8 ms·
Man, I wish I too shared such optimism in EU or other massive multi government bureaucratic organizations to pull something off like this in 2021. But sorry I s
by dmix 6y ago
Man, I wish I too shared such optimism in EU or other massive multi government bureaucratic organizations to pull something off like this in 2021. But sorry I simply don’t.
From my experience with gov trying to do tech investment it’s going to go into the pockets of late career overpaid ‘executives’ with ‘industry experience’ and tons of (also overpaid and underqualified) contractors doing the same thing. Not the brilliant engineers you imagine are going to run it.
I mean I hope it goes to the engineers and smart dudes (who doesn’t want another successful semi competitor and 2nm?) and not the ones who know how to play the political game the best but I’m highly highly skeptical.
We’ll see in a few years I guess.
And I say that fully understanding ASML and the value they currently provide. But let’s hope who ever doles this out understands it as well as (maybe) half the people in this thread.
But even ignoring that giant elephant in the room, it adds the critical question why should ASML care to give special treatment to this one project? They are already killing it globally. Just because they operate in EU and this offers shit loads of money to potentially fleece?
- 11thEarlOfMar 6y agoI don't think this requires anything special from ASML. They'd sell systems to the TSMC/EU partnership. The key is that the process engineers in the fab would learn how to develop processes for the leading edge geometries. To some extent, ASML engineers would know this, too. However, the secret recipes (literally) of chip production are so closely guarded that even the equipment companies for the tools they run on are not permitted to see them. I completely agree that consortium-based organizations are less able to respond and react to changes or issues. That's why it needs to be treated as a strategic (defense) matter rather than a government/industry partnership, a-la Sematech in the US.
- supernova87a 6y agoI think I share your concern -- it's a question of "is a top down order to develop some technology (with required investment to 'force' it to happen) the best way to achieve that?" What often ends up happening is that because it's so "strategically important" and so much money has been directed towards it, you are obligated to have expensive ("experienced") people hired based on past reputation -- which may not be applicable to new technology, roadmaps get laid out which everyone has to say they're meeting successfully, and then at the end you find out somehow that it didn't produce what you wanted. Versus, a more bottoms-up approach where you hopefully create the conditions that incentivize or make possible some technologies and technologists to succeed -- without knowing fully who exactly or what exactly. I'm sure that both have some inefficiencies, but the 1st method is sure to pay the people who have already been successful. But then again, maybe for a relatively incremental advance in some technology, there are fewer unknowns and method #1 works ok.
- emn13 6y agoConversely, for the really novel stuff government investment in research has a pretty good track record, really - I mean, they don't always turn it into a product, but who cares? Hoping for a corporation to fund medium or long-horizon research seems unwise. Even if they do, they're not going to do it in a way that's very efficient (i.e. they'll inefficiently invest in stuff with a high moat, rather than broad applicability).
- bojan 6y agoConsidering what it has to do and support, the EU has surprisingly little bureaucracy, although obviously it's huge in absolute terms.
- AdrianB1 6y agoA relative is a middle manager in one of the biggest tech organizations in EU; only 60% of their huge organization is doing actual work and that 60% includes the management of the actual experts; the rest are "support functions" with very well paid positions for people appointed by the member states with strong political connections and no relevant skills vs. hired with interviews for the 60%. But even in the 60% there are a lot that are not meeting the skills required, but they are protected from any action by labor laws and unions, including forced salary raised based on how long they worked there, with no performance indicators. In practical terms, over 50% of that large (many thousands) organization should not exist. Another 50% of what is left is only needed because EU bureaucracy requires them to cover work that can be avoided. That qualifies for "too much bureaucracy".
- egman_ekki 6y ago> This book contains three basic ideas. The first is an output-oriented approach to management. That is to say, we apply some of the principles and the discipline of the most output-oriented of endeavors—manufacturing—to other forms of business enterprise, including most emphatically the work of managers. Consider Intel, which is a true manufacturing and production company, making highly complex silicon chips as well as computer-like products built from them. Our company now has over thirty thousand employees. Of these, about 25 percent actually work to make the products. Another 25 percent help them as they supervise the personnel, maintain the machines, and engineer and improve the manufacturing process. Another 25 percent work in administration, where they schedule production, keep personnel records, send bills to our customers, and pay our suppliers. Finally, the remaining 25 percent design new products, take them to the marketplace, sell them, and service them after the sale. Andy Grove, High Output Management, about Intel (at the time when they were successful) I think there are many, many (not directly) productive functions that successful business needs to have and engineers often underestimate those (I certainly know I often do).
- mafribe 6y agoAs far as I understand, ASML is not the creation of government funding, but instead a spinout of Philip and ASMI, two successful Dutch electronics companies. ASML was successful in a competitive market, dominated by Japanese companies pretty quickly. In other words, very different model from what the EU is currently trying.
- 11thEarlOfMar 6y agoYou are correct. The discussion about ASML is relevant in that state of the art chips cannot be produced without one of their TwinScan lithography tools. ASML is the only company that can build tools with 7nm and below capability. In the context of the EU developing the ability to fabricate state-of-the-art chips, ASML would sell them the TwinScan tools they needed, which are just one species of about 40 that are used for chip fabrication.
- mafribe 6y agoASML would also sell those tools without prodding by the EU. ASML would dearly love to sell them to SMIC and other Chinese foundry too, if the US would let them. Lithography machines are a functioning market that is interfered with for geo-strategic reasons.
- jjeaff 6y agoIf ASML is the only company that can do what they do and anyone else would need decades to catch up, how are they not the most profitable company in the world? All of the world's most profitable companies depend dearly on chips. If they are literally the only company that can provide this critical part of the chip production, why aren't they taking a larger portion of the profit?
- roenxi 6y agoAt some point their customers will stop worrying about using 7nm and go back to double-digit nanometres. CPU tech from the 2000s is adequate for most workloads, let alone stuff from 2010. I'm sure there is a lot of stuff where 7nm is necessary to get certain outcomes, but it is a small market compared to what could be reasonably done with older process technologies.
- 908B64B197 6y ago> Not the brilliant engineers you imagine are going to run it. I have no doubts brilliant European engineers are going to work on the hard problems of shipping a next gen chip process and design... In America! What the EU should do is earmark some of that money to be directly paid as engineering salaries. Of course they won't: that would reduce the piece of the pie of all the non-technical managers and bureaucrats!
- csunbird 6y ago> I have no doubts brilliant European engineers are going to work on the hard problems of shipping a next gen chip process and design... In America! Exactly. The salaries in EU are pitiful.
- usr1106 6y agoTech salaries in the US are not sustainable. We don't have enough planets to run economies and waste natural resources at that velocity. Well, probably the same is true for the "pitiful" salaries in the EU. Just to a lesser degree.
- neltnerb 6y agoYet tech salaries in Vancouver combined with real estate values make it so that most tech jobs are barely living wage, at least when my partner and I tried hard to make it work some years ago. Even with universal health care (which wasn't quite what they have) it couldn't possibly make up the difference for someone with college debt. I'd argue that if you got a CS degree from an American college you're probably just forced to work in the US simply because the other prospective employees can accept a lower salary due to a lower debt load. Maybe the best of the best can find a way to earn American salaries overseas but I really don't think that's as common as the HN crowd might assume.
- 908B64B197 6y agoThe situation in Vancouver is caused by out of control immigration quotas. Real estate there is a way to park assets in a safe jurisdiction. Seattle has a much healthier market both for devs and real estate and is only three hours away. Only difference in healthcare is while american top devs get it on top of their compensation packages, canadians pay for it with their prohibitive income tax rates. It's pretty easy to take the total spending percentage that goes to healthcare and multiply it by your income tax to see how much a dev is really paying for healthcare.
- usr1106 6y ago> Man, I wish I too shared such optimism in EU or other massive multi government bureaucratic organizations to pull something off like this in 2021. But sorry I simply don’t. I can understand your concerns. But if you look at mid-range airplanes the European monster Airbus did respectably catch up with incumbent Boeing. In with the Jumbos they clearly surpassed it engineering-wise but failed in their commercial predictions that such planes would be needed. So EU is neither a guarantee for failure nor for success.
- neltnerb 6y agoPlus side, the amount of money seems realistic. TSMC's latest node at 5nm/3nm was like $16B and so 10x doesn't sound outrageous if playing catchup. I do wonder how many people will be involved; it could be like ITER where everyone builds one fab that's truly the best in the world. Or it could be that they want to spread it out amongst the existing incumbents they can find who I am sure will do nothing with it. Concentrate the funding and I imagine it could do pretty well. Distribute to (more than one) commercial entity (preferably zero to retain public access) and I doubt it will go anywhere.
- wott 6y agoAirbus was not related to EU (or EEC). It was basically a France-Germany bipartite cooperation, with tiny bits of Spain and Netherlands, and UK being in/out. UK didn't belong to EEC when they were supposed to take part in the beginning of the project. Spain didn't join EEC until 20 years later. Netherlands were in by accident, in a way.
- usr1106 6y agoWhile Airbus is older than the current EU it was and is a multi-national cooperation in the same spirit, where political and national interests need to be taken into account, even if it reduces economic efficiency. They even needed the Beluga https://en.m.wikipedia.org/wiki/Airbus_Beluga https://en.m.wikipedia.org/wiki/Airbus_Beluga because they could not concentrate manufacturing of big parts in a single country. Boeing should have a clear advantage because they have no such political interference. But it doesn't seem to have prevented Airbus from being a serious competitor. Whether the political interference is older than the EU or from the EU of today should not make a significant difference here.
- jxramos 6y ago> Not the brilliant engineers you imagine are going to run it. I'm still so fascinated by our minds' ability to interpolate and extrapolate to the platonic ideal of any ungrounded and unconstrained idea so readily. Yes big bureaucratic mass of brain trust experts going at it nonstop, how can you not yield great returns? I'm starting to awaken more and more to how messy and data dependent this world is. Ideals and declaration and promises just don't go far enough. But it's all so complicated and situational to figure out in advance why they don't go as far as we'd expect.
- throwaway2048 6y ago>From my experience with gov trying to do tech investment it’s going to go into the pockets of late career overpaid ‘executives’ with ‘industry experience’ and tons of (also overpaid and underqualified) contractors doing the same thing. Not the brilliant engineers you imagine are going to run it. Are we talking about VC and large corporate investment or government investment here.