Apple has pushed this misleading narrative a bit too long to defend their massive China outsourcing.
They've actually been diversifying iPhone manufacturing away from China for a few years already. As of April 2024, 14% of all iPhones were already manufactured in India. That's around 30 million phones per year. And Apple plans to double their India manufacturing again by 2028.
Results don't matter as much as PR, this is time when this is unfortunately valid. Just look at US elections.
Measurable results affect rational aspects of our minds, PR attempts to attacks directly emotions bypassing the former, ie to induce impulsive shopping.
Also, what actual security? Apple is as vulnerable as cheap chinese phones against state actors using 0days. Apple devices are still being stolen for spare parts, Apple doesnt secure each component AFAIK and thieves know this (very recent case with friend of a friend, they even knew how to bypass that built in airtag tracking). I haven't seen anything but very well crafted PR statements on this topic. All money-accessing apps on absolutely any phone are a security risk.
> This should be the top comment. Apple are doing business the way business is done, just like last time. Results don’t matter, it’s economic policy via press release. Form over substance.
If the Trump administration has any competence, they will rub those old promises in Apple's face until Cook actually does something meaningful.
The whole Trump administration is all about form over substance, though. I would not expect Trump to do anything actually productive about it, as long as Tim Cook sings his praise (and pays his dues).
I am a free trader in principle. However you have a country (China) with an authoritarian government that makes favored industries subsidized.
Of course the standard economic argument is that China using its GDP to make goods cheaper for our own citizens to purchase is better for us - they are subsidizing our economy. However it ignores the strategic disadvantage by our country losing its manufacturing capabilities.
The graphs may show economic advantage. It’s hard to quantify the long term strategic and militaristic disadvantage to not being able to make anything yourself if a world war occurs.
> However you have a country (China) with an authoritarian government that makes favored industries subsidized.
This is overlooking the forest for one tree. The thing is, mean chinese manufacturing wages are $25k/year (purchasing parity adjusted! $15k unadjusted) for a 49h week.
That is the reason that so much manufacturing/industry has shifted there, not some nebulous "Chinese government subsidies" (not saying those are not a thing, just that they don't really matter all that much).
> It’s hard to quantify the long term strategic and militaristic disadvantage to not being able to make anything yourself if a world war occurs.
Certainly. But forcing low-skill industry to stay at a relevant size in a high-wage country is expensive business (compare agriculture, which is subsidized basically for exactly this reason) and not straightforward (see Jones act).
Presenting tariffs as a viable alternative to taxation is just beyond ridicule, but that has not stopped people so far either...
Salaries are just a small part of the reason industry works in China.
The bigger picture is that China invests in the development of an industrial chain. This has many aspects: infrastructure, education, training, housing, and of course tax incentives. The USA decided to stop investing in practically all of these. Even scientific research, the last area in which the US used to lead, is now in jeopardy from both sides: competition from China and internal cuts.
I'll concede that having a solid baseline of infrastructure, (political) stability and a motivated/educated workforce is necessary, and China did well in building this up.
But I strongly disagree with your conclusion.
Lets assume that the US did all of those perfectly:
- Brilliantly educated workers with the perfect ratio of industry specific knowledge/experience
- Cheap housing near industry hubs built by the state
- The best and cheapest to use ports, roads and railway networks on the planet
- No tax on manufacturing workers income
Some of those are ludicrous/unrealistic for the US to provide.
But even if you managed to do all this-- that still does not make US manufacturing industry competitive. Because those US workers will still want a locally competitive wage instead of <10$/hour.
The reason the US is not competitive is exactly because it doesn't spend the money needed for that. China did it. To be more concrete, if the government spends money to build housing, workers don't need to pay so much to have a home. If the gov spends money on public transportation, workers don't need to buy expensive cars just to get to the job. If it spends money on free health care, then workers don't need to pay for expensive insurance. If the US spent money on (near) free higher education, workers wouldn't need to pay high costs on student loans. These are all items that make the US uncompetitive with other nations.
> The reason the US is not competitive is exactly because it doesn't spend the money needed for that.
You are making some kind of logical jump here that I can not follow. I just listed basically the absolute best that the US could have ever done-- but even in that absolute dream scenario (tax free income for manufactoring workers? I mean in what world do you see something like that ever actually happening?), the US is still not competitive in a direct comparison, because US workers have just no reason to work manufacturing for 10€/hour (when they make ~30/hour right now).
You can stack all the incentives you want-- the gap in wages/standards is so large that apart from straight up paying the difference (in either tariffs or subsidies, and that is a lot of money), you are not going to make US manufacturing competitive in a head-on comparison.
What you forget is that is exactly these inefficiencies that inflate US salaries. You need to make more money just to survive in most US cities, which forces companies to increase salaries.
> What you forget is that is exactly these inefficiencies that inflate US salaries. You need to make more money just to survive in most US cities, which forces companies to increase salaries.
I think you are reversing cause and effect here. Wages are not rising because things are expensive-- things instead become expensive because people are "rich" and can afford them.
I suspect (not an accusation!) that you would intrinsically like to see the US run healthcare and education in a government controlled way, at-cost, instead of allowing excessive private profits there (which I think is a good idea!).
But advocating for such changes in the name of making US manufacturing competitive is dishonest in my opinion, because I absolutely do not see those shrinking the wage-gulf sufficiently for US factories to compete head-on with China.
Furthermore, I don't even think you want to be competitive with China in this regard. Having a significant percentage of Americans working in/for factories to produce simple goods for 10$/hour strikes me as a step back, even if you would bundle this with a bunch of positive progressive improvements.
Absolutely. You do need a minimum baseline for infrastructure, government stability and workforce.
Most of Africa is just starting to slowly get there, Bangladesh is already very relevant for textile production.
I would expect the same basic trend to repeat that we saw with electronics manufacturing in 90s Japan:
First cheap products move (very wage sensitive), then the local sector expands, wages rise with the whole local industry moving up the value chain, then at some point local wages become high enough for the whole process to repeat with the next low-wage country...
I think trying to block this trend off with tariffs is a futile waste of taxpayer money which american consumers are gonna pay for.
Spending tax money to keep some degree of self-sufficiency in critical industries (like with agriculture) can be a solid idea if done sparingly and cleverly, but that is not how the current US admin has approached this...
"The company responded by intensively lobbying the U.S. government to intervene and mounting a misinformation campaign to portray the Guatemalan government as communist.[18] In 1954, the U.S. Central Intelligence Agency armed, funded, and trained a military force that deposed the democratically elected government of Guatemala and installed a pro-business military dictatorship.[19]"
> The graphs may show economic advantage. It’s hard to quantify the long term strategic and militaristic disadvantage to not being able to make anything yourself if a world war occurs.
Is the United States at risk of not being able to make anything ourselves? We have the second largest manufacturing output in the world.
Sour grapes. Most economists were just happy with this situation until recently. What I mean is, the current situation arises by the desire of Western businesses of getting hid of productive investments and concentrating only on capital investments. It has nothing to do with trading with an authoritarian government or not, which almost everyone believed was Ok until recently.
Pricing in externalities (such as national defense impact) is a basic function of economic policy.
I searched 'economics 101 strategic industries' and found this[1] within 30s which includes an overview of 'national self-sufficiency'. It presents the standard argument, including the parts you claim the standard argument ignores.
I personally favor decentralized planning over markets, but I find it unnecessary to slander economics.
Charitably, tariffs exist so POTUS can either lower taxes or increase jobs in US, but both would take time to pan out assuming things go well. So if a company is willing to onshore money or jobs, its achieving its intended purpose in their eyes.
No I think Mexico/Canada were largely about stopping immigrants and fentanyl smuggling. But China was targeted for not doing enough to stop manufacturing of fentanyl precursors.
Sorry to say but that's already been walked back on after Canada committed $1 billion dollars for extra northern border security and it made no difference in the tariffs discussions.
- According to CIS, the number of Canadian crime groups producing synthetic drugs doubled between 2023 and 2024
- There's a lack of Canadian agents who are tasked at preventing this and current legislations make it very inefficient between federal and provincial law agents
- There's an upward trend in Fentanyl seizures in Canada the last 2 years
- Fentanyl is now being produced domestically in Canada
All of that is within the control of Canada with better policies.
Let’s put it into perspective, because those numbers don’t give a baseline for what the problem is. Also they don’t necessarily have anything to do with trafficking.
Last year there were 45 lbs of fentanyl intercepted crossing into the US from Canada. Thats a backpack. There’s 500x as much coming from Mexico.
It’s unrealistic to expect that zero fentanyl will come into the US from Canada, and until that happens we will tariff all trade with them.
Zero fentanyl is a fantasy and will never be achieved. That thing is way too small and can be carried around too easily*
Tariffs are just a lever to get things done on the international front.
Canada has been neglecting security for a long time, so it's a wake-up call, and it's not a bad thing to put this out publicly to change things.
CBSA only inspects about 4% (some sources say even 1%) of all containers that ship in Vancouver. Quebec-Vermont* border control has been a joke for years.
Sure, Trump is using all kind of tricks to put pressure against his trade partners to secure some wins that will solidify him as a change agent to boost America, which will appeal to his electoral base. At the same time, it may actually bring good results to the US economy, bringing major investments in the country and negotiating better trade deals.
You may not agree with the means to get there, but you can't deny there's an argument to be made about his "America first" policies and why it could benefit the average people in the long run.
So therefore that allows the President to go back on a trade agreement he personally signed in his last term? I'm not going to disagree Canada should do more about reducing Fentanyl, nor that Canada can't control it with better policies. I am not clear on why this allows the United States to go back on agreements and allows the President to threaten with tariffs that seem to change weekly.
Yes this is one reason tariffs are so valuable to a corrupt POTUS. They have essentially unilateral and very fine-grained control over them, down to exempting specific companies or products outright.
Congress needs to step up on this, honestly. The entire idea that the President can unilaterally implement trade policy is as plain a violation of separation of powers I can think of, and SCOTUS is a fan of non-delegation doctrine.
Legislators step up when enough of their voting constituencies make it clear that they value something as a non-negotiable (assuming votes still matter).
Which means those who care about this are back to not only contacting legislators but also persuading a lot fellow voters that separation of powers is crucial and worth prioritizing over familiar well-handled and loved heuristics.
Yes a million times. For all the rhetoric about authoritarians, the Democrats never seem to want regin in Executive power when they are they majority. It is like a game of chicken where America winds up with a populist dictator from either the left or right.
Yes because populism is a reaction to government being generally unresponsive to people’s needs.
Congress has become increasingly unproductive and unresponsive. There are many popular policies that Congress essentially ignores, and many problems that go unsolved. So trust in government dwindles and people crave strongman solutions.
I’m not sure there is a solution. There are so many interlocking problems gumming up the process that any “we just need to fix X” solutions (where X is gerrymandering, money in elections, lobbying, the two party system, first past the post, corruption, income inequality, the electoral college, the slow death of journalism, consolidation of industries, etc) are nearly impossible and also probably insufficient because they all feed back into one another: they are both causes and effects.
So when people are mad about a downstream effect like the price of eggs and digging any deeper touches one of the topics above (“to fix egg price gouging you need to reinvent the political system” sounds a lot like “to make an omelette first you need to create the universe”), it’s really easy to throw your hands up.
A codebase accumulates messes during it's entire lifetime. Sometimes, it gets to the point to where a rewrite is a good option. Even if the rewrite doesn't have all of the features...Even if possible, it may be better not to keep all of the features. As some features aren't important enough to maintain. The same can be said about government structures.
That is a fair analogy of the problem. But the solution is inherently more complex.
1. The agents of political change are themselves subject to politics. (Unlike maintainers of a codebase, who are able to make top-down decisions about the code.) If you make a bad decision, it may be the last one you’re allowed to make.
2. There are no unit tests for politics. There is only history, which is an imperfect predictor.
Granted large political systems are uniquely complex. But complex & old software can share the attributes you outlined.
1. Politics does enter into programming teams. There's several articles of this genre. Where the most highly regarded programmer made bad decisions over a long period of time & exhibited hubris in doing so. This programmer was fired, but if he was a founder otherwise held some leverage, then he may not be fire-able. However, other staff can leave. It's more difficult to leave the impact of politics.
Entire industries can make bad decisions due to incentives. Such as I must learn technology X because companies hire for technology X. Companies choose technology X because developers know technology X. Technology X could have fundamentally flaws. Compounding complexity in the ecosystem as additional tech is made to fix the flaws...which will also have flaws.
These cycles of compounding bad strategic decisions yet good tactical decisions can last decades.
2. In some ways there are unit tests for politics. Geopolitical signals can be sent. So if X happens, there's a tacit agreement that Y will be the response. Unit tests are great for small units. However, complex systems can be difficult to comprehensively test. This is why there is fuzzing...which is effectively a monte carlo simulation. Monte Carlo simulations are widely used in political analysis.
3. There's non-deterministic inputs. And the system can be more complex than what's possible to model. Leading effectively to non-determinism in making decisions about how to modify the system.
Clearly a non-sequitur, but I'll bite anyways. Paul Pelosi is a VC, and I'm sure that most of the Pelosi net worth is due to his income, not hers. Since the laws preventing members of Congress from trading on information they receive as part of their duties, you can't say that the Pelosi's have violated any securities laws.
And both sides of the aisle benefit from this. Whether it's legalized insider trading or jumping to corporate jobs when out of office, it's a corrupting influence. All members of Congress, SCOTUS and POTUS should have to place their assets into blind trusts. That won't stop this corrosive influence, but it is the bare minimum.
> Since the laws preventing members of Congress from trading on information they receive as part of their duties, you can't say that the Pelosi's have violated any securities laws
The feds can and do go after people for using family members and friends to execute trades. So if Nancy Pelosi told her husband some material non-public info, and Paul Pelosi traded on it, that would still be insider trading.
There was a guy at Microsoft who was caught once using a friend to place trades. He said he talked himself past his ethical concerns by reasoning that members of Congress do it.
Edit: to be clear, the absence of a prosecution does not mean that the Pelosis did not insider trade. Nor that they did. We can't tell from this distance, only speculate.
"Paul Pelosi, 83, sold 30,000 shares of Google (GOOGL) stock in December 2022, just one month before the tech giant was sued over alleged antitrust violations."
> Since the laws preventing members of Congress from trading on information they receive as part of their duties, you can't say that the Pelosi's have violated any securities laws.
iirc as soon as anything becomes beyond the border the President holds the keys for various reasons including the ever-vague “national security” but also due to being prescribed as the primary negotiator https://en.m.wikipedia.org/wiki/Treaty_Clause
Tariffs are not treaties, they're taxes/duties - only Congress has the power to raise them. Article I is extremely clear on that. Historically, tariffs were always raised by acts of Congress, not by Presidential fiat.
Trump is using extremely misguided legislation from the 1960s/70s where Congress allowed the President to enact tariffs for national security and emergencies. There is a very strong argument (in the sense it resonates with the conservative SCOTUS majority) that Congress cannot delegate its fundamental powers to the executive by legislation alone.
I think people are just too cowardly to bring a case in front of the courts to challenge the constitutionality of it all. Non-delegation doctrine is what the Federal Society want to use to kneecap all federal regulation. Trump operates on a spoils system so it's not in the interest of conservatives or businesses to challenge him, for fear of retribution.
Trump is using tariffs not to raise revenue, but rather use it as a stick to force companies to invest in USA.
Previously they were outsourcing and offshoring as much as they could get away with it. Which led to transfer of advanced technologies outside USA and America losing its manufacturing and technology edge
So how's that going? Outsourcing seems to be going strong, the tarriffs instead pissed off allies who are preparing counter-tarriffs, and the CHIPS Act is being dismantle as we speak (there goes our investment.
Tariffs are a form of taxation. If I want to import say tea, and the government is placing a tariff on that imported tea, I am effectively taxed by the government. And only Congress can impose new taxes.
Not saying you're wrong, but... I have seen claims that tariffs are a source of government income that Congress doesn't control. You're claiming they do.
I haven't seen a citation from either side. Can you substantiate your position?
I have already explained my thinking up this comment chain. I'm mostly replying to GP who misunderstands that the intent of the tariffs is besides the point.
TL;DR read Article I section 8, read up about the Trade Expansion act of 1962 and Trade act of 1974, and "non-delegation doctrine", and you can trivially find legal debate about the constitutionality of IEEPA. Rather than listen to random nerds on HN you should seek out this information yourself.
There is no jobs problem in the US though. Unemployment is at 4% which is mostly just job churn. Long term unemployment is only 1%.
US consumers, that’s all of you, are being hammered with taxes on imported goods most of which can’t realistically be produced in the US anyway, to solve a problem you don’t have.
A commitment like this takes years to plan. It can’t possibly be a response to tariffs announced weeks ago. This is all optics.
I fully believe that the real ("main street not wall street") economy is in worse shape than government numbers on unemployment suggest and both sides are to blame for different aspects of this problem.
But nothing Trump is doing is going to fix your situation.
In fact he (or rather Elon/doge) is very actively making things worse for you with the massive government layoffs, flooding the market with even more people to compete with you for jobs making finding work more difficult and also eventually dropping all of our wages.
>But nothing Trump is doing is going to fix your situation.
I'm aware. I'm sure he's responsible for at least 3 job freezes I ran into mid-interview this year. He's literally costing me job opportunities because no one can budget around this chaotic government.
4% too many and probably understated. The BLS repeatedly underestimated unemployment during the previous administration. Also the labor participation rate, which is harder to game, still hasn't reached pre-Covid levels yet: https://www.bls.gov/charts/employment-situation/civilian-lab...
In February 2020 it was 63.3% and in January 2025 it was 62.6%, for a difference of 0.7%. Also note the steady decline post-2008 and the multi-year plateau that jitters around 63%.
Having the plateau change from ~63% to ~62.5% isn't an unreasonable scenario.
False, USA has a big problem with manufacturing. All US jobs are service jobs to prop up consumer economy, that have no strategic benefit.
A lot of fake employment and low productivity jobs are in the government/NGO sector, paper pushers, DEI jobs, law/compliance type jobs - that should have been manufacturing jobs instead.
USA has no shipyards and infrastructure is crumbling precisely because of misallocation of resources and labor
if shipbuilders were not insulated from competition, they would have offshored their manufacturing long time ago and sold out to Hyundai or some other foreign conglomerate. (just like the rest of manufacturing left USA in 80s-90s)
Jones Act is the reason US has at least some form of domestic shipbuilding
The US domestic civilian shipbuilding doesn't exist. It's dead and already smelling.
For example, the Washington state needs new ferries, and their cost is literally almost 10x than the cost for the similar ferries produced in Turkey for the nearby Vancouver, BC. With this kind of cost gap, there is simply nobody who would buy the US ships unless they _have_ to.
Without the Jones Act, the shipbuilders would have adapted long ago to produce high-margin high-tech components for the ships for the international market. And likely in larger quantities than they do currently.
How is the Jones Act responsible for the failure of domestic ship building? Seems like the Jones Act didn't go far enough if we really cared about a strong domestic ship building industry.
The government determines the employment rate via surveys, i.e. they just go and ask people if they're employed. It's not a calculation from taxes or from employers or anything.
So it's up to the gig workers if they think they're employed or not. Presumably this depends on how often they do it.
Apparently, yes. I saw mention of discussion around the Trump administration potentially giving Apple a tariff waiver. And I believe in Trump’s last term, Apple did have some sort of waiver.
I’m on mobile but Googling for “Apple tariff waiver” and “Apple tariff exemption” will point you to several news items.
Maybe the next administration should keep up the tariffs (as Biden did to a degree). Cheap trade with China distorts the tech sector too. Jobs and Wozniak were the products of a system in which americans had to build products at home. Tim Cook is the product of a system where you can become a trillion company by hyper-optimizing foreign supply chains. Which is better?
You’re incorrect about history. Mercantilism not only restricted foreign trade, but restricted domestic industrial development by requiring the colonies to sell raw materials to Britain and buy finished goods from the Britain. Tariffs were a core pillar of the Lincoln Republican Party.
There’s been an isolationist wing in tech as long as I’ve been in it (early 2000s). I remember chatting with someone at Cisco/Juniper in the late aughts about Huawei ripping off their router designs down to the silk screening. Of course today Huawei makes their own state of the art routers with their own silicon, and some lower-end Cisco/Juniper gear is white boxed foreign equipment. And of course tech folks were complaining about immigration and outsourcing back in the early 2000s when Republicans were enthusiastically supporting both.
Having access to cheap oversea steel allows Americans to focus on building companies with significantly higher value-add. Onshoring low-value industries is a massive human capital waste and an easy way to depress wages.
It allows a very small portion of Americans to build companies with significantly higher value-add.
It destroyed the futures of a larger number of Americans.
Then again, why do we make the distinction "American"? If you have people who became unfathomably wealthy by shipping off strategic industries to the lowest bidder regardless of geopolitical implications, does nationality matter anymore?
No, the analysis (and it’s not exactly rocket science) says just the opposite: Way more downstream manufacturing jobs that rely on steel as input are lost, vs. domestic steel production jobs gained.
Do "normal Americans" pay taxes? From the numbers I've seen, ~1/3 - ~1/2 of tax filers receive more money from the government than they pay. To them, "refund season" is a cause for celebration rather than a stressful event.
The on-average crossover between negative and positive net total federal income (individuals will differ because of individual circumstance beyond just income level) tax when taking into account refundable credits (most notably, but not exclusively, EITC) is a bit below the median personal income but not that far below it, so certainly lots of individual "normal" (by most reasonable definitions) Americans do not pay net federal income tax .
But even if they don't pay net federal income taxes, they probably still pay a net positive amount in a variety of state taxes, federal payroll taxes, and federal consumption taxes (e.g., gas tax.)
Withholding has everything to do with it. Why do you think $10 an hour comes down to 1200 instead of 1600/month?
You can choose to withhold more or less, but the default taxation on w-2's do generally give a bit of a refund. Better to take out too much when you don't need it than slam down a gigantic bill when at once a year later.
"Refund season" is mostly a thing because the default w2 withholdings are set at a level where you slightly overpay on each paycheque, to avoid a surprise tax bill at the end of the year.
The problem with taxes is that it's a prisoner's dilemma. You need global cooperation at some base level of taxes, otherwise companies move to more favorable tax jurisdictions in the long term and offshore from there, which would hurt the US even more. It doesn't have to be all-or-nothing, but any marginal dollar of increased taxes in one place will have some non-zero effect of encouraging the next investment dollar to be spent elsewhere.
To be clear, I do think capital gains taxes are criminally low in the US relative to income tax, so I'm not arguing in _favor_ of lower taxes. I'm just saying why raising taxes isn't a panacea.
Creating an underclass that relies on economic elites paying taxes rather than being economically independent because you want to optimize for "high value add industries" is a terrible long term strategy.
> tax those people appropriately and pump that money back into the economy
So make the US to be like a far less successful country? Kill your economy by increasing taxes? The US economy is singularly successful because it has incentives to build businesses - see YC.
Have you tried living in a country that doesn't encourage businesses? They are often great tourist destinations. I'm in New Zealand and too many ambitious young people leave here: we have an emigration problem because our economy sucks. The government fixes the economy with 30% immigrants (disclaimer: I love immigrants). I have many friends that are never coming back here except for holidays. I hate the New Zealand government incentives for businesses (taxation and regulation) and I can see no way to fix them. Even our "business" political party ACT is completely fucked (latest story - they will be selling everything profitable to overseas "investors" - destroying the economy).
Taxation incentives matter to businesses. Be careful what you ask for because the majority have little understanding and vote for the wrong incentives.
Even business owners don't seem to understand incentive systems that well. Perhaps game designers do?
However I believe that incentives need to be marginal. If you already have a lot perhaps you need a big carrot as your incentive? I don't know any billionaires that I can ask how they feel about taxation incentives: I reckon you are making assumptions about what you think they should feel.
What makes Tim Cook make the US more money?
Taxation cliffs are shit. In New Zealand our Green party decided that 1 million was enough. Why would you bother growing a business after you reached 1 million? Retirement? A business is defined as being about making money (albeit some people do run "businesses" for other outcomes - why is Warren Buffett still working?).
High marginal taxation is also shit IMHO.
The hard part is to design the incentives so that productive people build your economy for the benefit of everybody.
If a government discourages business then the economy is crap and everybody suffers. See other economies.
Few people understand the incentives of others, and few people understand how wealth is created for all: the hoi polloi dismiss the wealthy as vampiric money grubbers. Anyone who uses the word capitalist in a derogatory way has been brainwashed. Most everything that makes our economies work is invisible non-monetary rewards. https://news.ycombinator.com/item?id=43162596
I can speak for my own financial incentives. My perception is that I have an effective tax rate of well over 50% in New Zealand (any retirement savings are not safe because our demographics and governments will screw our economy).
I do not feel the incentive to work in a business - My attitude means I now produce marginally less than I could for the New Zealand economy (I still pay taxes so they are advantaged but they could get a lot lot more from me). I now mostly selfishly concentrate on those closest to me. Why should I work if it isn't marginally beneficial enough for me? I'm no more selfish than my retired friends that I know (a wide variety of people from many walks of life).
(Reëdited to expand and clarify).
We can't decide how much is fair. Compare yourself to a dead king - what is fair? We can design systemic incentives so that we each make the world better for everyone. Not that that it is easy... Trite thoughtless dismissals of the most productive members of society are not helpful.
Edit 2: I guess this discussion is as close to work as it gets for me. Too much adulting. Should I get into politics? Are morals an impediment to helping others? There are too few politicians I admire, and too many I wouldn't want to shake hands with or be associated with. Every idiot has political opinions - how much of an idiot am I? Every politician is smart enough to win an election - they are not stupid yet they make too many horrific mistakes. What about the cryptically smart ones? I see how systems affect people that join a system. What would I become if I join our political system? Understanding our different systems is hard because they grow so weirdly with vestigial complexities due to history, complex interactions, and reflexivity.
So What Does Drive Political Disagreement?
If you’ve read The Psychopolitics Of Trauma, you already know my answer to this: it’s all psychological. People support political positions which make them feel good. On a primary level, this means:
\* Successful people want to hear that they deserve their success.
\* Unsuccessful people want to hear that successful people don’t deserve their success, lied / cheated / nepotismed their way to the top, and are no better than they are.
\* People want to knock down anyone who makes a status claim to be better than them.
People want to feel like their own identity group is heroic net contributors, and that their outgroup are villainous moochers.
People want to feel like their own identity group deserves more power.
* People want to feel like their preferred lifestyle and policies have no negative implications at all and they don’t have to feel guilty about them.
* People want to feel like they’re part of a group of special people poised to change the world, and everyone else is hidebound bigots who resist temporarily but will eventually be forced to recognize their genius.
People want to virtue-signal: demonstrate that they have the good qualities that their ingroup considers most important.
* But people also want to vice-signal: demonstrate their willingness to breezily dismiss the supposedly good qualities that the outgroup considers important.
>Do Medicare and Medicaid exist without businesses?
In a purely technical sense, yes. Because you don't necessarily need an American salary to pay taxes that cover these facilities.
It was very much a concerted effort for most other non-govt Healthcare to be tied to often American jobs. Which of course causes a cacophony of problems when less employers are even offering full time work.
>Why do you look at money as though that is all that matters?
It does not, but business these days sucked up enough money that it's starting to affect basic survival, let alone any pursuit of happiness. There's no point finding upsides when the common person is is so low on the totem pole.
Making hypotheticals of "well look on the bright side, you're not dead" doesn't help either. When America starts using that wealth to make sure no one in a first person country isn't dying on the street, we can discuss the subtleties of capitalism.
>Every poor person I've met avoids taxes.
Well I can't speak for New Zealand. You can't tax a poor person with no income. That's how bad the situation is here.
> Because you don't necessarily need an American salary to pay taxes that cover these facilities.
That is a weirdly employee centric view. I'm talking about the US economy. American salaries depend on American businesses. America has some of the best healthcare available in the world. If US businesses are fucked due to the beliefs of citizens (or whatever else), then the US socialised healthcare is fucked too. There's plenty of poorly run countries to compare against (including Cuba where I discovered their lies about their healthcare first-hand as a tourist). NZ socialised healthcare is okay but our economy is not improving and regardless of our desires for more, the social benefits have no choice but to match our economic output.
> it's starting to affect basic survival, let alone any pursuit of happiness
Only if you're one-eyed. US citizens are the rich. In a fair world we would tax all Americans at 90% and redistribute that to the poor in the rest of the world. Maybe same for NZ too (Wikipedia shows that NZ's disposable median income is ⅔ that of the US however it also strangly says that NZ's median wealth is nearly double that of the US -- I'm guessing because houses are more unaffordable in NZ). Income is usually a better measure within an economy of useful output (economies can't really save for next year). The US federal poverty line is about $16000 for one person - a hell of a lot of money for people in many countries.
> Making hypotheticals of "well look on the bright side, you're not dead" doesn't help either.
I guess you're referring to my comment "Compare yourself to a dead king - what is fair?".
My obfuscated point is that few people (maybe narcissists) would give up their modern life to live in past poverty. Antibiotics, freedom, technology, access to the intellectual output of the world. We are mostly a lot better off than the past. Most people don't value that instead they are money-centric (as many of your comments are). Most people seem to compare themselves to people that are wealthier than themselves and then complain about how they are not getting their fair share. Few people compare themselves against the global poor and then talk about how much they should share their wealth downwards. They talk about how others should share their wealth - they rarely seem to consider how they should share their own wealth. Especially ironic given that it appears that the majority of commenters on HN are the wealthy of the world (and often part of the tech overlords - e.g. YC).
The US is often a parasite upon other countries. If you were to say that the US pays it back to poor countries with technology (mostly from rich companies), then you would be implicitly arguing that wealthy US companies deserve to be wealthy. I recall that weapons are the biggest US export (nice!)
I guess I'm saying is really take care not to kill your geese laying golden eggs (even if you think the geese seem to be keeping too much golden egg to themselves): the socialised good that you have depends on those geese (US businesses). The bad is bad but don't destroy the good.
An economy is a delicate balance - as shown by many failed economies.
> When America starts using that wealth to make sure no one in a first person country isn't dying on the street
I get the value add argument, but lots of people just need income to pay for living expenses. Without an income, those people become disaffected and sometimes violent. Then they embrace right-wing protectionism because, while their gadgets are cheaper, they have no income to buy cheap gadgets.
Nor can they move to these offshore places (where the cost of living is lower) because immigration laws exist in part to control worker mobility.
> Having access to cheap oversea steel allows Americans to focus on building companies with significantly higher value-add. Onshoring low-value industries is a massive human capital waste and an easy way to depress wages.
That's the talking point, but it's bullshit. A lot of those "low-value industries" are fundamental capabilities, and China sure as hell isn't going to let the US own the "higher value-add" areas. They dominate those next, and the US free-trade business elites will be fine with it as long as they get to make some money for themselves.
Being a high value-add area is endogenous to how hard it is for others i.e. China to reproduce. In other words, if it were easy to make GPUs they wouldn't be so damn expensive.
> Being a high value-add area is endogenous to how hard it is for others i.e. China to reproduce. In other words, if it were easy to make GPUs they wouldn't be so damn expensive.
China's going to put the money into making GPUs, and they're going to get it right, probably sooner than later. Then they'll drive the American manufacturers out of business, like they've done in many areas before. Their government isn't beholden to the profit-focused capitalist attitude that is one of the West's biggest vulnerabilities.
Also, it's pretty foolish to 1) forget power and security doesn't come from rarefied high value-add stuff, 2) that there are a lot of people that can't be employed doing stuff like making GPUs.
>> Isn't this why we declared independence in the first place? To get away from the British restricting free trade?
No. I'm not sure where you got that idea. If you look at something like the Boston Tea Party, it wasn't high taxes on tea that were being protested against, it was lowered taxes on tea that undercut the smuggling operations of people like Sam Adams and John Hancock. "No taxation without representation" makes better press than "No undercutting my smuggling operation" though.
In the early years of the US, between 80 and 90 percent of federal revenue came from tariffs. Not exactly free trade.
> In the early years of the US, between 80 and 90 percent of federal revenue came from tariffs...
To be fair, the Federal Budget back then was 2%-ish of GDP. And their political consensus gave the Federal Gov't very few things that it had the power to tax.
Govt spending as a share of GDP is probably a good measure of how involved the government is in the economy. There are arguments that too much government involvement leads to stagnation, which considering many of the economies the US regularly out-grows have higher share of govt spending as a % of GDP has some merit. Its an interesting economic question what level that would be which changes depending on how you approach the problem.
I want to say “that’s not what isolationism means”, but I realize it starts to feel vague just like the word “fascism”, used when convenient but varies wildly in rhetorical meaning… to be more specific is better, I like what George Washington had to say about it in his farewell address because it shows the nuance of the topic across the spectrum, it’s not as simple as isolation good vs bad:
The great rule of conduct for us in regard to foreign nations is, in extending our commercial relations to have with them as little political connection as possible. So far as we have already formed engagements let them be fulfilled with perfect good faith. Here let us stop.
Europe has a set of primary interests which to us have none or a very remote relation. Hence she must be engaged in frequent controversies, the causes of which are essentially foreign to our concerns. Hence, therefore, it must be unwise in us to implicate ourselves by artificial ties in the ordinary vicissitudes of her politics or the ordinary combinations and collisions of her friendships or enmities.
Our detached and distant situation invites and enables us to pursue a different course. If we remain one people, under an efficient government, the period is not far off when we may defy material injury from external annoyance; when we may take such an attitude as will cause the neutrality we may at any time resolve upon to be scrupulously respected; when belligerent nations, under the impossibility of making acquisitions upon us, will not lightly hazard the giving us provocation; when we may choose peace or war, as our interest, guided by justice, shall counsel.
Why forego the advantages of so peculiar a situation? Why quit our own to stand upon foreign ground? Why, by interweaving our destiny with that of any part of Europe, entangle our peace and prosperity in the toils of European ambition, rivalship, interest, humor, or caprice?
It is our true policy to steer clear of permanent alliances with any portion of the foreign world, so far, I mean, as we are now at liberty to do it; for let me not be understood as capable of patronizing infidelity to existing engagements. I hold the maxim no less applicable to public than to private affairs that honesty is always the best policy. I repeat, therefore, let those engagements be observed in their genuine sense. But in my opinion it is unnecessary and would be unwise to extend them.
>when belligerent nations, under the impossibility of making acquisitions upon us, will not lightly hazard the giving us provocation; when we may choose peace or war, as our interest, guided by justice, shall counsel.
Ironic that as a Canadian, the US is moving from the nation that would be guided by Justice into the belligerent nation in this situation.
It also serves as a lesson to us that we should have learned from you and George Washington, and stood on our own first and ensured our own security before cooperating with others. We have a long way to go to get back there now, unfortunately under the position of potentially our closest ally and economic partner being belligerent, untrustworthy and unreliable.
I believe the body of people who are promoting such a narrative (and it’s clearly coordinated) have major conflicts of interests, they want to maintain the status quo for their own enrichment at the expense of any specific nation, because in reality plain old paying attention shows nearly the exact opposite is true in every sense.
They are very good at their propaganda which is exactly how they got in that position but they are not looking out for your interests in the slightest, they just want to “manufacture consent” for the forward march of the global hegemony, that oligarchy… nobody in their right mind, with eyes unclouded by hate, would come to these conclusions naturally.
It’s probably the scariest thing too, and it’s nothing new! Go read Chomsky’s book by the same name “manufacturing consent” and he lays out many examples that were happening in the 70s-80s, and they are following the same playbook today just with Ukraine and Gaza instead of Colombian Jungles and Vietnam.
>I believe the body of people who are promoting such a narrative (and it’s clearly coordinated) have major conflicts of interests,
I'm a Canadian. Operating SOLELY on the actions and statements of your executive branch, not the media's reporting but the wording of the government, their executive orders, their direct public statements. your government is increasingly belligerent, untrustworthy and unreliable.
Belligerant -> Constantly making subtle threads of annexation. Calling the Prime Minister of Canada the "Governor" of Canada. Constantly lying about trade deficits that are surpluses and drug and migrant problems that are actually a bigger problem moving north across the border than south.
Untrustworthy -> After renegotiating NAFTA to USMCA and hailing that as a great agreement, now its a shit agreement and he's putting tarrifs on to get more from Canada under the threat (and likely actuality) of causing economic harm to both our countries.
Unreliable -> You were our biggest and staunchest ally. Electing an administration that is actively hostile to our government and sovereignty means you are no longer reliable as an ally.
> they want to maintain the status quo for their own enrichment at the expense of any specific nation, because in reality plain old paying attention shows nearly the exact opposite is true in every sense.
The current actions of the administration are destroying hundreds of thousands of lives in order to enrich a few thousand at the top. They are alienating ally's, destabilizing peace and almost guaranteeing more conflict and war in the near future.
Chomsky warned of the damage a demagogue supported by a disenfranchised, hurting and angry populace can have on a country. This is happening now, and half the country is burying their head in the sand claiming its propaganda.
>This is total nonsense, nobody thinks they can annex Canada…
Trump defies typical expectations and does things everyone else thinks would never happen. Often because to do so would be very damaging and idiotic, but that never actually stops Trump.
However you do NOT threaten to annex, take over or otherwise threaten the sovereignty of an ally or friend. That is geo-politics 101, so regardless of how serious he is about doing it, the act of threatening it is belligerent and shows he isn't a reliable trading partner or ally.
Outside of annexing, the trade war is also actively hostile on top of being perpetuated on complete lies.
>they just want to “manufacture consent” for the forward march of the global hegemony, that oligarchy… nobody in their right mind, with eyes unclouded by hate, would come to these conclusions naturally.
The oligarchy has taken over. The few that benefit from the "global hegemony" you refer to, which is largely the 'interests of the rich' are now completely in control of the US.
The US administration absolutely is not looking out for the interests of the average Americans. Most of what they are doing directly hurts most or all Americans in one way or another, and the few things they do that help Americans benefit the rich the most.
Now is the time much of the US populace has always claimed it would stand up against tyranny, an oppressive "lord" class and kings. They are watching it and cheering it on instead.
China's economic power is certainly not rooted in their isolationist social policies. They're just as bullish about foreign investment as the US was at the height of the free trade era.
Apple went bankrupt under Jobs and Wozniak and was saved by hyper optimizing foreign supply chain company Microsoft only to rise 10 years later by focusing on hyper optimizing foreign supply.
There was a lot more than that going on and I think you've pretty generally mischaracterized the main problem with the mid-80's era Apple—which had nothing to do with domestic manufacturing and everything to do with not delivering new products that people wanted, at a reasonable price. You can claim overseas manufacturing solved the pricing component of that, but that's not at all clear: other companies were manufacturing in the US at the time and still out-competing Apple.
That timeline isn’t even close to accurate. Apple was doing quite well in the 80s when Jobs and Wozniak were there. In the early/mid 90s was when they started going downhill (well after Jobs was gone), and by that time they had already outsourced a lot of their manufacturing (computers in Cork, Ireland and Singapore, and motherboards/components in places like Taiwan).
I don't know about Microsoft, but I'm very clear that the "miracle" operated by Apple was exactly to perfect foreign supply chain at a time when Intel/Dell/HP and others were still heavily focused on the US. The quality of Apple products was already there since the beginning, but they had no way to compete with the PC market until they figured out Asian supply chains.
Anyone keeping count of how many trillions in hypothetical investments and millions of jobs large American corporations have promised in the next 3-5 years?
* What type of jobs? - "The 20,000 additional jobs, Apple said, will focus on research and development, silicon engineering and AI."
* Does the US have the required people, in terms of numbers and skills? - "The company is opening up what it calls a manufacturing academy in Detroit, where it will help smaller companies with manufacturing. It already operates an academy for app developers in the city. It’s also doubling its manufacturing fund in the US to $10 billion." - Sounds like they are upskilling, and will count the employees of companies joining the academy as "jobs created"
* Does this mean moving to US based fabs for the M-series chips? - "[M-Series] chips themselves, however, continue to be produced in Taiwan.
* Is this actually profitable, or is this just a political move? - Define profitable. It is cheaper than paying tariffs.
I might be reading it wrong, but that's the 20,000 ADDITIONAL jobs, which is going to be R&D, engineering and "AI".
Those 20,000 people won't be staffing the production lines. So how many manufacturing jobs, especially low skill, entry level with decent pay, will this create? The whole thing is framed in a way that makes it sound like Apple is creating thousands of manufacturing jobs.
> Does this mean moving to US based fabs for the M-series chips?
This is not really a practical option. A big part of the M-series success is TSMC's lead in cutting edge process nodes. And Taiwan does not allow export of technology for the latest nodes. It is available only there.
Developed, sure. Successfully integrated and commercialized, no. Organizations in US and Europe has done a lot of the prerequisite tooling and research. But they haven't successfully integrated it into an operation capable of producing leading nodes, yet.
> Does the US have the required people, in terms of numbers and skills?
For 30 years, IT managers at blue chip US corporations have exploited the H1-B visa program by saying, "No," and then hiring a never-ending stream of barely-capable Java coders from programmer mills in India, take 5 times longer to make an app than it should have taken, get promoted, and leave everyone holding the bag with shitty web app that we all hate because it's too slow, too bloated, and doesn't work like it needs to. And the companies who can't get enough of that bullshit in-house just hire it out to sub-sub-contractors that do the same thing. Can we not invest in our native population and education systems this time around? I'm so tired of the fact that 90% of the IT staff in my Fortune 250 is Indian, and I know people who would be better at their jobs living in my home town. It hurts our community and our country, in the long run, and by the VERY same logic as re-homing our chip production.
It sounds like you should be directing more of your anger to the C-suite than the people they’re hiring. If they couldn’t get even cheaper Indian immigrants you’d be complaining about code boot camp hires instead - what you need is a tech union which would give you the ability to push back against short-sighted decisions which make your life worse cleaning up messes.
Well, those Indians living in the US will have families of their own, and over time become part of the community you claim to be a part of. Very much like your ancestors did, except they likely didn't face the arbitrary constraints on immigration that Indians (and any other nationality) face today.
The same thing that happened in the UK will happen in America.
People in the UK who are against immigration are often talking about Poles who moved to the UK after the EU and not Indian families who have lived in those neighborhoods for generations.
The crazy thing is, it's not that long ago that Irish and Italian immigrants were not discriminated against. They didn't even consider Italian immigrants to be white.
This need to bend the argument back to the initial English colonization of America is stupid. These mediocre Indian IT drones are not putting everything they own in a boat and washing up here hoping to find a better life. They're the rich B students that could afford the process which become part of an idealized system that American corporations are now bending and exploiting to hire what are essentially indentured servants from a population of people who couldn't get the best jobs in their native country, so they settled on this backup plan.
And they DO have families of their own here (and bring over their in-laws), and a lot of them don't integrate well, for a variety of reasons. At least a third of my neighborhood is Indian. They glare at me on the sidewalk when I wave. And most of them remain inured in their caste system, and are difficult and unpleasant to work with.
Again, all the same arguments about developing our own chips domestically -- which I doubt many people have a problem with -- apply to developing our own, better education pipeline to fully develop domestic software engineers.
It will actually happen because it’s nothing new. The 500b is almost all wages for existing US-based employees. They are looking for a carve out from the new China tariffs (same as last time). Note - they made a very similar announcement 4 years ago https://www.apple.com/newsroom/2021/04/apple-commits-430-bil...
There's a lot of understandable skepticism about this announcement because there are so many PR announcements. I want to temper that with an alternative perspective.
I actually think Apple is a potential dark horse when it comes to AI hardware. What we have now is essentially a three-layered monopoly: ASML, TSMC, NVidia. This has been incredibly lucrative for NVidia. But, as we know, Apple doesn't like to rely on third-party hardware. They've invested heavily in ARM going back to buying PA Semi [1]. Apple replaced Intel chips (which originally replaced Power chips) with the M series in recent years. Apple is in the process of replacing Qualcomm modems in iPhones, which is not only a technical feat but a legal one given Qualcomm's patent dominance over 4G/5G.
Apple has the resolve for long-term initiatives that few other companies have. Apple Pay continues to chip away and get slowly better in a way that, if it were a Google product, would've been cancelled, rebranded, relaunched probably 3-4 times by now (Google Checkout, Google Wallet, Google Pay, Android Pay, etc).
Apple clearly sees AI as a strategic issue. They have loads of cash on hand to finance basically anything they want. And they won't want to be beholden to NVidia.
I expect Apple to have a significant impact here but it won't be tomorrow or even this year. It'll be over the next 5-10 years.
Apple's incentives have definitely aligned with replacing Nvidia entirely ever since they ceased diplomatic relations. But Nvidia also knows this, which is why they invest heavily in things Apple will never do. They write the official Linux drivers Apple wouldn't get caught dead supporting. They give users and integrators freedom to choose their OS, software and library stacks to better suit their application. They sell individual GPUs and unlocked edge compute hardware with no distribution terms or $99/year "developer license" bullshit. Nvidia is a hardware company in places where Apple tries shipping services instead.
Then there's also the software issues. Nvidia has invested in GPU-based compute nonstop for the past 10+ years. Apple invested in Nvidia, then invested in OpenCL after abandoning Nvidia, then abandoned OpenCL for Metal compute which would eventually become the proprietary Accelerate framework. Nvidia's eggs are all organized in one, valuable basket. Apple's investments are spread out all over the place, with much of the time and money going into projects that don't even exist anymore.
Apple has the TSMC advantage, but that's just about it. Their GPU designs aren't comparably efficient or compute-oriented to what Nvidia ships today. Additionally, Nvidia will continue investing in places that Apple principally refuses to support. Unless a serious tide change occurs at Apple, they aren't going to get a fair competition with Nvidia.
Nvidia has proven the space is incredibly lucrative and Apple is best equipped for high end chip designs. Remember 10 years ago it was unthinkable for an ARM chip to compete with x86.
First Apple has to prove they have competitive designs. Apple Silicon GPUs simply do not compete with the efficiency of Nvidia's GPU compute architecture: https://browser.geekbench.com/opencl-benchmarks
Apple's obsessive focus with raster efficiency really shot their GPU designs in the foot. It will be interesting to see if they adopt Nvidia-style designs or spend more time trying to force NPU hardware to work.
I think performance per watt is way in Apple's favor, but raw performance is not.
That said, an M4 Ultra (extrapolating from Max and Pro) would likely compete with my 3090, and with 192GB of memory (for 10x the amount it should cost) will out perform my 3x3090 AI server. And honestly, cost less than my 3 3090s + rest of the computer + electricity.
It won't outperform a bunch of A/H 100s (or even a single one, or any other cards in the enterprise realm) though, but it will cost an order of magnitude less than a single card.
Careful when comparing performance and efficiency. As a rough factor power increases quadratically as you increase clocks on a design, so you can quite easily make a high performance design low power by under-clocking it. The same is not true for the reverse.
Sorry, I was coming at this from the consumer side (since apple is a consumer product company). The majority of LLM use (by consumers) is in inference, not training, so I'd hazard to guess, the majority of people would rather have inference machines than training machine.
That's not necessarily news, unless I am missing something. Craig made an indirect mention of this during last year's WWDC regarding the private cloud compute.
The US is a great place to have your headquarters and a terrible place to have your not-so-cheap labor.
Their actions will drive prices higher and, indirectly, wage higher. Businesses without a war chest will not be able to keep going and fold, the labor market will collapse.
The rest of the world will trade amongst each other and I suspect to save themselves, some big tech companies will relocate their headquarters.
I think you’re overlooking that the robots require maintenance and facilities which are costlier due to labour costs which will ultimately be passed on to customers.
The US market demand is already depressed. Prices go up, demand will fall further.
Try to impose those prices elsewhere in the world and people will move off of apple products. Apple profits will fall, it will lead to a negative feedback loop.
That could also lead to a depression. I haven't heard a lot of politicians here (Andrew Yang in 2020? does he even count as "a politician"?) with good plans for what to do when automation hits jobs even harder.
“We grew from 10 customers to 100 customers in a year. At this rate we will have 20% of the world’s population in a decade!!!”
The first cohort of customers of any company is always the easiest to obtain with the lowest acquisition cost. You solve the easiest problems first.
This is Cohort Analysis 101. Not to mention Waymo still hasn’t shown to be able to operate in less than ideal weather conditions or proven that the unit economics will make sense or be economical especially taking into account maintenance, or utilization ratios.
It’s been operating safely in each market they’re in. The AI keeps getting better. They have no competition (please don’t bother mentioning Tesla vapor ware). Path to high growth seems pretty sure at this point.
And the markets they are in are low hanging fruit with good weather. I’m not saying Waymo is less safe than human drivers. I am saying that it will only take one fatal accident by any self driving car for people to lose confidence, investigations to start, rollouts to be paused etc. I’m also not saying that is a logical response.
> I am saying that it will only take one fatal accident by any self driving car for people to lose confidence, investigations to start, rollouts to be paused etc.
Uber and Cruise are both great examples of this, but it seems like the effect is mostly localized to the company itself that has the issue.
Uber hit and killed a jaywalking pedestrian, resulting in their self driving tech being sold to Aurora. [0]
Cruise hit a pedestrian that was flung into the cars path that a human driver hit previously. This resulted in GM completely abandoning Cruise and their future seems foggy at best. [1]
Which is very very few markets, and all of them share weather patterns that are very similar.
When Waymo can demonstrate reliably going from Chicago to Ann Harbor in the middle of a snow storm thats when we can start talking about how its good enough.
Sure wouldn’t, and for what it’s worth that’s why the scenario is a great litmus test. If it can do that, it should be able to handle anything else thrown at it
Google also had literally hundreds of failures and Android is not an amazing financial success by any means and Google still ends up paying Apple over $20 billion a year because people with money buy iPhones.
Google is not exactly known for its success rate getting products out of the door that aren’t ad related.
In the phone market. The Motorola acquisition was a major failure and Pixels aren’t taking the world by storm.
The entire “Other bets” haven’t led to any major successes.
There are only two tech companies that have shown any ability to do hardware at scale as mass consumer products in the last 25 years - Apple and Tesla.
This is all intended policy to benefit Trumps super donors. They can then scoop up marketshare and competition for pennies, then lobby to get the tariffs lowered or removed, but the higher prices - that we will be used to paying at the point this all comes together - will not go down.
All else being equal, companies are going to use the source of labor that results in the cheapest product they can produce. No one is forcing companies to move this kind of manufacturing to the United States. A 10% (let me reiterate that: TEN PERCENT) tariff on incoming goods is inflationary, but by very little, and quickly absorbed by companies and consumers. No one is moving their labor supply from China to the US to avoid a ten percent tariff; US labor is more expensive than that, and there are fifty other places around the planet you could find cheap low-skill labor that aren't on Trump's shitlist.
But you won't believe any of that, because you want all this to happen. You're a doomer; doomers and preppers secretly want the doom they predict to happen, even if they won't admit it to themselves.
Your "logic" (masking) conveniently avoids the point I raised: That these tariffs are being enforced to the tune of 10%. That isn't enough to alone justify this level of investment, or relocation of significant production capacity. Obviously, Apple agrees with this, because the investments they're making aren't as far as I can tell a relocation of capacity from China to the United States, but rather greenfield investment in high-skill research and development. Apple has also made significant investment into advanced silicon manufacturing in the United States; something they also did not rely on China for previously.
> The world is a bigger market than USA and just about every other country has cheaper labor and no tariffs.
Have you done zero research into this? The EU imposes a tariff on Chinese EVs. India imposes insane tariffs on all imported electronics. China tariffs Australian wine. Russia tariffs agricultural products from the EU. Brazil tariffs all imported automobiles. The list goes on. Tariffs are everywhere, everyone uses them for something.
Until things are actually built, I take press releases like these with a grain of salt. Similar to the stories about Mark Zuckerberg removing tampons from men's washrooms the week before the Presidential inauguration, I believe that a lot of these stories are intended for an audience of one.
It probably won't lead to anything without a weaker dollar to support US exports. First you need tariffs, then you need massive investment in reshoring, then you need a weak dollar so those new factories can profitably export.
It won’t happen. The supply chain is far too complex. Not to mention that the labor market in the US is not willing to do a lot of the work that you see in China and isn’t large enough even if there were enough willing people.
And then you have the rare earth minerals that aren’t available here.
Deepseek seems to prove there's no super secret sauce that makes these models irreproducible outside the US and that the companies here are suffering a bit from the glut of cash/credits leading them to burn tons of extra processing power that could have been optimized away.
A nice model, does not a billion dollar company make. The hard part of AI is not the model; Apple needs people to do the 80% rest-of-the-work; how do you make AI useful to the average person? How can we get inference on edge devices as cheap and efficient as possible? Models are boring. Everyone fully expects that we'll see an N% increase in intelligence every six months now. Yawn. The exciting thing now is: What are we using AI for?
Spending a couple billion dollars also doesn't make an actual billion dollar company. It's yet to be proven that all this spend on LLM training and running can actually get translated into an actual profit.
The US AI industry does have easier access to them but again Deepseek proves countries outside the US can get enough access to them to produce similarly powerful models. Unless the US really clamps down on exports of cards, which will be hard given they're made overseas and only designed in the US, it's not like they're a unique resource only US companies can access.
1: Take over Taiwan to get TSML under their control
2: Find a way to make a deal with ASML for the needed lithography machines
3: Somehow aquire the knowhow that Nvidia has
Taiwan would require a conflict with the US. ASML is a dutch company but seems to be somehow under US control. I have not yet figured out the exact setup there. And Nvidia is a US company.
They are trying, and will eventually catch up, the same way they have in software and in hardware in many other spaces. Maybe it will take another 10 years. Maybe another 10 months.
This idea that EUVL is—and always will be—outside of the reach of China is, frankly, silly. It's a silly strategy to maintain dominance. They will straight up steal the technology if needed.
We should stop pretending like we can roadblock the technological development of the largest country. It's just going to make the fall that much harder. Once they do attain the ability to manufactuer EUVL domestically, capital is going to flood out of US tech stocks like no tomorrow.
You haven't seen anything that hints at it, because most news about what happens in China doesn't leave China. The propoganda machine is hard at work reminding you that China is terrible and eveyrone in China is poor. You'll learn about China catching up only after it's already happened.
That’s just a dumb red herring strawman take on news about China. You can be perfectly informed about what is happening in China, you can also be misled by their propaganda as well. China will eventually catch up in the two areas it is lacking (performant and economical semiconductors and jet turbines) since they are throwing billions at it, but they won’t pass the west overnight (it will still take a few years, maybe a decade).
With the combination of very real geopolitical risk (which was a topic of considerable discussion at a tech conference I attended late last fall) and the current political climate, there's a significant mindset that the US should be pulling back a lot of things to its own borders where practical even if not optimal at the moment.
Tesla makes a lot of promises that it can’t keep and losing money and market share globally.
I doubt Waymo is going to be a big deal in much of the US over the next decade. Even if they do figure out all of the technical issues. People will accept hundreds dying from car crashes. But not one dying from autonomous cars.
Does this mean competing with Asahi to run a Linux kernel, or will this be an attempt to run AI workloads on XNU?
Consider the cost of GPUs, losing what could be double digit percents on overhead might not make this very competitive. The macOS microkernel can still beat NT in some situations (like not having filters slowing filesystem down to a crawl), but it lags significantly behind the investment in Linux performance over the years by every other major company.
I believe the intention is to use their own M-Series CPUs - to get what they call "Private Cloud Compute". The cpu on your phone will encrypt data and a request, send it over the network to am M-series CPU which will decrypt and process/send back an encrypted response.
The idea being there's no VMware, kernel or piece of hardware that can have backdoors built into unless someone files off the top of the chip and somehow probes the silicon
> Apple said that it, together with Foxconn Technology Group, will later this year begin producing the servers that power the cloud component of Apple Intelligence — a system called Private Cloud Compute — in Houston. That marks a relocation, at least for some production, from overseas. Next year, it says a 250,000-square-foot facility for such manufacturing will open in the city.
> The Private Cloud Compute servers use advanced M-series chips already found in the company’s Mac computers. Those chips themselves, however, continue to be produced in Taiwan.
I don't think it's AI servers for Apple silicon. I think it's just regular x86 Linux servers to power Apple's AI cloud services. It's a commitment to internal investment rather than a product.
Apple doesn't like B2B and Steve Jobs was very vocal on this (there are various videos where he explains why). Ever since they can afford to, they reduced their B2B to the bare minimum needed. So don't expect anything server-like from Apple.
I'm not sure. Because the moment you enter the system room, the ecosystem is a completely different universe.
You can't easily sell "Good / Better / Best" version of a single model, and tell "These are the options, take it or leave it". Servers are customized to the screws they come with and are expanded throughout over the years. So, the logistics are somewhat different for these kinds of devices.
Plus, macOS is not a CLI first operating system for server operations, and macOS Server is not updated for some years. Allowing Linux would be a different offering, and allowing macOS to work with all kinds of hardware from ordinary Ethernet to 100G+ Ethernet and 400gbps Infiniband (plus all the other interconnects) will be a fun exercise in testing flexibility of both macOS and Apple development teams.
So it's quite complicated. All servers are built to order SKUs. Dell keeps configurations "per server" in their databases, for example. If you have a Dell server, enter its service tag to support site, and you'll get the configuration of the device as it left the factory.
They don't have to go all the way though and fully compete with Dell EMC/HPE. I'm not sure what the original commenter was thinking but in my mind they could simply sell a Mac Studio variant with dual PSUs, better networking, a rackmount chassis, etc. Basically have their existing consumer machine placed in a more datacenter friendly factor.
I mean places like Github and AWS are painfully racking up Mac Minis for their deployments and this theoretical server model would simplify everything. It also becomes an option for on premises AI inference using MLX, especially if they manage to get ANE support working in conjunction with the GPU for faster prefill.
The support and software stack for the server model would be the exact same as the consumer variant and they certainly wouldn't have special Linux offerings, Infiniband, and all that. If there's networking beyond their existing 10G it's going to be built into the board and they aren't going to support random 3rd party cards. The unit also doesn't need to be upgradable either.
Thanks to ireland all big US corporations saved hundreds of billions of dollars past few years so now they can get back to US with this massive cash for anything they want (ofc. nothing will get back to EU as long as they ignore tax heavens)
"Four years ago, a few months after President Joseph R. Biden Jr.’s inauguration, Apple announced an “acceleration” of its U.S. investments, pledging to spend $430 billion and add 20,000 jobs over five years. In January 2018, during Mr. Trump’s first term, the company said that its “direct contribution to the U.S. economy” would be $350 billion over five years and that it planned to create 20,000 jobs over that period.
Apple did not immediately respond to a request for comment."
Is this just lip service? What happened to those previous investments?
This happens all the time with large investments in $X that make for impressive sounding press releases. If you (can) dig into the details, invariably a lot of the money is in previously committed/spent allocations in a whole bunch of different buckets (or, per bombcar's sibling comment, money that may never be spent at all).
This is an insane amount of cash to throw at any problem. They could have Apple rockets mining asteroids with $500,000,000,000. There is no way all this cash goes into AI. What will actually happen is they will take 1/10th this cash to an over-valued startup and acquire them.
Between land, hardware, having to build multiple power plants, the labor costs involved in all of this, and setting money aside to run all of those for however long, then yeah I can see where it gets up to that price. 20k engineers is easily 5 billion per year in salaries, probably more.
Assume the human resources at $100,000 per head, and you get $2B/yr. Four years comes to just $8B for human resources. Assume land costs $10B. Assume construction costs $100B.
There's a lot of noise I can see behind the scenes on investor confidence. Noise as in "everything is fucked" sort of level of noise. Thus I expect this is being said to try and stop the AAPL stock collapsing in the upcoming recession that the analysts are predicting more than a tangible expansion and recruitment goal.
I also take issue with their being 20,000 people on the market who are still able to contribute something useful. They will be culled quickly and quietly down the line in the annual corporate lay offs.
It is not the time to make grand gestures unless you're trying to gain political favour, at which point any respect I have at least is gone.
Do we think Apple will once again sell servers to customers?
I guess they could sell servers to customers who want to run the latest Apple Intelligence models on-prem, even though that probably wouldn't make much of a difference, since you probably still have to trust Apple.
> Apple said that it, together with Foxconn Technology Group, will later this year begin producing the servers that power the cloud component of Apple Intelligence — a system called Private Cloud Compute — in Houston. That marks a relocation, at least for some production, from overseas. Next year, it says a 250,000-square-foot facility for such manufacturing will open in the city.
[...]
> Apple will also expand data center capacity in Arizona, Oregon, Iowa, Nevada and North Carolina, all states with existing Apple capacity. The company confirmed that mass production of chips started at a Taiwan Semiconductor Manufacturing Co. facility in Arizona last month. Bloomberg News recently reported that plant is building chips for some Apple Watches and iPads.
I think they might build a cloud offering. Something like Cloudflare workers but AI centric, perhaps running Swift on the Apple equivalent of V8 isolates.
Makes sense from a business perspective - there's significant growth potential for them as their presence in web tech is approximately nil.
It would make sense for Apple to fork their next highest-end Mac Studio motherboard, make relatively minor changes to it (e.g. add a higher bandwidth NIC and strip out unnecessary I/O) then wrap multiples of those into a rack mount chassis, with commodity-grade cooling and power supply solutions appropriate for the context.
Combined with a properly headless fork of their OS stack (think Darwin, not OS X Server) they could spin up a highly competitive solution using entirely "B-team" resources.
...then it would be piped through their design-council, run through 5 more iterations to get a unique unibody case for it, accompanied by an optional proprietary Apple rack and a price-tag triple of the competition.
That's along the lines of how it usually rolls whenever Apple tries to make something purely utilitarian, it's the most considerate and "fresh" look at a product, but ultimately designed to be used and then disposed when finished.
A purely utilitarian IT-appliance without a individual end-user doesn't seem to be possible in their product pipeline, you usually end up with something "Prosumer": Impressive on its own, yet of degraded maintainability and scalability.
It's like asking Bugatti to design a public transport bus. It would surely be an impressive bus, but not one you would want to maintain over years at a scale of hundreds.
I don't get why folks keep saying x86 linux servers here for AI, if anything it'd be M series arm based servers, either running linux or macos. Realistically I'd imagine a set of scaled up mac mini arm servers for running inference or fine tuning on them as more likely being the "ai servers" than x86 based anything. Power is the key thing that they'll be optimizing for, and that's where ARM shines.
ARM is not a GPU architecture, nor is Apple Silicon the most power efficient GPU/CPU available to datacenters. The factor Apple is optimizing for is their own profit - they're terrified by the notion that another company might dictate their software margins for them.
Don't they need gpus (for training)? Apple already did a footshoot wrt gpus in the apple ecosystem. unless they have some sort of apple-internal ai chips ready to train models.
Apple's Private Cloud Compute is on racks of M4 chips which have NPUs and GPUs on-die and unified memory access to however much RAM they want to put on them. All of a sudden they're competitive with NVIDIA, but they don't let anyone else use that platform.
Apple has no interconnect technology comparable to what Nvidia ships to datacenters. The larger Nvidia clusters measure their addressable memory in terabytes, the value of Unified Memory at that scale is practically negligible (if not wasted bandwidth).
You're making some pretty handwavy generalizations here without a solid grasp on why Nvidia dominates GPGPU compute.
> Four years ago, a few months after Mr. Biden’s inauguration, Apple announced an “acceleration” of its U.S. investments, pledging to spend $430 billion and add 20,000 jobs over five years. In January 2018, during Mr. Trump’s first term, the company said that its “direct contribution to the U.S. economy” would be $350 billion over five years and that it planned to create 20,000 jobs over that period
Anyways, the land (obviously in Texas) is already purchased and has been sitting empty. The unbuilt factory keeps getting more expensive though.
Oh god. Hopefully they’re lighter than the old xserves. We had one still running up until a few years ago when we finally removed it. You could break a toe if it dropped while pulling it out of the rack. People are still selling them on eBay.
I suspect that was the RAID drive bay (ridiculous item). That was a 3-4U monster.
The Xserve, itself, was a 1U unit that was pretty much the same (or lighter than) any other 1U server (we also had HP and Dell servers that were heavy). The weight distribution could be weird.
That stupid drive bay was a proprietary nightmare. The disks cost a fortune.
However, from what I can see, this will be for "internal-use" servers. I don't think they will be selling iron; just services run on the iron.
I don't think it's AI servers for Apple to sell. I think it's just regular x86 Linux servers to power Apple's AI cloud services. It's a commitment to internal investment rather than a product.
They have hiring positions for running a Darwin-based server OS, and their private cloud compute is on Apple Silicon. I doubt it's going to be swaths of x86.
I'm curious what they will look like, given that these are not for anyone else to buy. Maybe Apple made a different form factor/configuration that suits their datacenters better?
expect of course they won’t… few months into every new administration (see 2017, 2021…) they’ll make a splash announcement like this… and then wait for the next administration to make it again
A strong majority of Apple's revenue comes from outside the US. If Apple is tied to US protectionism, it is precisely the sort of org that suffers.
Now to be real, Apple has been announcing these multiple hundred billion dollar investments every four years like clockwork. They did it early in Trump's first term, they did it again early in Biden's term, now they're doing it again. But for all of the "Yeah, this is what tariffs are for! Hoo ya!" rhetoric, note that companies like Apple have far, far more to lose than what America "gains" by acts like this.
No, I don't. The Americas in totality account for 42.3% of Apple's revenue. But of course "The Americas", despite the confusion of many Americans, is not only the USA, so while that is already a minority, South America is 14% of that 42.3%, then subtract Mexico and Canada...
The high revenue generated from the Americas is largely due to Apple’s strong performance in their home market, the United States. Apple has the largest market share among smartphone vendors in the U.S. by a large margin. Although international sales have a growing share of Apple’s total revenue, the U.S. still counts for around 42 percent of Apple’s net sales.
Actually look at the data. That is a bullshit, erroneous summary by some lazy clown.
The Americas in totality account for 42.3% of Apple sales.
I mean...this is all so silly regardless. Even if the US accounted for 100% of sales in the Americas (protip - it doesn't remotely), that would STILL mean the majority is outside of the US. But in reality in US is only a fraction of Americas sales, and it's a strong majority sold elsewhere.
Texas is big tech's choice to skirt employee protections. I'm sure these are the type of jobs, similar to Foxconn per the article, that Americans are looking forward to.
I know a lot of tech workers in Texas, specifically in the Austin area. They seem to be doing very well. I'm quite proud of America's working conditions. A lot of workers in other countries would marvel at our opportunities and be grateful these investments are happening here as opposed to elsewhere.
California's housing crisis is a result of small-c conservatives wanting their property values to rise forever. Prop 13 and it's consequences have been a disaster for the state.
Over the course of a decade go from a max of 2%/year to something like 10%/year. Maybe some extra tax breaks for those who lost equity when selling a long-term residence (doubt, probably just slow their growth)
The opposite damage I'm talking about is forcing people to lose their houses for inability to pay current-FMV property taxes. Mostly older/retired/fixed income people. This would apply to any phase-in schedule too.
While that’s certainly a key component, Texas is also home to the largest potential solar and wind capacity in the country. There’s also a ton of land to build on.
Unlikely. Texas is the largest exporter of crude oil and natural gas, the largest in capacity for refining petroleum, and huge exporter of petrochemicals in the US. More solar and wind means more oil for refinement and export.
I do find it slightly offensive that you would insinuate that hiring in Texas is solely about less worker protections and not that we have plenty of skilled workers in one place and with a lower cost of living.
I'm a worker not an owner,and I prefer living in Texas at this stage in my life and have turned down offers to move back to California.
Working in tech in a big Texas city easily puts you in the top 10% of the cushiest jobs in the US. Based on how I've been treated here, I really doubt that worker protections (or lack thereof) is the real driving force behind more tech jobs moving here. We are far from being oppressed here.
The most likely driving force of tech moving to Texas is that mid career professionals like myself don't see a future in California due to the insane cost and bad vibe of raising a family there. It's a great place for people just out of college, but Texas is a better place to settle down unless you are pulling an outrageous salary. The other big advantage California has is VC and startup networks being located there, which is also something that primarily benefits early career people rather than those of us that need a stable job at an established company.
It's also worth pointing out that Texas has long had a large technology industry presence. The dominance that California experienced during the early 2000s through to mid 2010s is an outlier and it shouldnt surprise anybody that things are evening out.
IMO we've really got to start pressuring our own governments to take control of their networks, as well as the companies the population is going through (not just Facebook, but even international contractors for services). Letting a foreign government have this much control over the data of your populace and the ability to feed whatever algorithmic message they like is a path bound for disaster in the long run. The powers of the world are way too consolidated as is, and a company can turn into a state actor at the drop of a hat. I don't think we can maintain kayfabe about the country/corporate divide. I also think this can be done without impacting freedom of speech for your population, as long as you don't consider corporations people.
Most countries don't have the resources to do much, but even then they can try their hardest not to be beholden to any single foreign country coughChina, Americacough.
Reverse technology transfer from countries like China is kind of fair. But EU companies should be very wary of Trump tariff blackmail that forces them to build production lines in the US.
Not a lot of them, but Trump accuse EU of tariffs because it has VAT. That’s not the same thing because VAT apply to domestic companies as well so there’s no real unfair advantage.
You could make the claim that it’s still worse for foreign goods as VAT can be offset with VAT from supplier purchases, but that only works if you produce goods in the same country as you sell them. But US companies have even larger advantage already as they don’t have any VAT, so in the end the playing field is kind of level.
This doesn’t stop Trump from pressuring and squeezing and making false claims to justify his standing point.
Expect an unstable trade environment due to tariffs and retaliation. This will hurt small businesses that have even more complex environment to operate in.
>But EU companies should be very wary of Trump tariff blackmail that forces them to build production lines in the US.
That's the whole point of tariffs, to encourage domestic production.
Put another way, what is the difference between what you wrote and
>But US companies should be very wary of EU tariff blackmail that forces them to build production lines in the EU.
?
Whether a particular tariff is economically viable is a reasonable debate. Calling Trump's tariffs "blackmail" without assigning that epithet to all tariffs from whatever source is not.
Some part of me thinks they are billing an overbudget here to report that they actually didn't need to spend that much so that they can beat lower guidance (same play for MS, Google, Meta). We've heard that actually training these models doesn't cost even a billion dollars.
Funny I turned Siri off because i didn't want apple intelligence running amok. The follow-on problem --> lack of Siri killed my Carplay because Siri is required (also use itf for setting alarms/timers). The kicker? I can't seem to turn Siri back on after look through all the menus.
I.e. My preference for apple CarPlay supersedes my concerns on GPT running over my contents. Though the UI/UX has made it next to impossible to turn it back on.
Apple Intelligence and Siri are still separate (though Apple like to make it look like they are fundamentally intertwined). You can turn Apple Intelligence off and leave Siri on for CarPlay.
How did you turn Siri off in the first place? That's where I'd start...
The part of Siri that causes the most trouble is the speech recognition - which uses a voice recognition model that we now colloquially refer to as "AI." The part that works reliably, the part that sets your alarm or sends the message, is an action that's hardcoded.
IMO, moving towards AI just leads to increased uncertainty and undesirable outcomes, which is something several journalists reviewing Apple Intelligence have attested to.
If your phone is new enough for Apple Intelligence, Siri is now under that umbrella. There's no "just Siri" option anymore, unless you're rockin an iPhone 14 or older.
When I put in timers -- for some reason my timer frequently/randomly just sets to 79 hours and a random assortment of minutes and seconds. I have no clue why. I always have to double check otherwise I might be waiting awhile.
It feels like it was a residual timer or something but I have never set anything like that - it is quite strange.
I don't know if this is an actual problem you have, but since Siri appears to be composed of independent voice-to-text and text-to-action systems, you can say "start a one three minute timer".
The problem is AI current best use case is creative work, art, music, programming, but skilled creative professionals is a/the core userbase for Apple products.
Apple is stuck and it’s AI will never be good enough until those creatives embrace it. Right now it’s disdain when mentioned.
An oft-cited quote goes something like this: "we wanted robots/AI to automate boring, routine, meaningless jobs to let people be free to pursue arts, music, creativity. It's a sad state of affairs that AI is taking over arts/music/creativity stranding people with boring, routine, meaningless jobs"
AutoCAD came to the Mac when Intel was shitting the bed (with aggressive OEM contracts for first party system integrators that prevented AMD adoption across HP/Dell/Lenovo-lines) and Windows 11 was being forced on users.
WINTEL played the monopoly game too hard and is starting to lose ground.
When Siri first debuted it would automatically beep, so I could immediately tell if the phone did not recognize recognize "Hey Siri" (just "Siri" didn't work). A couple of iOS updates later this went away, which means I can't tell without actually picking up the phone and looking at it whether the command was accepted.
Even more annoyingly, sometimes there is a beep! ¯\_(ツ)_/¯
Whoa, whoa, whoa, on occasion, if the planets align just right, I can also get Siri to set a reminder (and at least half the time Siri gets it 80% right).
If you haven't tried OpenAI's advanced voice mode, it's a mind blowing version of exactly what things like Siri really ought to become with a little more development. If that's what you mean by LLM Siri, I totally agree.
Being able to chat casually with low latency, correct yourself, switch languages mid-sentence, incorporate context throughout a back-and-forth conversation etc. turns talking to these kinds of systems from a painful chore into something that can actually add value.
It's the other way around. The model is impeccable at "understanding text." It's a gigantic mathematical spreadsheet that quantifies meaning. The model probably "understands" better than any human ever could. Running that backwards into producing new text is where it gets hand-wavy & it becomes unclear if the generative algorithms are really progressing on the same track that humans are on, or just some parallel track that diverges or even terminates early.
Only if you wildly oversimply to the level of being misleading.
The precise mechanism LLMs use for reaching their probability distributions is why they are able to pass most undergraduate level exams, whereas the Markov chain projects I made 15-20 years ago were not.
Even as an intermediary, word2vec had to build a space in which the concept of "gender" exists such that "man" -> "woman" ~= "king" -> "queen".
3 lines? That's still going to be oversimplifed to the point of being wrong, but OK.
Make a bunch of neural nets to recognise every concept, the same way you would make them to recognise numbers or letters in handwiting recognition. Glue them together with more neural nets. Put another on the end to turn concepts back into words.
... Oh interesting. And those concepts are hand picked or generated automatically somehow?
> For a less wrong but still introductory summary that still glosses over stuff, about 1.5 hours of 3blue1brown videos
Sorry, my religion forbids me from watching talking heads. I'll have to live with your summary for now. Until I run into someone who condensed those 1.5 hours in text that takes at most 30 min to read...
> Oh interesting. And those concepts are hand picked or generated automatically somehow?
Fully automated.
> Sorry, my religion forbids me from watching talking heads.
What about professional maths communicators who created their own open sourced python library for creating video content and doesn't even show their face on most videos?
You're unlikely to get a better time-quality trade-off on any maths topic than a 3blue1brown video.
He's the kind of presenter that others try to mimic because he's so good at what he does — you may recognise the visuals from elsewhere because of the library he created[0] in order to visualise the topics he was discussing.
Simplifying to that point is more of what a Markov chain is. LLMs are able to generalize a lot more than that, and it's sufficient to "understand text" on a decent level. Even a relatively small model can take, e.g. even this poorly prompted request:
"The user has requested 'remind me to pay my bills 8 PM tomorrow'. The current date is 2025-02-24. Your available commands are 'set_reminder' (time, description), 'set_alarm' (time), 'send_email' (to, subject, content). Respond with the command and its inputs."
And the most likely response will be what the user wanted.
A Markov chain (only using the probabilities of word orders from sentences in its training set) could never output a command that wasn't stitched together from existing ones (i.e. it would always output a valid command name, but if no one had requested a reminder for a date in 2026 before it was trained, it would never output that year). No amount of documents saying "2026 is the year after 2025" would make a Markov chain understand that fact, but LLMs are able to "understand" that.
I’m confident that LLM’s will not have hallucination problems in the type of requests that I send to Siri.
I don’t ask Siri for facts (just like I don’t ask LLM’s for facts). As long as it can correctly, understand what and when I ask to be reminded about something, that would be a huge improvement for me.
That and being able to map “Bedroom Fan”/“Bedroom Fan Light” to “Bedroom Fan Lights” without having to specify aliases (and even then it hearing me wrong).
I’ve see Home Assistant working with LLMs and it can understand groupings that I never explicitly defined which is very nice. I can say “Turn off all overhead lights” and it will find all my overhead lights and turn them off. Siri/Alexa can’t handle those tasks currently.
They are AMAZING at understanding. I think even more reliably than generating. And with context, and back and forth.
Try asking ChatGPT to remember some obscure film you can only remember very hazy details about - really random stuff - I bet it will identify it for you I a few tries.
You can also totally miss spell words, use messed up grammar and it has no problems at all
Over on Android it's the opposite situation. The voice interface to Google Assistant was very reliable for simple things like reminders and appointments, and even for general knowledge questions. It was part of why I didn't switch to an iPhone. Then Gemini came along, and that core functionality got a lot worse.
...that will grind your request to set email Vacation Mode through the world's worst speech-to-text, jam the text into Chat GPT, and spend the next three minutes reading you an uninterruptible 3 minute essay about violence.
I tried this with the new ‘apple intelligence’ that I thought could see my screen.
I had a birthday invite with clear date and time: so I asked it to add to my calendar.
It just said, “add what?” repeatedly until finally deciding it needed to send it to chatGPT to help. Which it did, then just returned the text in the image without taking any action. Then I say “can you add the event now?”
“Add what?”
So I try copying the text from the image in photos and giving that to Siri to add as an event. Surly this can work?!?
I'm a formerly non-mac guy who finally bought a brand new iPad. I got bored with the wallpaper but couldn't figure out how to change it. "Hey Siri, how do I change the wallpaper?"..."Sorry, I can't help with that". Tried a couple more questions and all it did was Google it for me. This is the latest M4 that was around $2k.
This is what our "AI accelerated" chips give us in return? What a disgrace
Yeah, the real stock market mover is the mouse cursor. That wiggly thing they did that grows the mouse cursor really send ripples throughout Wall Street.
He wasn’t stupid. He’d observe their own developers and see how they rely on the terminal and command line for their work. He’d ask pointed questions and demand thoughtful answers.
Then he’d find a way to make it the #1 AI developer platform or distort reality until it is.
Or you can take the extra minute to install them yourself with brew, this a complete non issue for anyone that understands the command line in the 21st century using MacOS. Also, I would never build anything against macos userland because it's almost never the target.
So what? Do you complain when apt, yum, dnf, pacman, ports, or any other package management system does a download? I bet you don't, so it's not really a usable argument. Secondly, yeah, not tainting my systems OS and system paths is a good thing and opt/ from the filesystem perspective is the absolute right place to put add on packages.
"The /opt/ directory is normally reserved for software and add-on packages that are not part of the default installation"
Desktop Linux is right here, desktops Linux has always been here!
I don't understand why people complain about Apple neglecting developers when desktop Linux provides a superior experience aside from the rare times you need to compile an iOS/Mac specific application.
I don't get it either, it takes twenty minutes to burn a USB stick and run the installer. It takes me longer to remove the bloatware and set up my preferred settings on a proprietary OS than it does to install Linux nowadays, and that's been true for a decade at least now.
It's just not hard! It's not more work! And yet the meme about it being more trouble just. won't. die.
You people are supposed to be technologists! Why won't you spend 20 minutes of one time setup to get a better experience?
Not all hardware works? I don't see anybody complaining about having a limited set of hardware options when they buy Apple! Canonical maintains a list of fully compatible computers; just pick one, buy it, and you wind up with a computer just as easy to use as Mac OS but without the endless paper cuts of using a system that has no respect for you at all and thinks it knows better
I use Apple laptops primarily for the hardware. But Linux has never really been a great experience on Mac laptops when it comes to battery life, reliable suspend/resume, etc. etc. I used to use Yellow Dog Linux on a G4 PowerBook way back in the day, but I haven’t had much luck with Linux on Mac hardware since then.
Most of us have .dotfiles, I can snap any macos installation into my preferred configuration in about 5-10 minutes unattended depending on internet speeds. I do most of my work in a terminal, as long as that works, I'm good on Linux, MacOS, and BSD's. They all have pros and cons.
Edit: In case it's not clear from my initial, gut-driven snark: I definitely think if you use a reasonably popular distro (commercially backed or not) in 2025, you should never have any trouble connecting peripherals to it, with the possible exception of Bluetooth, which I hear also applies to macOS.
I'm on Windows, and my work laptop has no audio via its headphone jack.
I have no idea how this is supposed to even work, and since it's not my computer I don't mess around trying to install drivers. I just use my phone to call in for Teams.
Things happen, let's not act like any OS is perfect.
MacOS is definitely not perfect. I'm being snarky. But it has been my anecdotal experience as both a user and observing colleagues that MacOS is more reliable and stable for desktop use than Linux. This is unsurprising since it's easier to build a stable walled garden than an open ecosystem.
Macs are generally more reliable, but if you buy a year old ThinkPad Linux will be just as stable .
The only issue Linux really has is when new chipsets come out you might need to wait 6 months or so for the drivers to be updated. But to be completely fair, on one of my laptops I had no webcam support for like six or seven months until Windows update decided to finally install it for me.
If you need a significant amount of hard drive space, Macs are almost always exorbitantly expensive. I make music so I find myself dual booting between windows and Linux. I don't want to speed 3k+ on a MacBook just to get a 4TB SSD I can add to any Windows PC for 200$.
Plus on Linux you can customize your personal experience to a much greater level. If you dislike X,Y,Z you can disable it or find an alternative.
Both OSX and Windows are cramming so much monetization into the OS, there's a very real feeling that I'm just sharing my computer with a giant corporation rather than actually owning it.
It’s less convenient when you are on the go, but you can pack an external SSD and offload stuff to it. A friend of mine had one velcroed to the back of the screen.
It's actually cheaper to own a MacBook Air for things that need to work 100%, like a coding interview, and then a secondary laptop when you're playing video games or making music .
That's basically what I do now, my old M1 MacBook air is more than good enough for LeetCode and I'm more or less know it's never going to fail.
I haven't personally experienced that problem. Updates on Mac have always been smooth for me. But I'm a sample of one and it's probably workflow dependent.
Presumably these users have audio in other contexts? Are they running the web app version of teams? Do other web apps play audio? From 10000 feet up, I wouldn’t start by blaming Linux here (even as a non-Linux-desktop user,)
This isn't a hill I want to die on, but isn't it the case that even if the problem is in MS software compatibility with Linux that still results in desktop Linux being a less reliable platform for day to day use?
Last year I was using a Windows laptop for work and Teams was very unreliable with audio and video. And don’t even think of using the nice camera on top of the expensive video conferencing monitor on my desk.
Tons of peripheral devices do not work well or reliably on Linux, and I literally cannot remember the last time I have had ANY issue with Bluetooth on macOS. Certainly not in the last decade.
The last time I did was this morning. I get dropped connections constantly, microphone not working in Teams (solved by reboot), pegged connections preventing handoff, etc.
I went through 5 distros a month ago dealing with fractional scaling issues on my 4K monitors. Decided it is not worth dealing with a went back to macOS so… No.
I very recently tried again to adopt Linux on the desktop. I'm really sick of feeling like a frog in a pot of water. It's becoming harder and harder to bypass their literal gatekeeping of which applications I can run on my computer, and with every new version of macOS the temperature in the pot keeps rising.
The main problem I have with living in a Gnome desktop environment, is with the keyboard. I'm not willing to abandon my use of Emacs control+meta sequences for cursor and editing movements everywhere in the GUI. On macOS, this works because the command (super/Win on Linux/Windows) key is used for common shortcuts and the control key is free for editing shortcuts.
I spent a day or so hacking around with kanata[0], which is a kernel level keyboard remapping tool, that lets you define keyboard mapping layers in a similar way you might with QMK firmware. When I press the 'super/win/cmd' it activates a layer which maps certain sequences to their control equivalents, so I can create tabs, close windows, copy and paste (and many more) like my macOS muscle memory wants to do. Other super key sequences (like Super-L for lock desktop or Super-Tab for window cycling) are unchanged. Furthermore, when I hit the control or meta/alt/option key, it activates a layer where Emacs editing keys are emulated using the Gnome equivalents. For example, C-a and C-e are mapped to home/end, etc.
The only problem is, this is not the behavior I want in terminals or in GNU/Emacs itself. So I installed a Gnome shell extension[1] that exports information about the active window state to a DBUS endpoint. That let me write a small python daemon (managed by a systemd user service) which wakes up whenever the active window changes. Based on this info, I send a message to the TCP server that kanata (also managed by a systemd user service) provides for remote control to switch to the appropriate layer.
After doing this, and tweaking my Gnome setup for another day or so, I am just as comfortable on my Linux machine as I am on my Mac. My main applications are Emacs, Firefox, Mattermost, Slack, ChatGPT, Discord, Kitty, and Steam. My Linux box was previously my Windows gaming box (don't get me started about frog boiling on Windows) and I'm amazed that I can play all my favorite titles (Manor Lords, Hell Let Loose, Foundation) on Linux with Proton.
I had an old i5 Mac mini laying about I wanted to use desktop Linux on the other day. The last time I tried, was about 20 years ago. I note nothing has changed since.
I made this. It makes it easy to use all of the most common gnu tools via brew, without having to do gsed for sed, etc... all with working man pages. It also lets you switch back easily in a shell session if you need the mac native ones for some godforsaken reason:
So are the BSD tools by some definition of "behind". Another way to look at this is to say that GNU tools as typically seen in modern Linux are bloated (I know, Linux and "bloat" are kind of a meme, but it is generically true for the most part when it comes to the command line utilities feature creep over the years, so it's a useful and descriptive word).
I have to work with old machines and legacy operating systems quite a bit in my day to day and I always am going to prefer something lighter and with less ways to shoot myself in the foot w.r.t. POSIX compliance. MacOS is Unix certified so I appreciate them being somewhat reserved in the features they add on top of POSIX.
Modern GNU userland utils are nice and fun but if you are looking for compatibility it's best not to use them. Consequently, the MacOS situation doesn't bother me especially given you can install more up to date tools if you want. I think keeping the defaults older and more compatible is a good thing.
There is poor cross-UNIX compatibility if you're doing anything complicated, anyway. I maintained a large test suite for about a year that was written in POSIX sh and targeted Linux, macos, {Free,Open,Net}BSD. It wasn't fun because every program behaved in slightly different ways, half of them undocumented (for example, I remember having lots of pain with how different versions of tail handled SIGPIPE).
In the end it was was easier to rewrite in Perl than to keep maintaining that thing, struggling for hours to find ways of implementing every little bit of functionality that worked reliably on every OS. You'd add or fix something, and the tests would break on FreeBSD. You would fix it there and it would stop working on NetBSD. And so it goes.
I prefer a lot of the BSD variants of the typical POSIX tools (i.e. bsdtar vs GNU tar, ksh or similar instead of bash, etc etc). Usually because they add less extensions on top of what is required by POSIX, but are still easily acquired in a modern Linux distribution. I mostly just alias them.
If I write a script using BSD esque tools I can be reasonably sure they will work on any Unix-like, whereas if I write/test my script on a machine using GNU utils, I'm fairly likely to accidentally use a GNU extension that would cause the script to fail on an older Unix-like OS. For instance, I do a lot of work migrating code off of AIX,and I need the scripts I develop to work on AIX when I'm gathering environment information from customers. I can't just assume they will have a ~2020+ implementation of Unix userland tools with all the GNU extensions and nice features. Sometimes the machines have been sitting quietly in the back of a data center not being updated for quite a while and will have more "90s style" of Unix tools.
It’s like having the tools in a different computer. You can mount your local filesystem onto the container, but if feels like WSL - there’s always an “impedance mismatch” between the two sides.
I prefer to use the tools running locally on the same OS I’m working with. For that, MacPorts is great.
I detached this comment from https://news.ycombinator.com/item?id=43158188 so we could pin the latter to the top. What you posted was fine! I just don't want to take up extra real estate at the top of the thread.
We used to call this an "Industrial policy" or an "Economic development policy". Back in the golden era when a strong labor movement coexisted with a Red Scare. 78 years after Taft-Hartley and 44 years since PATCO, not so much.
We have maybe fifty or a hundred million people rotting away in areas where jobs are scarce and housing is plentiful, because we used government policy to shut them out of areas where jobs are plentiful and housing is scarce. We systematically exported jobs from places that aren't big cities because they can be performed overseas and our aristocracy can still profit from them by owning those people overseas.
I don't know if returning to a little more deliberate of an economy is even a partial salve for the place we've found ourselves, but I don't think this laissez faire thing is sustainable for a whole lot longer. We are overleveraged, and arrogantly delusional about our sway at the moment; "Ownership" is not some valuable skill. The fall of an anchor currency and global conversion to an alternate financial network would be a spectacular thing, an astroid striking terrain, which might leave craters on entire other continents from secondary ejecta. World wars have been fought over less.
"We" wasn't big government. It was a million homeowners who decided that the neighborhood they moved into should be frozen in amber forever. Everyone wants housing to be cheap but also for their property values to rise onto infinity. They push back against any attempt to change this and then complain about the inevitable results.
Say what you want about Trump, this is the kind of deal that wouldn't have happened with any of the previous administrations, both Democrat or Republican. It's the kind of deal that keeps MAGA loyal to him, despite all the noise about DOGE.
