You missed:
1) new flashy instructions that are implemented to win benchmarks, hyped to hell and with no regard to competent ISA design, so there needs to be a v2 extension in the next gen. And then a v3. And then a v4. (Which then transition from "Intel Mouthpiece says new instructions will cure cancer etc etc" to "We feel non-standard extensions like these are harmful to the x86 ecosystem as a whole" once #4 happens).
2) new flashy instructions that require the CPU to downclock.
3) new flashy instructions that not work on *some* CPUs - including *future* Intel ones.
4) new flashy instructions that are then removed from future chips because everyone else does them better than you.
5) new flashy instructions added just to chase a particular fad ... which stopped being cool 3 months ago.
6) TDP lies that are now out by a factor of *4* or more in some cases. (Up from 2x just a couple of years ago, which was already bad enough! And yes, I know what Intel's definition of TDP is this week. Doesn't in any way change my comment).

(I do, oddly, wish Intel luck after the board's decision to wreck the company; but that doesn't change the last decade-plus's worth of absolutely shit products, let alone your laundry list of older Intelisms).

(AKA "Lie Strategy #93" in the company Handbook)

> "If you can do 30% more code with the same number of people, are you going to get more code written or less?" he asked.

Less, in this case. Once the 30% claim (which itself is bullshit) has been around a few months, IBM will "rightsize" the oldest 40% of the devs. (40% because "the AI will be even better in the next release, so we might as well do it all at once. Also, they're old, so fuck 'em anyway").

The AI-slop code still won't be working by then; the now-senior staff will be spending all their time trying to repair it instead of finishing the feature set; and the juniors will be doing the same, just by by feeding ever-longer queries in with phrases like "and doesn't cause an RCE if it reads a crafted JPEG" being added each time.

Okay, the idea of IBM *caring* whether the code actually works or not is a bit of a fantasy, but I feel pretty confident about the first part of that prediction for some reason...

That depends on your position, not your perspective. If you're stanning Twitter-the-company, then sure: that's a coherent stance. Not merely "not hypocritical" for Musk etc, but practically required from him.

If it's the content that matters to you, I'm sure sufficient disho^W mental gymnastics could let you hold either position. Obviously, this is the largest cohort by a mile - as evidenced by Twitter still existing. Given all the success of politicians, lifestyle/shitcoin/etc and other scam artists on the platform, it's hard to be shocked by that. Of course a gullible audience with flexible morals can be made to see any viewpoint as valid. So yeah: intellectually dishonest, but not technically hypocritical even under the ever-fluctuating meaning of English. (Though given the rate and direction it's adding them, I wouldn't put money on that being case within a year).

When the IIC constantly rants about "Freeze Speech" ever time he's getting his ass kicked about something though; while also banning Twitter users for Free Saying something? That's practically the poster child for hypocrisy, and he's rightly getting blasted and mocked for it.

This, I have to disagree with. "Stupid folk gonna stupid" is about as not-worth-mentioning as you can get.

To be somewhat pedantic, you're just confirming that it wasn't YouTube (which you in fact haven't done, and I thought that would have been obvious to you) - you're not actually arguing the truthfulness of his wider point (which is good, or you'd have looked rather silly). The way you've phrased your reply though means the opening sentence conveys a nearly-opposite meaning to what you're actually saying.

There are quite a few ways these "New! !" scams go, but the two major ones are:
* The "XYZ History will only *display* 1mo of data instead of 6mo" PR spiels, explicitly intended to mislead, have unsurprisingly turned out to be "misinterpreted" by gen pop on numerous occasions (not helped by trade blog shills, to be fair).
* The "actual claim" lies of "We don't record XYZ Data unless you don't not opt-in", OTOH, have repeatedly been shown to be lies far FAR more often than not. (But it was an honest mistake every time; or cosmic rays; or a "rogue" programmer (man, there are so many of these now, which is really weird for someone who spend his whole career in development) etc, but i digress). Technically, your response is also in this group.

I'd love to know how you imagine "the next generation" *could* "fix this". Seriously.

I mean we've had over half the solution for 50+ years, and we haven't done anything with even that much as a free square.

But IBM does, and that's the whole point of this 15-year trainwreck.

> Wayland has been in development for how long now? 15+ years or something?

Longer - closer to 20 at this point.

> It better be the most awesome graphical layer and compositor ever.

It isn't. Hell, it isn't even as capable as X11, while also being no more performant and only fractionally less insecure. (In fact, it's no better than Mir was 10+ years ago).

> Which, if I recall correctly, was/is the entire point of the Wayland project.

It wasn't. The "point" of Wayland was exactly the same thing that happens every time someone tries to take over a piece of code that's out of their depth: "This sucks, so I want to rewrite it because *I* can do it *properly*".

Since then it's just become a way for IBM to favor GNOME by pushing an impossible maintenance burden onto every other DE, and forcing app developers to migrate to GTK4. It's wasted literally millions of FOSS hours at this point, all for literally no gain that wasn't much more achievable by just spending that time refactoring, fixing, and extending X11 - you know, the way that we got to X11 in the first place. But that takes *work*, and it doesn't help IBM screw over everyone else, so, yeah, can't say I'm surprised they didn't go with that. :/

The plan is "Have the waiter put ice in my gin and tonic" while the poor and the elderly die in droves.

it's $36M/yr, i.e. less than 0.1% of their revenue and probably less than 1% of their annual profit - IOW, not even worth the effort it would take to make bothering to comply with the law worth it. As usual.

Maybe worrying about bullshit grifters should come after making some progress on the corrupt government and ignorant populace?

Besides, by then this wave of overhyped garbage will be as dead as shitcoins, VR, and the previous dozen wastes of energy, so it won't even need a "culture shift" to deal with it...

A Pi.

Practically all the alternatives offer better HW, but bring with that a hellscape of unmaintained out-of-tree patches to actually get working X, LAN/WiFi, etc etc. Unless you're just collecting SBCs for fun, the Pi is still the only meaningful horse in this race, despite its significantly lower specs and its own multitude of software problems (mostly around acceleration for video encode and decode, and anything else to do with the godawful Broadcom VideoCore).

I love tinkering with these things, but until the SoC providers actually start putting a non-zero amount of effort into supporting their chips you need to have teenager levels of free time just to get them "sort of, mostly" up and running. On a Pi4 though, I can set up basically anything I want - whether that's a headless server running as NAS / router / DNS / etc, or a full desktop, or both - *running my distro of choice* in about ten minutes, and know that (nearly) everything will Just Work.

If spending the weekend compiling (cross- or otherwise) random tgz's from .cn domains just to get working Ethernet is still a luxury you can afford, and doing so still interests you, then yeah: pick pretty much any of the dozens of "better Pi's" out there, and you will indeed get something with 2-8x the CPU power; and/or a Mali GPU that, technically, on paper, is massively more capable and performant than POSVC6; and so on. The "problem" is that if you want an SBC for HW projects, the Pi's GPIO, software support, and massively-larger ecosystem make it easily the best option. If you want an SBC for SW projects, the support and ecosystem again make it the only reasonable choice.

If you want something "significantly more powerful" than the Pi4, you're into active cooling territory and would probably be better off with an N95 NUC or similar, or even an x86 ITX build, because the intersection of genuine needs for "more than 4x 2GHz of CPU" *and* GPIO is microscopic - which is why the dozens of companies making such devices are invariably years behind with their OS support. :(

Yes we do: it's called the McGill scale. VAS (which is what you're talking about) is a garbage system only really appropriate for simple injuries, not chronic pain.

> If we could see how much pain someone is really in with a test, it would

not make any difference at all. The frauds would still commit fraud, and the same corrupt doctors who falsified their "whiplash" cases would claim the pain is variable and they were just tested during a low point. The drug-seekers would likewise still lie about their pain, because they're addicts. The patients with actual chronic pain, which often genuinely *is* highly variable, would end up being randomly denied adequate medication because you can't take a measurement at a single point in time and expect that to accurately represent their condition over a span of multiple months. It's as accurate as declaring a patient dead because a millisecond reading of their heart rate came back as "none".

In short, you have a nice imaginary set of outcomes, but they're based on such a complete lack of understanding of any of the scenarios you're talking about that all those outcomes only exist in your head. The only thing this will be used for outside of the lab is denying treatment based on intentionally bad "science" to save insurance companies money.

Very much so.

You know what the funny part is? The whole "adding acetaminophen" tragedy came from *one* idiot, and is based on literally *nothing* except his own imagination. No testing, no studies run, nothing: just utter bullshit pulled out of thin air, and discredited over and over again for decades since. Norco/Vicodin still does it to this day though, and it's impossible to get hydrocodone in any form without at least 30x as much liver-destroying poison added to the pills as there is actual medicine.

Yep: it went from +500mg to +350mg. Hepatic failure is at ~4000mg/day, with 3000mg/day generally considered a nonlethal but harmful dosage.

Liver failure via acetaminophen has killed *far* more patients than hydrocodone itself has - all for no reason at all, as the analgesic is of no benefit whatsoever to patients whose pain is severe enough that they need opioids in the first place.

"It depends". If your doctor is competent, and you have a similar response to different opioid types, you can cycle between e.g. hydrocodone and oxycodone and give the receptors time to recover.
If your doctor is incompetent though, or only one of them works for you, the problem is much worse than "you need a bigger dosage" - you can burn the receptors out completely, such that one type no longer works at all - and the recovery time for that appears to be "effectively never".

"Simple" pain meds like opioids are actually dirt cheap in the USA. Some of the neuropathic ones are over $500/mo though, and not covered by insurance (because "insurance" in the USA gets to choose whether or not it pays for anything). So yeah, I know plenty of chronic pain patients who simply can't afford to fully medicate, and are forced to live in misery instead.

The bigger problem is as GP says though: because of the "Opioid Crisis" political theater bullshit, doctors are essentially barred from prescribing anything like the counts and dosages patients with severe chronic pain need; and in the rare event that you can get something close to adequate prescriptions the local pharmacy will fail to be able to fill those scripts at least 3 or 4 times a year.