This doesn't seem like the giant step I thought it was:
> Unlike the costly bleeding edge 2 and 3 nanometer chips made by giants like TSMC, TI’s chips are made on cheaper, legacy nodes: 45 to 130 nanometers.
What's worse:
> Making chips takes an immense amount of water, and about a quarter of Texas is in drought. Luckily, Sherman has water rights to nearby Lake Texoma.
Indeed lucky for TI, but probably not so for area residents.
Texas is a big state. It’s lazy journalism to generalize the state as droughty, which is implied by that sentence.
Lake Texoma has been hovering by the “full” mark pretty consistently for over 55 years. Recently, its water level has significantly declined to—wait for it—100% full!
If you monitor water maps, the east half of Texas’s water supplies don’t often get far outside of “full”.
Also the water doesn’t disappear from the universe. It either evaporates or is pumped back out. People who lose their minds over water have been saying for 50 years we were about to enter a desert world
Clean water does disappear. Not from universe bit from places you need it most. That there's plenty of it in the air or sea or Saturn rings, is little consolation.
These fabs don’t evaporate the water though, they use it as process water, and then treat it to wastewater standards before discharging to the municipal wastewater system.
Assuming the municipality recycles their wastewater, which they would do in any drought prone region, this water will become clean water again.
I read they either dump it into a local river while monitoring it or back to water treatment plant. Some datacenters have started reusing water but for some reason using it more than once is not appealing. Definitely the biggest problem is it’s not always a closed system and dumping it into a river potentially damages the river and makes the area lose their water. A percentage of the water is also lost via evaporation.
It’s definitely not a closed system unless the water from the waste water treatment plant is pumped back upstream of the source of the municipal water which is not how most of these work.
You should definitely bring this attitude into the housing affordability discussion - remind people that they indeed can find cheap housing in rural Africa.
There's lots of more boring chips in phones and mundane devices that are fine on older processes; things like audio stuff and battery controllers.
If you want to get away from relying on China, then those chips are the ones that are in everything from your USB hubs, keyboards, phones, washing machines - everything
Less sexy, but in many ways more critical to keeping stuff going.
The trailing edge is also where Moore's law in its original sense is still going strong. The fabrication processes get cheaper and more refined over time, while this is becoming a lot harder with new leading-edge nodes.
This is misleading, modern chip fabs recycle their water basically 100%. It's not being evaporated for cooling like in some data center deployments, it's just part of the process.
That is of course the ideal, however local economic conditions (low/subsidized water prices, etc.) may dictate a different path by incentivizing another way of doing things (open cooling systems, etc.).
They already have fabs nearby, and so, fab suppliers/services in the area.
They have a large employee pool in the vicinity, with modest salary expectations.
The southern US mostly doesn't have blizzards and ice that regularly knocks out electric supply lines, and/or makes it impossible for employees to get to/from work for days/weeks at a time.
Water is about the cheapest resource around. Even in water-stressed areas (while it may cost several times what it would in wetter parts of the US) it's still an insignificant expense for any industry that doesn't need an absolutely obscene continual supply.
Roof-top PV solar power works out a lot better in the southern US than further north.
> The southern US mostly doesn't have blizzards and ice that regularly knocks out electric supply lines
Except for when they do, then they are literally paralyzed until it melts. This may only happen once per decade, bot wow it sucks. One ice storm in Birmingham was so bad kids were stuck at school for 3-4 nights.
The transfer of wealth from tax payers to tech giants has never been better.
Enormous state subsidies.
For doing something the free market would do if it made economic sense.
How many bailouts is Intel on now?
The idea of forcing a merger between Intel and TMSC at what could be called
be barrel of the gun, would at least make it (indirectly) Taiwan that would
have to take over paying for Intel.
You can’t build a modern fab (which this one doesn’t seem to be? But even old fabs are extremely expansive) without subsidies. They’re extremely expensive to build, and depreciate so quickly, no private party is willing to put the funds.
It's not socialism, it's corrupt state collaboration with capitalist corporations to suppress labor. That's fascism.
Where state, employers, and workers collaborate through formal national
mechanisms to collectively manage the economy, prioritizing the needs of the
nation and protecting corporate interests while suppressing labor.
Technically then they aren’t wrong to equate this to the USSR - it’s important to remember that both china and Russia are “transitioning” to communism, although both are kind of stuck in the in between phase where the people who should be ushering in that change found instead that they can wield authoritarian power to enrich themselves and subjugate others.
Other side of things is you’re absolutely correct that this feels like a fascism, which is what this administration continues to feel like…over and over and over and over again
The American people have been lied to for so long about what Socialism actually is to even be able to recognize it when it's happening right before their eyes.
No this is what happens in authoritarian states all the time. Dear leader has a conversation with a “friend” and suddenly that friend’s enemies have some new limitation from the government. Socialism takes the form of generally higher taxes on profits from personal or corporate income which pay for things for the whole society.
The US hasn't been anything remotely akin to the capital of Capitalism in more than half a century. The US has a hyper regulated, highly taxed economy, to go with a gigantic government. The capital of Capitalism would be the opposite in most every respect. Go back to the late 19th century in the US as one example, if you want to see low regulation and low taxation.
Holy shit, somehow I missed Trump trying to sell half of Intel to a Chinese company that doesn't want it. TSMC is a great company, but this sounds disharmonious with Trump's jingoist and xenophobic rhetoric.
The deal hasn't actually closed, though, has it? The linked page says, "However, TSMC is unlikely to agree to this proposal."
You have likely not experienced it as governments and monetary systems have absorbed these risks (ie. they have moved them elsewhere, hidden away, until some major explosion happens).
Chips are like agriculture: It does not matter that the home grown product can only compete with government support, you never want to be in a position where you cannot provide for your country’s minimal needs purely from local sources.
It depends what the minimal needs are, but basically the US has this for chips already. Yes the US only makes older chips, but in fact those are the ones that really matter, that you can’t do without. The ones that are in everything.
However making everything domestically just isn’t really viable for any country any more. The scale of our technological civilization and the diversity of goods and materials it depends on is more that any one continent can support, let alone one country.
It would be possible to collapse that down to one continent if absolutely necessary, but it would be incredibly economically painful and the US would need to give up on a lot of non-essentials and other priorities to devote resources to duplicating capacity that already exists elsewhere.
Buying rocks and exporting chips is fine. Of course it's not that extreme but the sentiment is there. It's about expertise capital and concentrating that domestically.
I think it is not about one country doing to all. But technology transfer should happen within like minded countries and within a smaller circle. So may be two continents max.
Are there going to be Congressional hearings at least where Citizens can see a debate with witnesses testifying under oath? Will the Congressional research office publish statistics and data we can read together to debate?
Or will some backroom deal be made without any Congressional approval? I got a lot of questions and basically there is no way to get answers since Republicans gave the president:
1. unilateral authority to set tax rates (tariffs and illegal export levies)
1. impound Congressionally appropriated funds on an ad-hoc basis
1. unilaterally close down federal agencies created by Congress
I refuse to debate the merits of this industrial policy while the above three things are true. There is nothing to debate here because no one outside the administration's inner circle (industrialists bribing) who knows what is going on.
I feel like my 7th grade Civics class has been a complete waste. Or, it lets me see what is happening in the correct light.
Shame you've been downvoted. Regarding federal agencies, the Constitution indeed means what it says when it says "The executive Power shall be vested in a President of the United States of America." If some agency can wield executive power, it is under the exclusive control of the president. Looking forward to the likely overturning of Humphrey's Executor soon. It's gone in all but name anyway.
Perhaps this will spur some future Congress to take back what they've delegated (like sweeping tariff power)
This is a completely meaningless statement. It sounds good, but contains no actual argument. why are chips like agriculture? What makes that statement not apply to literally every product category that exists?
Agriculture is self-evident, we need food on a daily basis to live. I don’t need three square microprocessors a day.
Just like you don’t want to wait many months to tool up to make food if there is conflict…
You don’t want to wait many years to tool up to make chips that go in weapons, either.
Else, you risk being structurally defeated very early, and this leads to others concluding that they can take you on. It undermines other preparations.
Yes, we understand that, but it’s not a good comparison.
Producing chips and agricultural goods are fundamentally different, starting with limits like climate, seasonality, and land. At most, you could compare it to mining minerals for chip production
The only things that matters for this consideration are:
A. How bad is the consequence of not having them?
B. How long does it take to tool up domestic production if there's a supply disruption?
C. How long can you tolerate the shortage of them?
D. How easy is it for a disruption or disaster to disrupt international supply?
The answers to A and D are fairly different, but both are "bad enough" to trigger this kind of thinking. The ratio of B to C are probably similar between food and critical integrated circuits.
Farming equipment needs microchips, as does practically any modern device, vehicle, machine, etc. If you don't have domestic microchip production then pretty much everything else you make will have foreign dependencies that could be embargoed.
Microchips for farming are made in several counties including US. There are highly specific processors, like those required for ML, but that’s not what farming can live without.
Having narratives as broad as yours, without any evidence and reasoning, is the fastest way to achieve Russia-level disaster with government monopoly in the name of “security”.
The only sense it makes if you want corruption at state level and to go on war with at least half of the world.
I think there's surely a little bit of a middle ground?
Subsidizing and protecting local production has a cost: there are gains from trade and operating less efficiently makes everyone's standard of living goes down.
But you don't need to go 100% into protectionism. You can ensure that you have enough domestic industry to
- better weather changes and disruptions
- have a starting knowledge and capital base to use to ramp up production if you have to.
Just having a few percent of domestic demand for advanced ICs covered by local production is enough to make the worst international shocks much softer. Critical needs can still get filled. So a certain amount of subsidy can make sense.
Of course we'd think that, we're on the Internet, discussing it. The people who aren't, aren't here to rebut that. I met this woman who lives part-time in Romania, and was saying that they don't have smartphones still, which is why she has to spend her time in Romania, because she can't just have this particular subset of elderly people in Romania get information from her off of Instagram.
>It does not matter that the home grown product can only compete with government support,
Even that isn't a given, because unless you have amassed a certain amount of technocratic and governmental competency chances are it can't compete even with government support and you just produce crony dysfunctional companies.
And of course there's economic trade offs. If you're politically ordering your economy to make chips, it doesn't make something else, and whatever it was making and trading for chips it was better at, and so you get fewer chips, that's comparative advantage. Industrial policy (and tariffs) do not increase aggregate production, they reduce it. And given that the circle of items you "can't do without" seems to be a bit of a moving target these days, at some point you're actually more brittle because you've replaced large chunks of the market with state production.
People take it as an assumption that cronyism will always happen if the government invests too strongly/consistently in a certain thing. But cronyism is a policy and structural failure, it's usually because the incentives for the different parties involved encourage it to happen. Institutions can be designed carefully if policymakers actually want to do it.
USPS is a great example of an organization that's managed to largely avoid this. Whenever you mention that people crawl out of the woodwork to complain about the 7 different times they lost their package, but their logistics at scale is still unmatched by the private sector, while also not completely negating the value of private sector alternatives (which so often is argued would happen if the government actively started doing anything new).
>People take it as an assumption that cronyism will always happen if the government invests too strongly/consistently in a certain thing.
No, it's the opposite. It's a conclusion taken from empirical evidence looking at success rates in the real world. As the US has engaged on its most recent bout of industrial policy, industrial activity has declined, not gone up. Again, this is expected. If you through protectionism make chips and steel in the US inefficiently, everyone in the US using those products as inputs suffers.
There's some limited cases like developing countries engaging in catch up growth , but there's virtually no evidence for effectiveness of these policies in cutting edge technologies, which isn't surprising because they by definition tend to rely on supply chains and knowledge sourced from all across the globe, and to this protectionism is particularly disruptive. And even in developing nations most of the time practices like ISI (import substitution industrialization) fail, it's devastated Latin America in the 1950s to 1980s.
I don't know what USPS proves in this case because it's sustained by a government monopoly. Obviously you could if you wanted privatize mail delivery.
It is established economic theory not just some opinion .
Constraints in form of regulations, subsidies, tariffs by government intervention will produce inefficiencies.
This is the core economic philosophy behind both the modern centre right[2] i.e. neoliberal democrats (80s-today) and the right wing republican conservatives (pre MAGA -2016).[1]
Even communists would agree this is correct, it is understood in different levels since Adam Smith invisible hand, from far left to centre left just posit that different polices that benefit the people not the economy are worth the cost .
That is actually true for all other groups except anarchists and libertarians all agree government intervention is needed in some form, they just disagree on how .
——-
Postal services anywhere in the world are inefficient by design. They are governed by universal service obligation principles not efficiency same with telecom providers and other utilities.
The question then becomes how much inefficiency is acceptable given the objectives . There are no right answers, you could have competing private couriers who dump unprofitable low density routes to USPS while serving profitable ones themselves (UPS, fedEX, Amazon ) or you could have gigantic postal service which is sole delivery provider socialist style, there are going to be some problems either way.
It is just matter of preference which set of trade offs we are okay with .
—-
[1] Post MAGA it is a populist party economically there is no fixed ideology to be characterized
[2] the centre right label is economic and technical to contrast say the socialist left party of FDR/New Deal 1930-70s, not meant to offend.
But that's exactly the thing, where we care about an industrial capacity for its strategic reasons rather than because we're trying to maximize efficiency of revenue extraction. Pure market efficiency is great if you're trying to maximize profits, but that's not necessarily the goal here because the benefits are indirect, just as they are with universal postal service (or universal healthcare, etc).
Maximize economic output not necessarily profit outside of that I agree with your comment, nobody is questioning whether we should have other goals.
The problem is however first can we all define the goals consistently and agree on the definition.
Next can we measure the value/benefits of the goal. This is typically subjective so everybody paints the picture best suited to their preconceived ideas.
Finally perception drives the political will for a social program. UBI style programs or just giving money in welfare instead of in-kind benefits for food or healthcare etc or rehabilitation over punishment are more efficient and while shown to work, will rarely get political support because of how they are perceived.
Given that we all have different understanding of the goals, measure them differently and have different moralities on what should be done as society (for welfare, crime etc) and have varying degrees of influence over policy what can we do to approach a problem better
How? What if we just decide we will never go to war with Mexico or Canada and get comfortable with the idea of importing from our allies? There is no serious future risk from doing that
100 years ago the British Empire ruled the world, now it's a small island you don't hear much about...
The US is only about 250 years old, I'd be cautious about predicting the future
You're off by about 150 years. 100 years ago was 1925, the British has already largely collapsed losing the US, Canada, Australia. 100 years isn't forever but it's a long time
And we haven't had any serious threats from Canada since 1812. I think the most reasonable estimate is 100-200 years
~1920 is the peak of the British Empire in term of territory, anyways, the details are meaningless, what matters is that things move fast and just because you're at the top of your game right now doesn't mean you'll be in the same position in 100 years
I could also take the example of world wars, in France ww1 was deemed "la der des ders", which meant "the very last war" or "the war to end all wars", well 20 years later we were at it again
Or simply look at China, you don't even have to go back 100 years in the past to see drastic changes.
That's only the case if you include Canada and Australia, which were functionally independent at that point.
I'm not claiming nobody will invade France or Taiwan in the next 100 years, I'm claiming that the US is special. We haven't been invaded since 1812 and haven't really been attacked since 1941. It's reasonable to predict we won't be invaded or go to war with our neighbors for 100 years since it hasn't happened for 213 years
No, based on the history of conflict we can say that the more time that passes with neighbors not invading one another, the less likely they will in the future
If we include the idea that either one of them is allied to a major power at war with the US over a hundred year horizon, right now that looks pretty likely, and arguably is one of the things the current US admin are trying to stop before it becomes inevitable.
The US admin is very clearly pushing us in the opposite direction. You believe that Trump's actions make war with Canada less likely? What's the mechanistic explanation?
So the argument is that the neighbors will never ally and be involved in a war that puts US on the opposite side? What is the argument for the sovereign neighbors to always be neutral or on US side come what may?
Germany was trying to get Mexico to join them for wwi, with enticement of getting back the land they lost 70 years prior. That loss was immense for Mexico in land, pride, and economics
Mexico was dealing with its own internal issues (revolution) which made it difficult for them
A slight turn of events and the us would have a huge southern front I deal with, and a base of support for disaffected native Americans and African Americans.
Russia also dropped out of Wwi due to its intern revolution
It is easy to look back and see manifest destiny as a given. There were a lot of contingencies
You understand that for basically every violent entity in the world (except China and Russia), avoiding the US Mil is their best path to success and continued survival?
"War" as a concept does not have to include two militaries fighting each other.
If the cartels are waging a war against the US, so is a large number of American oligarchs. When will they start being on the receiving end of drone strikes and special operations?
When you stretch the definition of war to absurdity, so can I.
It's an extension of a peace through strength philosophy. If you lose your critical sovereign capabilities you become weak and vulnerable. You no longer get to decide who you do and don't "never go to war with".
Yes I understand the reasoning, it's just obviously wrong. This doesn't happen and hasn't happened in hundreds of years. This is not why people get invaded, otherwise Switzerland would have been invaded many times over the last century
That's exactly why Switzerland didn't got invaded in the last century. Every side depended on them for money supply, so they couldn't risk bringing the war to Switzerland. Also it had a track record of staying neutral, so they didn't feared it may pick a side, because it largely made money from selling stuff from both sides.
This kind of answers your own question: the reality is you are only a reliable US ally [1] if you can hold them by the balls TSMC style. Given that countries go to great lengths to develop and maintain such dependencies. Canada's current weakness is at least in part because it has failed to do so.
[1] Edit to add: This was/is poorly worded - I mean that the US will only guarantee that you remain an ally while they are in some sense dependent on you, and while doing so they may work to break that dependence, which you may interpret as them trying to abandon you.
You are claiming that Canada will stop exporting to the US in the future? Populism does the opposite, they would likely stop importing. There's almost no risk to us that they would not export chips
My claim is Canada has no unique product that it offers that the US is sufficiently conspicuously dependent on for it to guarantee any real sense of true independence, and so it finds itself subject to the whims of a foreign state.
Right now there is a very large Canadian boycott of US products, services and tourism. I also had to explain to a client this week that because of the import tariffs on Chinese goods to the US the US assembled products are now no longer competitive with alternatives. The fact you were seemingly unaware of this kind of demonstrates the level of effect of it.
The Nintendo Wii SoC was actually fabbed in Canada and exported, but that facility has changed into something slightly different because the whole east coast/hudson river valley fab world went sideways a while ago.
> We import tons of food and energy from them and have no alternative on time scales or 10 years
More importantly for Canadians that food or energy has no alternative competing market to sell into. Consequently the Canadians are totally dependent on the US market to even set the price of it. This applies to many other sectors as well.
Canada is currently having a huge desperate push to export to non US markets because of the levels of uncertainty that have been created. And I say this as someone not totally dismissive of the US position, but they need to do a far better job of bringing their allies into the tent with them.
Historically there have been plenty of incentives to stop exporting. Its called an embargo. Usually in an attempt to get the host country to change foreign policy, though I can't think of any situation where it actually worked. Examples: Napoleon's "Continental System" against Great Brtain, US oil embargo against Japan prior to WWII, Confederate States of America cotton embargo against the UK during the early years of the American Civil War
Yes it's definitely possible, but very rare and as you pointed out (especially in the confederacy's case) it usually harms the exporter much more than it helps.
I'd say that in any case of a serious Canadian export embargo, it will have been in retaliation to US trade policy or US invasion, not the other way around.
We had essentially no risk that Canada would embargo us, there was no possibility of this happening for the last 150 years until we became the aggressors
At any time, Canada could decide to stop subsidizing its uncompetitive chip makers for the same reason so many people in this thread want the US to do, and the US would then become dependent on someone else who might be their enemy (eg. Chinese occupied Taiwan or China itself).
Russia has been totally isolated from global market and they are still producing hundreds of cruise missiles a month just fine. America could easily figure it out long enough to survive a few yrs of conflict. Even stockpiling 10yr old chips would be good enough 95% of military industry. Then emergency investment and smuggling will cover the rest.
Meanwhile cutting off your markets and wasting hundreds of billions on a long term bet with a small probability that another global war will happen is pretty dumb financial thinking.
Tariffs and corporate welfare will actively make a country poorer and create unproductive zombie markets while raising taxes on everyone. Not to mention diverting budgets and new revenue away from actual national security investment.
Up until relatively recently we were the SOTA and #1 semiconductor exporter. When people talked about the "american manufacturing sector" a significant portion of it was actually that.
Those foundries didn't go away, they're still manufacturing with the same capabilities they used to (and they're much cheaper now since they're competing with the better ones in Taiwan.)
It's good to hear TI hasn't given up on high performance SOCs as it was beginning to look like they had. But most of this stuff is still here. Freescale and many other American companies are still making the same (better even) chips they always have which is more than enough for cruise missiles (more than enough for decent PDAs and smartphones really) even without "stockpiling."
Yes and those Russian semiconductors aren't sophisticated or high end at all which supports my point. The military isn't making cellphones, they make missiles and fly jets with decade old computer chips. A few years of war with China is not going to magically eliminate all computer chips. The global market will still exist in some form and wars are far more motivating than a few government grants.
Not to mention China (and/or Taiwan) is still going to want to sell to someone to survive, and those countries can smuggle them into America - just like Russia does for it's drone industry and oil. America is much more capable in that regards with NATO and it's huge purchasing power.
I still think TI and Apple should be investing in foundries and domestically. It should just make sense as a business otherwise it's going to be a very expensive embarrassment.
I don't know if it was on purpose or not, but I have heard it said more than once that Republican led states are able to greenlight projects faster, are more business-friendly environments, and generally have less red tape compared to Democrat led states. Love it or hate it, but greenlighting projects is a big component in allocating funds.
On the other hand, if they were so much better, shouldn't those "red states" be the ones with the much better economies?
I'm looking at China and what environmental price in the form of polluted land they pay and will be paying for a very long time. The big problems in Western countries also all originated in times with less or no regulation of such things. Just because that's not in the headlines all the time does not mean the problem is any less while the public is not paying attention.
When "it works" and overall success is the only criteria, the Vikings and Mongols surely count as extremely successful. Regulation is meant to take the price into account, in the cases of those two peoples millions of dead and a lot of pillaging and conquering. Regulation is definitely a burden, if you don't have to care about anything but the bottom line it's much easier.
Definitely not true. Right now Red states are openly attacking businesses that don't agree with the prevailing ideology. In Florida, the governor tried to destroy the state's biggest employer. In Texas they have been trying to prosecute out of state businesses. Alabama has more taxes on businesses than California.
I thought it is because they have worst economies and it is always an attempt to prop them up.
The other reason is that while republican party is purposefully trying to destroy economies of blue states, democrats were not trying to purposefully destroy economy of red states.
This might be true, it’s definitely repeated, but it’s generally not the real reason.
The real answer is just politics. Blue states have (generally) healthy economies, with a variety of economic actors and many businesses. Businesses often will shop around various states to build a factory looking for tax cuts. The politician can be associated with new jobs, and the business gets a discount, so it appears to be a win-win (if you ignore the lack of tax revenue). No one needs a tax break to start a business in NYC, LA, nor Silicon Valley, so you don’t hear about all the businesses that open there.
Nationally, policies like the IRA are big boosters for the economy, and democrats are focused on getting it done because it’s good for society and the national priorities. They won’t focus on where the money goes, and will allow the money to go to run down republican states as economic stimulus. But you’ll notice it’s usually capital intense factories that end up in these situations, not white collar jobs.
Voters in Red states destroyed any motivation by foreign companies to build factories there (tariffs) and Democrats to focus infrastructure spending there (CHIPS, IRA being worthless in getting votes).
And now Republicans are ok with impounding Congressionally assigned funds.
So I guess outside of facebook brain rot and doom-scrolling, there is no way to convince voters to vote with policy (requires attention span of more than 30 seconds and high school level reading skills).
In Michigan, there were several IRA-catalyzed investments worth many billions for battery facilities or various components — and the announcements were met with vehement protests because some of the companies involved were Korean or Chinese. It didnt matter that they were going to create many thousand good jobs and use tons of local construction, the weird conspiratorial bigotry was just too strong (fanned by the now-President and VP).
Just no idea where to go from here. Feels really dire to be honest.
Oh how I wish we would be able to split the country and not have to deal with the deplorables any longer. Unfortunately we would also have to split with the deplorables the national debt and the nuclear arsenal. Imagine what would happen then…
While that sounds like a solution it’s of the “simple but wrong” kind. Just like you, they are product of propaganda, though theirs is less based in reality than yours. If you don’t counteract the source strategy (like the lack of education, biased fox “news” and adjacent media), nothing prevents it from being repeated in the resulting halves of your solution, Zeno style.
Honestly, no. That’s enlightened centerist nonsense. Being for things like freedom of speech, free and fair elections, a non-totalitarian form of government, and being against economic suicide, masked goon squads kidnapping people off the streets is not some sort of insane propaganda that I should be questioning.
They still haven’t fully torn down a Texas Instruments plant in Stafford, TX that closed a decade ago. But now we’ve got an In and Out and Costco on the former campus.
> Making chips takes an immense amount of water, and about a quarter of Texas is in drought. Luckily, Sherman has water rights to nearby Lake Texoma. “It was about acquiring more rights, ramping up our production and being able to provide for the mass quantities of water it takes to run a semiconductor facility,” said Teamann, adding that the fab has almost doubled the amount of water Sherman uses.
I'm a bit biased because my home state is Ohio and they have it in their constitution that the state can't have a stake in any private company and can't even lend credit to any private company. And this amendment was written in blood, in the early 19th century the state nearly bankrupted itself investing in and taking stake in private companies.
* The state can't risk taxpayer money on ventures that might not pay off or lose them
money. How the state "gets around this" is by issuing zero recourse loans. The advantage is that when economic development money is handed out there's not an asset on the balance sheet. It's treated like it was spent. The value the state gets from spending the money has to be independently worth it for taxpayer without considering financial returns.
* It eliminates a whole category of conflicts of interests where the government will get squeamish regulating or punishing bad behavior because it would hurt the taxpayers' investment.
* It also eliminates vectors for corruption as well as the negative effects of the government having direct influence over specific businesses. No backdoor regulations from the state's ownership stake that don't go through the legislature.
So I'm very heavily in the camp that government shouldn't ever be allowed to have stake in any private company. The line between government and private enterprise should be the wall this admin likes to talk about. I certainly didn't expect it would be republicans I would be trying to convince that state ownership of business is a bad thing.
> So I'm very heavily in the camp that government shouldn't ever be allowed to have stake in any private company. The line between government and private enterprise should be the wall this admin likes to talk about.
completely agree, private sector should be private and public be public, and the two should never co-mingle (and that includes politicians owning stocks and bonds etc)... right now there is so much corruption (sorry, "investment") it boggles the mind
> I certainly didn't expect it would be republicans I would be trying to convince that state ownership of business is a bad thing.
because they are competing with china and they are wiping the floor with the competition, so they feel they need to do the same i assume
> I'm a bit biased because my home state is Ohio and they have it in their constitution that the state can't have a stake in any private company and can't even lend credit to any private company.
The Ohio government works around this by cooperating with private “development corporations.”
America is the last place that is short of capital for industry investments where it requires gov taxes going to it, they have a huge domestic financial market and tons of foreign investment (but those require legitimate plans, not national security woo woo). This is just propping up weak megacorp industry like they did with Boeing, instead of fostering real progress.
Sure.. calculators and MSP430s for remote power meters are keeping TI from closing up shop, but TI doesn't have the capitalization structure to bring up a fab for the types of chips people say they want. I mean sure... If you want to make 28nm chips, they're fine, and you can do a lot with 1 and 2 GHz parts, but... We keep saying we want to make the chips in the states that they're making in Shenzhen and Taipei... And honestly, a $1.6B grant from daddy warbucks may not be enough to prevent TI from taking the money and dropping out of the program in a few years.
And this comes from a place of love... My family's been invested in GSI for almost 100 years
> calculators and MSP430s for remote power meters are keeping TI from closing up shop
calculators have consistently been a minor percentage of TI’s business (~5% of profits per source below). I doubt MSP430s in particular amount to a huge percentage either
Depends on which line of DACs. And calculators are an almost irrelevant amount of TIs revenue. They don’t report it individually, but it’s categorized under the miscellaneous “other” bucket which is only 6% of their business and includes DLP and “other charges” related to M&A. $947 million with all those other things means you’re talking about probably 100-300 million in revenue. There’s other businesses within TI that do more revenue than that by themselves.
I’m not saying that at all. I’m saying it’s nearly meaningless in the grand scheme of their overall revenue. They make roughly 50% margin on their $17 billion in revenue right now. If somehow calculators cost nothing (100% margin) it wouldn’t make a meaningful difference on their overall margin.
Saying "TI makes money on calculators" is a pretty misleading statement outside of any other context. Its a tiny part of their profits and revenues. It's like suggesting McDonalds is an ice cream shop. Sure it's on the menu and they make a profit on it but it's a small side business after selling burgers and fries.
> Sure it's on the menu and they make a profit on it
My point is, the statement "calculators and MSP430s for remote power meters are keeping TI from closing up shop" isn't based in reality. Both of these products are tiny parts of their business. If that's all you know of TI, you don't know TI. Like thinking McDonald's is an ice cream shop and completely being ignorant of the burgers and fries, saying those ice cream cones are keeping McDonald's from closing up shop.
Even for the MSP430, it's a small product line of their wireless and microcontroller products.
I’m really confused how you got that from the comment you’re replying to, and why you’re continuing to defend misinformation you’re spreading in the original comment. you implied TI primarily makes money from calculators and MSP430s. this is easily provably false
the person above made an analogy —- they didn’t claim TI loses money per calculator
Microcontrollers and calculators are a small part of TI's revenues. Most (>70%) of their revenues come from analog devices like amplifiers, DC-DC converters, ADC/DACs, and things like that.
They make important chips and many top of the line products of their segments but they're not things like server grade CPUs or GPUs.
Yes. They make most of their money from low margin parts. That's not as good a story as you might think it is. Though... making money is certainly better than not making money. And yes, they have a decent mixed signal story.
But... everyone seems to think TI will be competing with TSMC's and Samsung's small-node parts. And they probably could, but they would need to a) build a fab that can make 5 or 3nm parts and b) build a sales channel for new parts. I was alive in the 2000s so remember TI doing an exceptionally poor job of step b.
Their analog division has >50% margin, a good bit more margin than their MSP430's and graphing calculators. That's not far off from TSMC's overall margin.
It's a better story than your misleading statements acting like TI only makes calculators and old microprocessors and flat out inaccurate ones about profit margins.
I only saw their OPM broken out by division. OPM was around 37-38% in 2025Q2. Do you have numbers for NPM broken out by division? But yes, if they could get volumes like analog or mixed signal with margins like "other" or "embedded" that would be pretty awesome.
as I understand it (from outside), such calculators are part of american curriculum and is pretty much demanded on SAT and similar school-independent testing
I'm guessing the commenter above was asking what trading places and commodities futures have to do with the referenced article. I'm trying to figure out if you view Trump + cronies as the good guys in Trading Places (Dan Aykroyd + Eddie Murphy) or the bad guys (Ralph Bellamy + Don Ameche). Or if you think the old institutional guys conspiring to ruin young Dan Aykroyd's life over a bet were the bad guys or the good guys? I'm not trying to be snarky, it's just that there are a lot of opinions offered on this site and we don't all view the world the same way.
There are a great number of opinions expressed on this site. I generally think "cronies" is pejorative and you generally think "cronies" is pejorative, but I take it on faith that there is at least one person out there thinking "hunh... cronies? yeah. That sounds good. I don't know why people get so upset when I say that."
And I do not know whether the person using it in a sentence is using it pejoratively. And heck, there are PLENTY of heresies I subscribe to: testing software is more often useful than it isn't, the web was done by amateurs, adding comments to code documenting the programmers' intent is useful, the rust borrow checker should have just been an add-on tool for C++, etc. So it may be ME who is against the grain on this one.
Which is to say... just be careful out there when you assume things about a large audience.
This doesn't seem like the giant step I thought it was:
> Unlike the costly bleeding edge 2 and 3 nanometer chips made by giants like TSMC, TI’s chips are made on cheaper, legacy nodes: 45 to 130 nanometers.
What's worse:
> Making chips takes an immense amount of water, and about a quarter of Texas is in drought. Luckily, Sherman has water rights to nearby Lake Texoma.
Indeed lucky for TI, but probably not so for area residents.
Texas is a big state. It’s lazy journalism to generalize the state as droughty, which is implied by that sentence.
Lake Texoma has been hovering by the “full” mark pretty consistently for over 55 years. Recently, its water level has significantly declined to—wait for it—100% full!
If you monitor water maps, the east half of Texas’s water supplies don’t often get far outside of “full”.
More data: https://www.waterdatafortexas.org/reservoirs/individual/texo...
Also the water doesn’t disappear from the universe. It either evaporates or is pumped back out. People who lose their minds over water have been saying for 50 years we were about to enter a desert world
Clean water does disappear. Not from universe bit from places you need it most. That there's plenty of it in the air or sea or Saturn rings, is little consolation.
These fabs don’t evaporate the water though, they use it as process water, and then treat it to wastewater standards before discharging to the municipal wastewater system.
Assuming the municipality recycles their wastewater, which they would do in any drought prone region, this water will become clean water again.
It’s essentially a closed system.
> then treat it to wastewater standards before discharging to the municipal wastewater system.
Are there PFAS chemicals in the semiconductor photoresist and then in the water?
Do municipal wastewater standards require removal of PFAS?
Similarly, they're all okay with dumping water used to wash out coal plants into the rivers: https://www.google.com/search?q=trump+10%25+coal+plant+wash+...
> It’s essentially a closed system.
Source? Especially the claim that they are using external municipal wastewater treatment seems highly implausible, they are using nasty chemicals.
I read they either dump it into a local river while monitoring it or back to water treatment plant. Some datacenters have started reusing water but for some reason using it more than once is not appealing. Definitely the biggest problem is it’s not always a closed system and dumping it into a river potentially damages the river and makes the area lose their water. A percentage of the water is also lost via evaporation.
Source: https://dgtlinfra.com/data-center-water-usage/
Datacenters and fabs are different things.
It’s definitely not a closed system unless the water from the waste water treatment plant is pumped back upstream of the source of the municipal water which is not how most of these work.
guy has never seen 'Waterworld'
You should definitely bring this attitude into the housing affordability discussion - remind people that they indeed can find cheap housing in rural Africa.
The Aral Sea says otherwise.
Do you really think that that's what the authors were thinking? You can't accuse people of not using their brains when you haven't engaged yours.
Also you seem to be unaware of https://en.wikipedia.org/wiki/Overdrafting
There's lots of more boring chips in phones and mundane devices that are fine on older processes; things like audio stuff and battery controllers. If you want to get away from relying on China, then those chips are the ones that are in everything from your USB hubs, keyboards, phones, washing machines - everything Less sexy, but in many ways more critical to keeping stuff going.
The trailing edge is also where Moore's law in its original sense is still going strong. The fabrication processes get cheaper and more refined over time, while this is becoming a lot harder with new leading-edge nodes.
> Making chips takes an immense amount of water
This is misleading, modern chip fabs recycle their water basically 100%. It's not being evaporated for cooling like in some data center deployments, it's just part of the process.
That is not entirely accurate, as per the article:
> TI will use about 1,700 gallons of water per minute when the new Sherman fab is complete, with plans to recycle at least 50% of that
Just for context, around 200 times that amount simply evaporates from the Lake Texoma’s surface every minute.
Would this be the calculation you used? https://g.co/gemini/share/0639e6364e50
Basically yes. I googled average lake evaporation rates, and did the same multiplications as Gemini.
You mean you sis the math yourself? Novel idea these days. /s
That is of course the ideal, however local economic conditions (low/subsidized water prices, etc.) may dictate a different path by incentivizing another way of doing things (open cooling systems, etc.).
Why are they all building factories in places like Arizona and Texas?
Aren’t there plenty of areas where water is ample and land prices relatively cheap.
Land in Texas and Arizona *is* rather cheap.
They already have fabs nearby, and so, fab suppliers/services in the area.
They have a large employee pool in the vicinity, with modest salary expectations.
The southern US mostly doesn't have blizzards and ice that regularly knocks out electric supply lines, and/or makes it impossible for employees to get to/from work for days/weeks at a time.
Water is about the cheapest resource around. Even in water-stressed areas (while it may cost several times what it would in wetter parts of the US) it's still an insignificant expense for any industry that doesn't need an absolutely obscene continual supply.
Roof-top PV solar power works out a lot better in the southern US than further north.
> The southern US mostly doesn't have blizzards and ice that regularly knocks out electric supply lines
Except for when they do, then they are literally paralyzed until it melts. This may only happen once per decade, bot wow it sucks. One ice storm in Birmingham was so bad kids were stuck at school for 3-4 nights.
Also, very, VERY long histories of manufacturing there - TI has been in Arizona since 1969, for example.
Taxes and labour laws/no unions
Also limited to no environmental regulations and very… accommodating… administrations.
Republicans.
[dead]
What the fuck. This is terrible news.
The transfer of wealth from tax payers to tech giants has never been better. Enormous state subsidies. For doing something the free market would do if it made economic sense.
How many bailouts is Intel on now?
The idea of forcing a merger between Intel and TMSC at what could be called be barrel of the gun, would at least make it (indirectly) Taiwan that would have to take over paying for Intel.
You can’t build a modern fab (which this one doesn’t seem to be? But even old fabs are extremely expansive) without subsidies. They’re extremely expensive to build, and depreciate so quickly, no private party is willing to put the funds.
>For doing something the free market would do if it made economic sense.
It didn't make sense.
Bailout?
Merging Intel and TSMC?
Sorry I am not following.
Desperate measures to save Intel: US reportedly forcing TSMC to buy 49% stake in Intel to secure tariff relief for Taiwan
https://news.ycombinator.com/item?id=44801486
I would never have expected the US as the “capital of capitalism” to start managing their economy like a soviet country in 1975.
It's not socialism, it's corrupt state collaboration with capitalist corporations to suppress labor. That's fascism.
Technically then they aren’t wrong to equate this to the USSR - it’s important to remember that both china and Russia are “transitioning” to communism, although both are kind of stuck in the in between phase where the people who should be ushering in that change found instead that they can wield authoritarian power to enrich themselves and subjugate others.
Other side of things is you’re absolutely correct that this feels like a fascism, which is what this administration continues to feel like…over and over and over and over again
It’s sad to see the top 2 are going down similar economic routes TBH.
> corrupt state collaboration with capitalist corporations to suppress labor.
How does the Intel thing suppress labor?
Probably that it gives one company (TSMC) a bigger monopoly on chip fabrication? Not the original commenter though
I didn’t say socialism. There was that but late Soviet Union was mostly corruption and oligarchy… very similar to the current theme in the US gov.
The American people have been lied to for so long about what Socialism actually is to even be able to recognize it when it's happening right before their eyes.
No this is what happens in authoritarian states all the time. Dear leader has a conversation with a “friend” and suddenly that friend’s enemies have some new limitation from the government. Socialism takes the form of generally higher taxes on profits from personal or corporate income which pay for things for the whole society.
So seizing the means of production is now something that Authoritarian Capitalists do, and socialism is just when taxes. Gotcha.
The US hasn't been anything remotely akin to the capital of Capitalism in more than half a century. The US has a hyper regulated, highly taxed economy, to go with a gigantic government. The capital of Capitalism would be the opposite in most every respect. Go back to the late 19th century in the US as one example, if you want to see low regulation and low taxation.
> The US has a hyper regulated, highly taxed economy, to go with a gigantic government
This is pure hyperbole. I don’t know who told you that any of that is true but they’re taking advantage of your trust.
Holy shit, somehow I missed Trump trying to sell half of Intel to a Chinese company that doesn't want it. TSMC is a great company, but this sounds disharmonious with Trump's jingoist and xenophobic rhetoric.
The deal hasn't actually closed, though, has it? The linked page says, "However, TSMC is unlikely to agree to this proposal."
> somehow I missed Trump trying to sell half of Intel to a C̶h̶i̶n̶e̶s̶e̶ Taiwanese company that doesn't want it.
fixed your typo
Is the US willing (and able?) to go to war for Taiwan?
I certainly hope not; we have known for 80 years that a war between two nuclear powers would be a tragedy unprecedented in human history.
It used to be, though the previous Trump administration xenophobia set it on a policy of reducing the support for Taiwan.
Now their xenophobic tendencies against China are stronger.
For how long though? The end of an independent Taiwan can't be just dismissed as a possible outcome at this point, no?
You must mean "Taiwan, territory of China"
/s
Free market doesn’t plan for irrational wars.
They absolutely do.
You have likely not experienced it as governments and monetary systems have absorbed these risks (ie. they have moved them elsewhere, hidden away, until some major explosion happens).
> barrel of the gun
Ah, that locus whence political power comes?
MBAs forgot that when the carrot stops working the next step is the stick
And that the people responsible for their demise might leave with a parachute the stock holders won't (hi Boeing)
Chips are like agriculture: It does not matter that the home grown product can only compete with government support, you never want to be in a position where you cannot provide for your country’s minimal needs purely from local sources.
It depends what the minimal needs are, but basically the US has this for chips already. Yes the US only makes older chips, but in fact those are the ones that really matter, that you can’t do without. The ones that are in everything.
However making everything domestically just isn’t really viable for any country any more. The scale of our technological civilization and the diversity of goods and materials it depends on is more that any one continent can support, let alone one country.
It would be possible to collapse that down to one continent if absolutely necessary, but it would be incredibly economically painful and the US would need to give up on a lot of non-essentials and other priorities to devote resources to duplicating capacity that already exists elsewhere.
Buying rocks and exporting chips is fine. Of course it's not that extreme but the sentiment is there. It's about expertise capital and concentrating that domestically.
But at the expense of what? We’re not lopping off the rock farmer end of the economy to redirect efforts toward chips.
I am not sure about this. There was a chip shortage from Taiwan a few years ago that led to an auto shortage https://www.npr.org/2021/04/28/991270369/taiwan-races-to-rem...
I think it is not about one country doing to all. But technology transfer should happen within like minded countries and within a smaller circle. So may be two continents max.
Are there going to be Congressional hearings at least where Citizens can see a debate with witnesses testifying under oath? Will the Congressional research office publish statistics and data we can read together to debate?
Or will some backroom deal be made without any Congressional approval? I got a lot of questions and basically there is no way to get answers since Republicans gave the president: 1. unilateral authority to set tax rates (tariffs and illegal export levies) 1. impound Congressionally appropriated funds on an ad-hoc basis 1. unilaterally close down federal agencies created by Congress
I refuse to debate the merits of this industrial policy while the above three things are true. There is nothing to debate here because no one outside the administration's inner circle (industrialists bribing) who knows what is going on.
I feel like my 7th grade Civics class has been a complete waste. Or, it lets me see what is happening in the correct light.
[flagged]
Shame you've been downvoted. Regarding federal agencies, the Constitution indeed means what it says when it says "The executive Power shall be vested in a President of the United States of America." If some agency can wield executive power, it is under the exclusive control of the president. Looking forward to the likely overturning of Humphrey's Executor soon. It's gone in all but name anyway.
Perhaps this will spur some future Congress to take back what they've delegated (like sweeping tariff power)
This is a completely meaningless statement. It sounds good, but contains no actual argument. why are chips like agriculture? What makes that statement not apply to literally every product category that exists?
Agriculture is self-evident, we need food on a daily basis to live. I don’t need three square microprocessors a day.
Just like you don’t want to wait many months to tool up to make food if there is conflict…
You don’t want to wait many years to tool up to make chips that go in weapons, either.
Else, you risk being structurally defeated very early, and this leads to others concluding that they can take you on. It undermines other preparations.
Yes, we understand that, but it’s not a good comparison.
Producing chips and agricultural goods are fundamentally different, starting with limits like climate, seasonality, and land. At most, you could compare it to mining minerals for chip production
I don't think they're referring at all to any similarities in how they're produced.
It's more to do with similarities in the consequences of not have any local supply in times of emergency.
The only things that matters for this consideration are:
A. How bad is the consequence of not having them?
B. How long does it take to tool up domestic production if there's a supply disruption?
C. How long can you tolerate the shortage of them?
D. How easy is it for a disruption or disaster to disrupt international supply?
The answers to A and D are fairly different, but both are "bad enough" to trigger this kind of thinking. The ratio of B to C are probably similar between food and critical integrated circuits.
Farming equipment needs microchips, as does practically any modern device, vehicle, machine, etc. If you don't have domestic microchip production then pretty much everything else you make will have foreign dependencies that could be embargoed.
Microchips for farming are made in several counties including US. There are highly specific processors, like those required for ML, but that’s not what farming can live without.
Having narratives as broad as yours, without any evidence and reasoning, is the fastest way to achieve Russia-level disaster with government monopoly in the name of “security”.
The only sense it makes if you want corruption at state level and to go on war with at least half of the world.
Good luck with that.
Hello from Moscow.
I think there's surely a little bit of a middle ground?
Subsidizing and protecting local production has a cost: there are gains from trade and operating less efficiently makes everyone's standard of living goes down.
But you don't need to go 100% into protectionism. You can ensure that you have enough domestic industry to
- better weather changes and disruptions
- have a starting knowledge and capital base to use to ramp up production if you have to.
Just having a few percent of domestic demand for advanced ICs covered by local production is enough to make the worst international shocks much softer. Critical needs can still get filled. So a certain amount of subsidy can make sense.
The Internet is made of microprocessors. Maybe we don't need the internet to live, but it seems pretty important these days!
Of course we'd think that, we're on the Internet, discussing it. The people who aren't, aren't here to rebut that. I met this woman who lives part-time in Romania, and was saying that they don't have smartphones still, which is why she has to spend her time in Romania, because she can't just have this particular subset of elderly people in Romania get information from her off of Instagram.
Sure. I could live in the middle of nowhere in the US and communicate exclusively via written letter…
…but the USPS will still use machines with microprocessors to sort that mail.
>It does not matter that the home grown product can only compete with government support,
Even that isn't a given, because unless you have amassed a certain amount of technocratic and governmental competency chances are it can't compete even with government support and you just produce crony dysfunctional companies.
And of course there's economic trade offs. If you're politically ordering your economy to make chips, it doesn't make something else, and whatever it was making and trading for chips it was better at, and so you get fewer chips, that's comparative advantage. Industrial policy (and tariffs) do not increase aggregate production, they reduce it. And given that the circle of items you "can't do without" seems to be a bit of a moving target these days, at some point you're actually more brittle because you've replaced large chunks of the market with state production.
People take it as an assumption that cronyism will always happen if the government invests too strongly/consistently in a certain thing. But cronyism is a policy and structural failure, it's usually because the incentives for the different parties involved encourage it to happen. Institutions can be designed carefully if policymakers actually want to do it.
USPS is a great example of an organization that's managed to largely avoid this. Whenever you mention that people crawl out of the woodwork to complain about the 7 different times they lost their package, but their logistics at scale is still unmatched by the private sector, while also not completely negating the value of private sector alternatives (which so often is argued would happen if the government actively started doing anything new).
>People take it as an assumption that cronyism will always happen if the government invests too strongly/consistently in a certain thing.
No, it's the opposite. It's a conclusion taken from empirical evidence looking at success rates in the real world. As the US has engaged on its most recent bout of industrial policy, industrial activity has declined, not gone up. Again, this is expected. If you through protectionism make chips and steel in the US inefficiently, everyone in the US using those products as inputs suffers.
There's some limited cases like developing countries engaging in catch up growth , but there's virtually no evidence for effectiveness of these policies in cutting edge technologies, which isn't surprising because they by definition tend to rely on supply chains and knowledge sourced from all across the globe, and to this protectionism is particularly disruptive. And even in developing nations most of the time practices like ISI (import substitution industrialization) fail, it's devastated Latin America in the 1950s to 1980s.
I don't know what USPS proves in this case because it's sustained by a government monopoly. Obviously you could if you wanted privatize mail delivery.
It is established economic theory not just some opinion . Constraints in form of regulations, subsidies, tariffs by government intervention will produce inefficiencies.
This is the core economic philosophy behind both the modern centre right[2] i.e. neoliberal democrats (80s-today) and the right wing republican conservatives (pre MAGA -2016).[1]
Even communists would agree this is correct, it is understood in different levels since Adam Smith invisible hand, from far left to centre left just posit that different polices that benefit the people not the economy are worth the cost .
That is actually true for all other groups except anarchists and libertarians all agree government intervention is needed in some form, they just disagree on how .
——-
Postal services anywhere in the world are inefficient by design. They are governed by universal service obligation principles not efficiency same with telecom providers and other utilities.
The question then becomes how much inefficiency is acceptable given the objectives . There are no right answers, you could have competing private couriers who dump unprofitable low density routes to USPS while serving profitable ones themselves (UPS, fedEX, Amazon ) or you could have gigantic postal service which is sole delivery provider socialist style, there are going to be some problems either way.
It is just matter of preference which set of trade offs we are okay with .
—-
[1] Post MAGA it is a populist party economically there is no fixed ideology to be characterized
[2] the centre right label is economic and technical to contrast say the socialist left party of FDR/New Deal 1930-70s, not meant to offend.
But that's exactly the thing, where we care about an industrial capacity for its strategic reasons rather than because we're trying to maximize efficiency of revenue extraction. Pure market efficiency is great if you're trying to maximize profits, but that's not necessarily the goal here because the benefits are indirect, just as they are with universal postal service (or universal healthcare, etc).
Maximize economic output not necessarily profit outside of that I agree with your comment, nobody is questioning whether we should have other goals.
The problem is however first can we all define the goals consistently and agree on the definition.
Next can we measure the value/benefits of the goal. This is typically subjective so everybody paints the picture best suited to their preconceived ideas.
Finally perception drives the political will for a social program. UBI style programs or just giving money in welfare instead of in-kind benefits for food or healthcare etc or rehabilitation over punishment are more efficient and while shown to work, will rarely get political support because of how they are perceived.
Given that we all have different understanding of the goals, measure them differently and have different moralities on what should be done as society (for welfare, crime etc) and have varying degrees of influence over policy what can we do to approach a problem better
You should tell Taiwan that.
This is true. Also true for oil production.
How? What if we just decide we will never go to war with Mexico or Canada and get comfortable with the idea of importing from our allies? There is no serious future risk from doing that
> if we just decide we will never go to war with...
That's hubris. Although the US does indulge elective wars, one does not always get to choose with whom one will war.
Okay so make a prediction. What is the probability that Canada or Mexico will declare war on the United States in the next 100 years?
100 years ago the British Empire ruled the world, now it's a small island you don't hear much about... The US is only about 250 years old, I'd be cautious about predicting the future
You're off by about 150 years. 100 years ago was 1925, the British has already largely collapsed losing the US, Canada, Australia. 100 years isn't forever but it's a long time
And we haven't had any serious threats from Canada since 1812. I think the most reasonable estimate is 100-200 years
~1920 is the peak of the British Empire in term of territory, anyways, the details are meaningless, what matters is that things move fast and just because you're at the top of your game right now doesn't mean you'll be in the same position in 100 years
I could also take the example of world wars, in France ww1 was deemed "la der des ders", which meant "the very last war" or "the war to end all wars", well 20 years later we were at it again
Or simply look at China, you don't even have to go back 100 years in the past to see drastic changes.
That's only the case if you include Canada and Australia, which were functionally independent at that point.
I'm not claiming nobody will invade France or Taiwan in the next 100 years, I'm claiming that the US is special. We haven't been invaded since 1812 and haven't really been attacked since 1941. It's reasonable to predict we won't be invaded or go to war with our neighbors for 100 years since it hasn't happened for 213 years
This strikes me as hubris in the extreme.
My own death has not yet been a problem for me, but I can safely assume it will be.
Because everyone dies. If everyone lived to be 1000 you'd be wrong to worry about dying in 100 years
Surely based on history the odds of a conflict between neighbouring countries increases with time passed.
No, based on the history of conflict we can say that the more time that passes with neighbors not invading one another, the less likely they will in the future
It sounds like you’re making the assumption that things will remain static because the alternative is unfathomable to even consider.
100 years? Only a fool would attempt that. 100 years from now Uzbekistan could be fighting Brazil in orbit around Venus.
Definitely non-zero. If you were Mexican or Canadian you would not take the bet on even 20 years right now so who would bet on 100?
If we include the idea that either one of them is allied to a major power at war with the US over a hundred year horizon, right now that looks pretty likely, and arguably is one of the things the current US admin are trying to stop before it becomes inevitable.
This is the wildest interpretation I've ever heard about what's going on right now. Maybe look up the term "soft power" and why it works.
The US admin is very clearly pushing us in the opposite direction. You believe that Trump's actions make war with Canada less likely? What's the mechanistic explanation?
Didn't the US join 2 world wars it did not start (or was involved in starting) in the last ~100 years?
Not against our neighbors
If the US didn't enter WW2, we would have new neighbors. The technology has made the world a small neighborhood.
So the argument is that the neighbors will never ally and be involved in a war that puts US on the opposite side? What is the argument for the sovereign neighbors to always be neutral or on US side come what may?
Yes, that's right. This has been the case for over 200 years so I think it's reasonable it will continue to be true for at least another 100
Germany was trying to get Mexico to join them for wwi, with enticement of getting back the land they lost 70 years prior. That loss was immense for Mexico in land, pride, and economics
Mexico was dealing with its own internal issues (revolution) which made it difficult for them
A slight turn of events and the us would have a huge southern front I deal with, and a base of support for disaffected native Americans and African Americans.
Russia also dropped out of Wwi due to its intern revolution
It is easy to look back and see manifest destiny as a given. There were a lot of contingencies
The fact that they're much smaller than the US and right there. Both would have their key cities flattened within 30 days.
Mexican warlords (which we colloquially call "cartels") are fighting a small war with the US right now and have been for many years.
What does that mean? No they aren't? When is the last time cartels attacked the American military?
You understand that for basically every violent entity in the world (except China and Russia), avoiding the US Mil is their best path to success and continued survival?
"War" as a concept does not have to include two militaries fighting each other.
If the cartels are waging a war against the US, so is a large number of American oligarchs. When will they start being on the receiving end of drone strikes and special operations?
When you stretch the definition of war to absurdity, so can I.
It's an extension of a peace through strength philosophy. If you lose your critical sovereign capabilities you become weak and vulnerable. You no longer get to decide who you do and don't "never go to war with".
Yes I understand the reasoning, it's just obviously wrong. This doesn't happen and hasn't happened in hundreds of years. This is not why people get invaded, otherwise Switzerland would have been invaded many times over the last century
That's exactly why Switzerland didn't got invaded in the last century. Every side depended on them for money supply, so they couldn't risk bringing the war to Switzerland. Also it had a track record of staying neutral, so they didn't feared it may pick a side, because it largely made money from selling stuff from both sides.
This kind of answers your own question: the reality is you are only a reliable US ally [1] if you can hold them by the balls TSMC style. Given that countries go to great lengths to develop and maintain such dependencies. Canada's current weakness is at least in part because it has failed to do so.
[1] Edit to add: This was/is poorly worded - I mean that the US will only guarantee that you remain an ally while they are in some sense dependent on you, and while doing so they may work to break that dependence, which you may interpret as them trying to abandon you.
You are claiming that Canada will stop exporting to the US in the future? Populism does the opposite, they would likely stop importing. There's almost no risk to us that they would not export chips
My claim is Canada has no unique product that it offers that the US is sufficiently conspicuously dependent on for it to guarantee any real sense of true independence, and so it finds itself subject to the whims of a foreign state.
Right now there is a very large Canadian boycott of US products, services and tourism. I also had to explain to a client this week that because of the import tariffs on Chinese goods to the US the US assembled products are now no longer competitive with alternatives. The fact you were seemingly unaware of this kind of demonstrates the level of effect of it.
You are arguing my point. Canada may stop importing from us but will never stop exporting. There are no incentives for that and never will be.
We import tons of food and energy from them and have no alternative on time scales or 10 years
If we imported chips from Canada, that supply chain would be safe for at least 50 years, probably hundreds
The Nintendo Wii SoC was actually fabbed in Canada and exported, but that facility has changed into something slightly different because the whole east coast/hudson river valley fab world went sideways a while ago.
> We import tons of food and energy from them and have no alternative on time scales or 10 years
More importantly for Canadians that food or energy has no alternative competing market to sell into. Consequently the Canadians are totally dependent on the US market to even set the price of it. This applies to many other sectors as well.
Canada is currently having a huge desperate push to export to non US markets because of the levels of uncertainty that have been created. And I say this as someone not totally dismissive of the US position, but they need to do a far better job of bringing their allies into the tent with them.
Historically there have been plenty of incentives to stop exporting. Its called an embargo. Usually in an attempt to get the host country to change foreign policy, though I can't think of any situation where it actually worked. Examples: Napoleon's "Continental System" against Great Brtain, US oil embargo against Japan prior to WWII, Confederate States of America cotton embargo against the UK during the early years of the American Civil War
Yes it's definitely possible, but very rare and as you pointed out (especially in the confederacy's case) it usually harms the exporter much more than it helps.
I'd say that in any case of a serious Canadian export embargo, it will have been in retaliation to US trade policy or US invasion, not the other way around.
We had essentially no risk that Canada would embargo us, there was no possibility of this happening for the last 150 years until we became the aggressors
At any time, Canada could decide to stop subsidizing its uncompetitive chip makers for the same reason so many people in this thread want the US to do, and the US would then become dependent on someone else who might be their enemy (eg. Chinese occupied Taiwan or China itself).
> go to war
It won't be war. It'll be one-sided trade deals [1,2], and a slow erosion of economic and political sovereignty, culminating in a puppet state.
[1] https://en.wikipedia.org/wiki/Canada-China_Promotion_and_Rec...
[2] https://www.cbc.ca/news/canada/fipa-agreement-with-china-wha...
One sided trade deals in which they continue exporting to us and import less
There is plenty of risk that our neighbors stop importing and almost no risk they stop exporting
Russia has been totally isolated from global market and they are still producing hundreds of cruise missiles a month just fine. America could easily figure it out long enough to survive a few yrs of conflict. Even stockpiling 10yr old chips would be good enough 95% of military industry. Then emergency investment and smuggling will cover the rest.
Meanwhile cutting off your markets and wasting hundreds of billions on a long term bet with a small probability that another global war will happen is pretty dumb financial thinking.
Tariffs and corporate welfare will actively make a country poorer and create unproductive zombie markets while raising taxes on everyone. Not to mention diverting budgets and new revenue away from actual national security investment.
Up until relatively recently we were the SOTA and #1 semiconductor exporter. When people talked about the "american manufacturing sector" a significant portion of it was actually that.
Those foundries didn't go away, they're still manufacturing with the same capabilities they used to (and they're much cheaper now since they're competing with the better ones in Taiwan.)
It's good to hear TI hasn't given up on high performance SOCs as it was beginning to look like they had. But most of this stuff is still here. Freescale and many other American companies are still making the same (better even) chips they always have which is more than enough for cruise missiles (more than enough for decent PDAs and smartphones really) even without "stockpiling."
Yes and those Russian semiconductors aren't sophisticated or high end at all which supports my point. The military isn't making cellphones, they make missiles and fly jets with decade old computer chips. A few years of war with China is not going to magically eliminate all computer chips. The global market will still exist in some form and wars are far more motivating than a few government grants.
Not to mention China (and/or Taiwan) is still going to want to sell to someone to survive, and those countries can smuggle them into America - just like Russia does for it's drone industry and oil. America is much more capable in that regards with NATO and it's huge purchasing power.
I still think TI and Apple should be investing in foundries and domestically. It should just make sense as a business otherwise it's going to be a very expensive embarrassment.
Everyone talks about how the CHIPS, IRA, etc. purposely invested more in red states, but they don't talk about the mechanisms that made that happen.
I would be curious
I don't know if it was on purpose or not, but I have heard it said more than once that Republican led states are able to greenlight projects faster, are more business-friendly environments, and generally have less red tape compared to Democrat led states. Love it or hate it, but greenlighting projects is a big component in allocating funds.
On the other hand, if they were so much better, shouldn't those "red states" be the ones with the much better economies?
I'm looking at China and what environmental price in the form of polluted land they pay and will be paying for a very long time. The big problems in Western countries also all originated in times with less or no regulation of such things. Just because that's not in the headlines all the time does not mean the problem is any less while the public is not paying attention.
When "it works" and overall success is the only criteria, the Vikings and Mongols surely count as extremely successful. Regulation is meant to take the price into account, in the cases of those two peoples millions of dead and a lot of pillaging and conquering. Regulation is definitely a burden, if you don't have to care about anything but the bottom line it's much easier.
Definitely not true. Right now Red states are openly attacking businesses that don't agree with the prevailing ideology. In Florida, the governor tried to destroy the state's biggest employer. In Texas they have been trying to prosecute out of state businesses. Alabama has more taxes on businesses than California.
Red states just say they're better for business.
I thought it is because they have worst economies and it is always an attempt to prop them up.
The other reason is that while republican party is purposefully trying to destroy economies of blue states, democrats were not trying to purposefully destroy economy of red states.
This might be true, it’s definitely repeated, but it’s generally not the real reason.
The real answer is just politics. Blue states have (generally) healthy economies, with a variety of economic actors and many businesses. Businesses often will shop around various states to build a factory looking for tax cuts. The politician can be associated with new jobs, and the business gets a discount, so it appears to be a win-win (if you ignore the lack of tax revenue). No one needs a tax break to start a business in NYC, LA, nor Silicon Valley, so you don’t hear about all the businesses that open there.
Nationally, policies like the IRA are big boosters for the economy, and democrats are focused on getting it done because it’s good for society and the national priorities. They won’t focus on where the money goes, and will allow the money to go to run down republican states as economic stimulus. But you’ll notice it’s usually capital intense factories that end up in these situations, not white collar jobs.
That’s not the reason national projects get there.
Federal funds get funneled to red states to secure the votes of their representatives.
Voters in Red states destroyed any motivation by foreign companies to build factories there (tariffs) and Democrats to focus infrastructure spending there (CHIPS, IRA being worthless in getting votes).
And now Republicans are ok with impounding Congressionally assigned funds.
So I guess outside of facebook brain rot and doom-scrolling, there is no way to convince voters to vote with policy (requires attention span of more than 30 seconds and high school level reading skills).
In Michigan, there were several IRA-catalyzed investments worth many billions for battery facilities or various components — and the announcements were met with vehement protests because some of the companies involved were Korean or Chinese. It didnt matter that they were going to create many thousand good jobs and use tons of local construction, the weird conspiratorial bigotry was just too strong (fanned by the now-President and VP).
Just no idea where to go from here. Feels really dire to be honest.
https://www.politico.com/newsletters/power-switch/2024/09/13...
At some point we need to cut these people loose, understand they’re beyond saving, and move on with things.
Oh how I wish we would be able to split the country and not have to deal with the deplorables any longer. Unfortunately we would also have to split with the deplorables the national debt and the nuclear arsenal. Imagine what would happen then…
While that sounds like a solution it’s of the “simple but wrong” kind. Just like you, they are product of propaganda, though theirs is less based in reality than yours. If you don’t counteract the source strategy (like the lack of education, biased fox “news” and adjacent media), nothing prevents it from being repeated in the resulting halves of your solution, Zeno style.
Honestly, no. That’s enlightened centerist nonsense. Being for things like freedom of speech, free and fair elections, a non-totalitarian form of government, and being against economic suicide, masked goon squads kidnapping people off the streets is not some sort of insane propaganda that I should be questioning.
They still haven’t fully torn down a Texas Instruments plant in Stafford, TX that closed a decade ago. But now we’ve got an In and Out and Costco on the former campus.
Is fabric here a mistranslation of "fabrica"? Correct translation would be "factory."
Probably it should be “fabrication plant”.
Chip manufacturing factories are traditionally called fabs, short for that.
Title reads very weird in American English, I'd expect "factory" instead of "fabric" (even though in Dutch, I'm bilingual, "fabriek" means factory).
The actual title of the article used the word "project" not "fabric"
Any reason to use "fabric" here?
> Making chips takes an immense amount of water, and about a quarter of Texas is in drought. Luckily, Sherman has water rights to nearby Lake Texoma. “It was about acquiring more rights, ramping up our production and being able to provide for the mass quantities of water it takes to run a semiconductor facility,” said Teamann, adding that the fab has almost doubled the amount of water Sherman uses.
Luckily…
“As for federal support, TI got $1.6 billion of CHIPS Act funding, and a whopping 35% investment tax credit from Trump’s big bill passed in July.”
So how much ownership is the US gov gonna get in this one?
TI is a going concern, no bailout necessary
Intel is not getting any money that legislation (law) had not already allocated to them. The transfer of shares to the Gov is just a shakedown.
Just? You mean they should have received the stupidly large sum of money without anything for the tax payer?
Yes? That's what grants are. The government is buying a domestic chip industry with that money.
Why shouldn’t they get some equity in return for giving money to a for profit company?
I'm a bit biased because my home state is Ohio and they have it in their constitution that the state can't have a stake in any private company and can't even lend credit to any private company. And this amendment was written in blood, in the early 19th century the state nearly bankrupted itself investing in and taking stake in private companies.
* The state can't risk taxpayer money on ventures that might not pay off or lose them money. How the state "gets around this" is by issuing zero recourse loans. The advantage is that when economic development money is handed out there's not an asset on the balance sheet. It's treated like it was spent. The value the state gets from spending the money has to be independently worth it for taxpayer without considering financial returns.
* It eliminates a whole category of conflicts of interests where the government will get squeamish regulating or punishing bad behavior because it would hurt the taxpayers' investment.
* It also eliminates vectors for corruption as well as the negative effects of the government having direct influence over specific businesses. No backdoor regulations from the state's ownership stake that don't go through the legislature.
So I'm very heavily in the camp that government shouldn't ever be allowed to have stake in any private company. The line between government and private enterprise should be the wall this admin likes to talk about. I certainly didn't expect it would be republicans I would be trying to convince that state ownership of business is a bad thing.
> I'm a bit biased because my home state is Ohio and they have it in their constitution that the state can't have a stake in any private company and can't even lend credit to any private company.
The Ohio government works around this by cooperating with private “development corporations.”
I’m not a fan, but it’s been going on for a very long time.
It’s not all that different from development tax incentives which are very common.
I think the grants had some requirements to build up domestic production.
They should be doing neither.
America is the last place that is short of capital for industry investments where it requires gov taxes going to it, they have a huge domestic financial market and tons of foreign investment (but those require legitimate plans, not national security woo woo). This is just propping up weak megacorp industry like they did with Boeing, instead of fostering real progress.
Sure.. calculators and MSP430s for remote power meters are keeping TI from closing up shop, but TI doesn't have the capitalization structure to bring up a fab for the types of chips people say they want. I mean sure... If you want to make 28nm chips, they're fine, and you can do a lot with 1 and 2 GHz parts, but... We keep saying we want to make the chips in the states that they're making in Shenzhen and Taipei... And honestly, a $1.6B grant from daddy warbucks may not be enough to prevent TI from taking the money and dropping out of the program in a few years.
And this comes from a place of love... My family's been invested in GSI for almost 100 years
> calculators and MSP430s for remote power meters are keeping TI from closing up shop
calculators have consistently been a minor percentage of TI’s business (~5% of profits per source below). I doubt MSP430s in particular amount to a huge percentage either
one random source: https://www.meta-calculator.com/blog/ti-graphing-calculator-... (this is pretty easy info to find)
I can assure you that TI's margins on calculators and MSP430s are much higher than their margins on DAC*s.
Depends on which line of DACs. And calculators are an almost irrelevant amount of TIs revenue. They don’t report it individually, but it’s categorized under the miscellaneous “other” bucket which is only 6% of their business and includes DLP and “other charges” related to M&A. $947 million with all those other things means you’re talking about probably 100-300 million in revenue. There’s other businesses within TI that do more revenue than that by themselves.
Saying "TI makes money on calculators" does not mean "TI does not make money on DLPs."
Also... revenue, profit and margin are all different things.
I’m not saying that at all. I’m saying it’s nearly meaningless in the grand scheme of their overall revenue. They make roughly 50% margin on their $17 billion in revenue right now. If somehow calculators cost nothing (100% margin) it wouldn’t make a meaningful difference on their overall margin.
Saying "TI makes money on calculators" is a pretty misleading statement outside of any other context. Its a tiny part of their profits and revenues. It's like suggesting McDonalds is an ice cream shop. Sure it's on the menu and they make a profit on it but it's a small side business after selling burgers and fries.
So TI is losing money on every calculator they sell? News to me.
> Sure it's on the menu and they make a profit on it
My point is, the statement "calculators and MSP430s for remote power meters are keeping TI from closing up shop" isn't based in reality. Both of these products are tiny parts of their business. If that's all you know of TI, you don't know TI. Like thinking McDonald's is an ice cream shop and completely being ignorant of the burgers and fries, saying those ice cream cones are keeping McDonald's from closing up shop.
Even for the MSP430, it's a small product line of their wireless and microcontroller products.
I’m really confused how you got that from the comment you’re replying to, and why you’re continuing to defend misinformation you’re spreading in the original comment. you implied TI primarily makes money from calculators and MSP430s. this is easily provably false
the person above made an analogy —- they didn’t claim TI loses money per calculator
Microcontrollers and calculators are a small part of TI's revenues. Most (>70%) of their revenues come from analog devices like amplifiers, DC-DC converters, ADC/DACs, and things like that.
They make important chips and many top of the line products of their segments but they're not things like server grade CPUs or GPUs.
Yes. They make most of their money from low margin parts. That's not as good a story as you might think it is. Though... making money is certainly better than not making money. And yes, they have a decent mixed signal story.
But... everyone seems to think TI will be competing with TSMC's and Samsung's small-node parts. And they probably could, but they would need to a) build a fab that can make 5 or 3nm parts and b) build a sales channel for new parts. I was alive in the 2000s so remember TI doing an exceptionally poor job of step b.
Their analog division has >50% margin, a good bit more margin than their MSP430's and graphing calculators. That's not far off from TSMC's overall margin.
It's a better story than your misleading statements acting like TI only makes calculators and old microprocessors and flat out inaccurate ones about profit margins.
I only saw their OPM broken out by division. OPM was around 37-38% in 2025Q2. Do you have numbers for NPM broken out by division? But yes, if they could get volumes like analog or mixed signal with margins like "other" or "embedded" that would be pretty awesome.
What does GSI refer to? Googling did not lead to any obvious results.
The original name of Texas Instruments (Geophysical Service Inc.).
TI's older name: Geophysical Service Inc.
also... CNBC put the video up on YouTube, if that's your platform preference:
https://youtu.be/0ybDbElLnE0
will this fix xkcd 768 problem, tho?
Do teenagers even use those calculators anymore?
as I understand it (from outside), such calculators are part of american curriculum and is pretty much demanded on SAT and similar school-independent testing
The new calculators are an abomination of reverse-compatibility and have a separate python coprocessor tacked on. But now it has a color screen
[flagged]
What exactly are you talking about?
The movie Trading Places. It's a comedy and worth a watch.
I'm guessing the commenter above was asking what trading places and commodities futures have to do with the referenced article. I'm trying to figure out if you view Trump + cronies as the good guys in Trading Places (Dan Aykroyd + Eddie Murphy) or the bad guys (Ralph Bellamy + Don Ameche). Or if you think the old institutional guys conspiring to ruin young Dan Aykroyd's life over a bet were the bad guys or the good guys? I'm not trying to be snarky, it's just that there are a lot of opinions offered on this site and we don't all view the world the same way.
The use of "cronies" is generally pejorative!
There are a great number of opinions expressed on this site. I generally think "cronies" is pejorative and you generally think "cronies" is pejorative, but I take it on faith that there is at least one person out there thinking "hunh... cronies? yeah. That sounds good. I don't know why people get so upset when I say that."
And I do not know whether the person using it in a sentence is using it pejoratively. And heck, there are PLENTY of heresies I subscribe to: testing software is more often useful than it isn't, the web was done by amateurs, adding comments to code documenting the programmers' intent is useful, the rust borrow checker should have just been an add-on tool for C++, etc. So it may be ME who is against the grain on this one.
Which is to say... just be careful out there when you assume things about a large audience.
It should be self-evident which side he's on if one were to look at his hustles.