The point is to prevent "tearing". A lot of effort has been put into solving that problem.

If the pixels were updated in some random or semi-random pattern on the screen it would probably be unnoticeable, but I suspect that either a lot of architectural changes would be needed both in software and hardware, or you would effectively have to achieve a 480Hz full-screen refresh rate to achieve it without doing things like attaching an address to each pixel output so the rendering device didn't need to assume sequential pixels should be drawn sequentially.

Wait, are you GPT-3? Maybe 2?

What do you mean by "compiles cleanly to predictable ASM"? Do you mean individual lines always output the same assembly in any given context? That hasn't been true for a long time, so you can't mean that. Do you mean that if you follow the steps of the C compiler in a given OS state that you will get roughly the same (the same for all intents and purposes, but things like compile-time constants like, well, COMPILE_TIME might differ) output? Well, yeah, that's called running a program, you're just doing it in your brain.

Are you saying that every C compiler is deterministic in a (consequential) way that Rust compilers are not?

It's got to be something I haven't thought of. What am I missing?

I don't know Rust, but why would those structures be particularly challenging for bounds checking?

You carefully read it twice, and still latched on to a point that you made up.

Congratulations. You have objective proof that you're both stupid and so gullible that you can be fooled by someone as dumb as you. This will be highly valuable information in your life from now on, should you choose to accept it.

IIRC, Apple uses it to tag your photos.

We’ve spent the last five decades catching up to the safety and efficiency that the initial promises were based on. We sold those initial promises as achievable in 5-10 years and retrofittable onto existing plants, which everyone building them knew was bullshit, but they wouldn’t have been buiilt without the lies.

These days, we know how to make truly safe and efficient reactors, but a half century of lies means only experts or idiots will support them, and even then they’ve fucked up so much, legislation is utterly intertwined, which means politicians need to be convinced, and they’re rarely either experts or idiots—so basically, short of bribes, nuclear power is fucked until we get truly desperate, even though it can now be safe, profitable, and clean.

I dunno. It was probably a necessary step. I can’t remember where I saw the concept, or a googlable name for it, but the gist is, we often have to do something awful in the short run to get something good in the long. Provided the short-term necessary step dosen’t kill us, it was worth it.

Heartbeat apps that require nothing more than looking at the camera have existed for years. I’d be mildly surprised if FaceID didn’t require a heartbeat.

I haven’t read the second link, so I’d be happy to be proven wrong, but I seriously doubt the investigation found that the paper instead proves the null hypothesis. It’s extremely annoying that even purported “science” writers can’t be precise with language.

Prediction: This article will be quoted, without a direct link to the original paper or investogation, asserting that “The Mediterranean diet [b]does not[/b] cause better health outcomes,” instead of “does not necessarily” or “has not been proven to”. And then the imprecise language will be further extended to call it harmful, based only on the game of telephone going on here. Feel free to link this post and call me wrong on or after 2021-06-15, I’ll find the articles.

My last official predictiction was to say that even though the Facebook IPO was widely acknowleged to be a ripoff, you would have more money five years after than you started with. I actually think it was much stronger than that, but regardless, I was right. Feel free to find my post from, I think, the day of the Facebook IPO. If my claim wasn’t at least as strong as I’ve claimed, post-inflation (I’m pretty sure I was smart enough back then to take both inflation and opportunity cost into account and think I said you’d at least double your money, but again, please prove me wrong, I’m too lazy to look the post up but refuse to refuse to accept reality), I’ll do something extremely personally embarassing with my real face on Youtube, do my best to make it funny to whoever posts a link to my incorrect post, use my real face, and link it as a reply.

I’m pretty fucking good though. Unverifiable claims:

Two weeks prior to the Google acquisiton of the (UHF?) wireless spectrum, I called that they would win, with their bid to three digits of accuracy (11.x million $ IIRC). Reasoning behind it? Slashdot, actually. Back then, there were a lot of very capable and informed people here, and having just started to become truly competent and trying to figure out how to communicate that to my peers and managers, I was really tuned in to identifying actual experts from their posts, and just let my brain pick a random number. I can’t say the reasoning was solid or anything but luck... but I think it was and was.

On seeing OS X 10.0, I called that Apple stock would skyrocket within 3 years. Reasoning? A company dropping its ego, and more than that, getting their senior devs to make an honest attempt at standing on the shoulders of giants, up-front and not post-hoc? It’s so obviously a path to success, and yet so rare, it’s practically guaranteed to succeed.

Called iPod success. Laughed at “Less Space than a Nomad”.

Called iPhone release when everyone said they’d never do it.

Successes posted as a reminder of confimation bias in case I’m wrong—I’ve only called one thing verifiably correct on Slashdot. Even if I get this one right, 2/2 in 12-ish years is not statistically-significant. Just making sure I can’t rewrite my unverifiable correct claims in case they end up wrong in hindsight.

CDATA is trivially broken out of.

Sure, but GPS wouldn't work (or rarher, would be simpler) in that case. The weirdest things in the universe are proven by your phone existing.

In which case, your incredibly accurate studio monitor speakers that you hooked up to a consumer-level receiver with the HPF/LPF turned off (filters still aren't common on woofers and tweeters in consumer receivers) will play weird resonances from the tweeters and woofers as they recieve signals they either are too heavy to vibrate at or too light to deal with the amount of power it takes to move the heavier speakers they were intended for, and if you don't pair them right (overlapping frequency responses, set the H/LPF filters there) you'll get certain freqencies standing way out (and usually distorting) or you'll have your subwoofer playing 150Hz, which if it's bigger than 6" you'll then turn way up to even out the volume with the mids, but never manage to get "Nine Inch Nails - Closer" to properly be "Boom - (inaudible because of your brain Boom) - Tiss".

Seriously, if you're buying a stereo, play Closer. Make sure your sub is at least 10 inches, and has at least 50% db at 20Hz (closer to 100% the better). Set the mids to mid, treble to mid, base keep turning up to where it goes "Boom Boom Boom Boom Boom Boom". Slowly decrease the bass, at some point that weird thing your brain does that MP3 exploits will take over, and it'll become "Boom Tiss Boom Tiss Boom Tiss", but the Tiss will just "feel" impactful. If it's a single unit, speakers and all, that should be very close to halfway. Next, increase treble until the Tiss and other highs make it sound cheap, then back it off by a half-number (on a 1-10) until it sounds crisp. This will be +/- 2 numbers on a good pairing of speakers, and +/-1 on integrated tweets and woofs. Finally, bring the mids to where Reznor's voice is easily understandable without trying (this is a really, really well-mastered album) but the music is still in the foreground enough that you lose words by being distracted by new layers or progressions in the instruments.

Been my main stereo-benchmark song for ~8 years, and it's both saved me from some clunkers and got me incredible deals. Just make sure you have a lossless copy, and don't play it over bluetooth unless that's specificially what you want to test (although you'll do better by checking the A2DP protocols your devices support for that--bluetooth play is only worth it as a final smoke test).

Also, if anyone has other songs that really exercise a system's edge cases, I'd love to add it to my rotation.

Portishead - Roads, aside from making any white noise in the area sound really strange, should sound amazing with the Closer tuning--the bass that makes the wah-wah should almost be inaudible unless you're in a white-noise-free room. If it doesn't, the high/low filters between mids and sub is fucked, or even worse you've got a gap (4" "woofers"), and you'll never find a non-annoying song without a fill-in speaker + EQ that will be a huge investment of time to get a smooth response from.

Lossless purchases are extremely easy to find, and in the last 3 years have been mostly equal or superior to CDs.

Not that I'm aware of a single A/B/X test where people were able to tell the difference, but hey, it's as good, unless you have a janky playback system without correct high/low-pass filters.

Wait. You think your CDs, that even if you don't play them on an internet-connected device, can reproduce frequencies the human ear can't hear right next to your phone that can pick up those same frequencies, that you purchased online, or at the store with a credit card, or customer discount card, or that you googled first, you think you're immune to being tracked with them?

As for DRM, that's been a non-issue for digital music for at least 5 years, more like 8-10.

Yeah, no. Any audio content can be distributed on CD that (at least 3 9s) of human ears can hear. I have yet to see a band, from Radiohead, to Phish, to bands that only play local ~100-person venues, that doesn't have a CD of everything they've ever made.

I love live music, but unless you have a massive and/or high-quality PA system, expert mixers who've tuned to the venue, bartender, and at least 20 friends to enjoy it with you, CD is just plain better, usually even when it's a live show. And it's always available. Maybe there's a niche where that's not common, but that niche isn't going to kill the CD.

50's and 60's mastering was objectively terrible. I don't think there's a single frequency or intensity that record can reproduce that a CD can't. I might be wrong about that, but if I am it's a frequency you can't directly hear. If you think CDs suck, you're a victim of the Loudness Wars. Listen to CDs mastered in ~'92 to ~'96 ('93-'94 was the peak)--some of the best, clearest music you will ever hear, and I challenge you to give a single example of an alternate format that captures the live sound better.

CDs are dying because you have to go to the store or get them delivered. That's the entirety of it, and there's no reason for concern--compressed music will eventually get a no-comprimises engineer combined with a band intricate and talented enough to make it worth it, and incredible speakers are so cheap these days somebody just needs to make one that's Bose-compatible, Base-boosted-monitor-quality, and Apple-easy, and HiFi will be the new rage.

Your Youtube anectote makes me worry I might be wrong, that's straight fucked that anyone could listen to Youtube music and not be annoyed, but I've been right since cassettes, so I'm going to bet on this one as well.

You're thinking way too small.

If this is real, and stable (given that they didn't even bring up the question, and that the article is at the traditional Slashdot-level of understanding, I'm guessing it's not, because these articles are never anything excition), it'd absolutely transform our understanding of quantum mechanics. Things we think are impossible would become easy.

But it's not, and it's not, and it won't, and they aren't, at least not in any way that will ever show up on Slashdot.

Also... really? Electricity is already pretty easy to get like 90% efficient from hundreds of miles away. 2x electric would either be 95% or 180% efficient, depending on how credulous you are.