A disclaimer to a lot of PNNL's marine research should be "Let's completely forget practical considerations, such as reliability, safety, ..."

No, it's shoddy research.

What I found is that folks who keep pushing "everyone must return to the office" are usually horrible communicators who couldn't write a proper email if their life depended on it. They tend to organize meetings for everything, and then don't get shit done, and just want to look busy. They usually also come up with BS excuses to avoid using instant messaging and just expect people to show up to the office instead.

More BS by folks who don't know a thing about the electronics industry. The entire push to get Intel to spin off its fab business is almost entirely motivated by the quick buck it'd lead to for particular groups of investors, but it'd be horrible for the company and the overall industry.

Yeah, that ain't happening, you'd lose guaranteed compatibility with all your legacy code for no gain at all. And even from a hardware point of view there's no advantage to switching instruction set at this point, building a RISC CPU with the same overall capabilities and performance as an x86 CPU leads to roughly the same power usage and design issues (for reasons I can't be arsed to get into in a comments section that's guaranteed to turn into a fapfest for ARM, AMD, and TSMC by folks who know jackshit about IC design and manufacturing).

You have multiple update channels for Windows, you can defer the feature updates for half a year (or more) and still get all the security updates.

And yet here you are, quoting theoretical numbers. Funny how that works, isn't it? Those densities can only be achieved on highly repetitive structures (aka memory). In reality you cannot achieve that density on actual designs due to yield, thermal and timing concerns. They have little to no meaning for actual general purpose logic, so unless if you're planning to make memory chips you're kind of out of luck.

And again, you are blindly staring at process nodes and PR numbers without understanding how to actually compare them.

Let's be quite clear about this, the name and number is meaningless and TSMC could have just as well slapped yoctometer or lightyear behind it. You cannot directly compare transistor density between technologies from different fabs because it ain't a good metric for the practical logic complexity achievable, nor does it indicate actual device performance. The size you're actually interested in is how much surface area you need to implement common digital circuit elements (e.g. common gates, SRAM cells, etc.), and the number of transistors required to implement such an element can vary wildly between technologies, please check the PDKs if you want to get a better idea of these numbers. And that doesn't even consider electric and thermal performance, you can cram all the transistors you want into a certain space, if the leakage current or switching losses are too high, you're going to have a bad time. This is why plenty of devices are manufactured in older technologies as well, the actual performance gain achieved by going to a newer process node can be surprisingly small versus the additional costs made.

And in terms of the CEO acknowledging they're behind, there are two major factors at play there. TSMC has been playing a dirty PR game for the last decade now with process naming and publishing complete bullshit figures, and half of what Intel's been stating publicly is to reduce investor pressure to spin off their fab activities. As to the reason for said pressure, for certain groups of investors it'd be a windfall if Intel would spin off their fab activities. So please learn to take public statements about technology readiness from both TSMC and Intel with a serious grain of salt, there's a lot more to it than most folks seem to realize. And the real gap is more in manufacturing capacity than actual technological performance, but that's not really surprising given the amount of support TSMC is receiving from the Taiwanese government. For added fun, realise that Samsung is breathing down the neck of both, and Samsung is in a class of its own when it comes to throwing money at things to make problems disappear. And that doesn't even consider the fact that UMC, ST and GF aren't quite as far out of the game as you might believe, they are surprisingly competitive in terms of electrical performance for smaller chips and simpler devices. Basically, considering the statements without the context is quite careless.

However, I am also fairly certain that all of this was semi useless and that you're going to quote more garbage from the TSMC fan club/wikichip.

Not necessarily, there've been several instances where the older Intel node was denser than the newer TSMC node, but it wasn't practically usable for things like general purpose CPUs due to thermal concerns, fine for certain types of memory though. Intel ain't really behind TSMC like a lot of the online and investor crowd made them out to be. Technology node comparison is a lot trickier than most folks make it out to be.

Another person who doesn't know how this entire racket works. You write your parts in Word/TeX/whatever... and you submit it to the editor. The editor has a full copy of the book, but unless if they send a finalized pdf to you, you have to get it through the publisher. Because have you ever tried to cite something for a grant application without having the page numbers?

This part is definitely not true: "process victimized authors, publishers, and booksellers,"

I'm an author, I had to use this website to access the book I wrote several chapters of because the publisher decided to be greedy and charge the authors an exuberant amount for a copy. Fuck Elsevier.

No, there's nothing wrong with those universities. The system is completely different and works as intended. You're training folks who have a final say when human lives might be at stake. These folks go out there and design public infrastructure, medical implants, aeroplanes, etc. So no, it ain't because you signed up that you're fit for such a job. So unless if you introduce a system like in the US where you need to follow training afterwards, that's how it's going to work.

Also, in a lot of European countries anyone who completed high school can start at university and go for any degree they wish (except for some arts programs and things like medicine). There's usually no binding entrance exam, nor is there any mandatory guidance. You select the university, you select the degree, tuition is often quite low too (e.g. about 750 EUR in Belgium). So it's not exactly a big disaster if you find out you're not cut out for it after a year, you can just switch to a different program.

They do exist, they're simply extremely rare. You'll have to go looking for them in the medical and aerospace industry, and they usually have an electronics engineering background and crashed over to software at some point during their career. I've never actually seen a software developer/engineer write critical software, it were always electronics or systems engineers, occasionally the rare physicist who went into software development.

What you forget to include in that statement is that you're by no means guaranteed to graduate at European universities. For example, in electrical engineering we started with about 35 people, we graduated with 6. Try running attrition rates like that at US colleges, highest number I've heard for a US college was slightly short of 50%.

An aspect many folks forget is that part of being a PE is telling your manager to go and shove his opinion on how long it should take up his rear exhaust port so you can do your job properly. Another aspect is that you simplify and add redundancy whenever necessary, this stands in stark contrast with the techniques many software developers currently use.

No worries, the same budget cuts are happening in European countries. They've consistently been cutting funding for everything while the economy is growing and taxation is going down. And trickle down economics don't work, all it does is increase poverty and reduce funding for essential services.