That depends on "repeated" and "nuisance". My experience in a Tesla is that the attention monitoring never has false positive alarms at all. And it 100% has caught situations where I'm head down in the cockpit on autopilot in ways that, honestly, I probably shouldn't have been. So there's a single data point saying you're wrong.
This kind of counter-intuitive/smarter-than-the-obvious pontification is a HN smell to me. Most of the time the obvious solution is correct. Driver attention management is one of the biggest (maybe the biggest) contributors to accident rate, and IMHO probably deserves some assistive technology attention. If the reverse is true, it needs actual evidence and not an attempt to reason from aviation evidence (a regime where inattention is NOT a major problem!).
Very different threat model though. Commercial aircraft aren't sensitive to keep-your-eyes-on-the-road failures with seconds-scale latencies, airlines require autopilot use, there is a copilot present at all times, the FAA very strictly regulates work hours and substance use, etc...
Sure, don't nag a pilot who is already very well backstopped by the existing solutions. Your uncle coming back from the bar at 2am doesn't have any of that.
Same protocol for a burning pile of tires at your favorite bumrocks exurban service station, amusingly. Yet no one loses their shit on HN about that horrifyingly dangerous technology.
I give up. This is a proxy war for something else. No one seriously is scared of batteries in the way people here keep trying to justify.
How much more likely are you to be injured on an eBike vs. getting run over on a Schwinn, though? Have you ever been afraid of the bike?
The presence of a different failure mode isn't the question at issue. Yes, they're new tech and require new techniques. Duh, as it were.
It's your abject and frankly irrational paranoia that I'm calling out. Chill the fuck out, as it were. Moving things have always been dangerous, and if you believe this represents a change in aggregate risk you are simply wrong.
And gasoline fires and high speed collisions aren't? This is the bit of paranoia around EV deployment that I genuinely fail to understand. Personal vehicles are by far the most dangerous apparatus with which we interact, already. And we're fine with it! Have you ever, even once, called a Civic or Vespa "scary"?
But put a battery in it suddenly they need new tech before you'll be comfortable?
You can, but people don't as a general rule. It's a traditional food prepared traditionally. And importantly, another thing early Alzheimers cases don't do is learn new cooking hacks.
> The most reliable source about a topic nowadays might not use their real name when writing. They might not have a journalism degree, or work for a mainstream media outlet. They might not have an academic background.
I really don't think that's true, though. Or it is in the specious sense by leaning heavily on "might". I mean, sure, in a handful of areas it is.
But in the real world, on the topics that make up the overwhelming majority of our discourse, the sources we trust most are human beings with names we know, and organizations run by human beings with names we know. Period.
And that remains true even in the spots where we admit to the presence of some anonymity. BTC was famously defined and documented by an anonymous developer. But even so, where do you go for your Crypto and BTC news? If, tomorrow, Satoshi were to return from the grave and start posting again, would you trust it? Probably not, right? At this point it would be more likely that someone found a crack on an old laptop and got access to his keys.
That's very much missing the point. It's absolutely true that the "approach" was similar (the bit encoding was software-generated in a timing loop and fed to non-ASIC hardware to send to the drive).
But the Disk II card was 8 chips you could get from Radio Shack, where the 8050 was a monster with a whole CPU/memory/bus (you can see the board appear in this video at 1:15 https://www.youtube.com/watch?v=S3d2cNSAB9A&t=69s).
And Woz's was faster. And hit market almost two years earlier. There's an aesthetic judgement to be made here too, and... it's not remotely close.
Again, we're not having an argument about health care, and I'm not trying to dismiss the book you cited that pertains to the argument we're not having. I'm saying that book is clearly partisan, as is your argument.
And I'm using that observation (which you keep reinforcing!) to justify my suspicion around the economic point you tried to make upthread (that, I guess, "hidden" health care "income" would increase labor's share if measured).
It's 100% clear to me that the labor/income point is spin, and to be blunt I don't believe it for an instant.
FWIW: Defense in depth is a security technique, and abuse detection isn't part of that domain. Security starts from the premise that the system is supposed to be undefeatable but might have holes, and then asking where the holes might lie to decide where to put backstops.
Here the system is "insecure" by design (literally they're trying to get the whole world to sign up for Claude Code for $200/month!) and they're trying to plug the hole that results from a "Except for Chinese Scrapers!" add-on requirement. That might be possible as an arms race kind of thing. But it's very unlikely to work by (as in the linked article) doing stuff like checking the system time zone.
> [my comment] was not partisan, and was based on reality and cold economics rather than ideology
That's what all partisans think. Nonetheless you put needless scare quotes around Medicare for All, dismissed it as a "total fantasy", and, I guess, helpfully suggested I read a book by (literally!) the head of the Trump administration's FDA as a reference for "What Broke American Health Care".
If that was all done in good faith, then you're in an echo chamber and need to escape.
Headline is, frankly, awful. This isn't the AI secretly doing stuff and hiding it. This is the very human Anthropic engineers trying to detect Chinese scraping via some frankly hamfisted and unimaginative URL trickery.
Is there a reference you can cite with corrected numbers? Honestly this sounds like excuse-making, especially when used as a jumping point into a decidedly partisan take (complete with scare quotes!) on the essentially unrelated subject of public health care financing.
The idea seems to have merit, but it's unconvincing to people outside your bubble and I'm dubious.
So... your solution is that Google implements authentication for DMCA takedowns using an authentication system that isn't Google's? Whose, then? Same headline, different villain. Probably still Google's fault somehow.
No, this is dumb and won't happen, for all the reasons that we yell about in other threads.
> A good start would be to require that claimants must verify their real identity.
I GUARANTEE that if this were tried, "YOU CAN'T FILE A DMCA COMPLAINT WITHOUT A GOOGLE ACCOUNT!" would rocket to the front page here and cause a(nother) general freakout about privacy concerns.
Depressing to see how much discussion on HN (!!) has resulted from a objectively terrible graph reading. I mean... in the process of our infiltration by finance bros and VC money, have we genuinely forgotten how exponentials work?
Well, there's an arguable point of fiscal policy there, and a conspiracy rathole that no one wants to excavate.
But none of that matters here. As grandparent comment indicates, you're making a pretty fundamental math error. This is a log chart. Government may have devalued currency. It did not do so exponentially.
> What’s more sustainable, changing your lifestyle to maintain the weight you lost, or being beholden to taking a drug to maintain that same lifestyle change for a hope at maintenance?
That's some pretty... charged language. But even so: the drug, clearly. People take drugs reliably as a matter of empirical fact. People likewise emphatically don't "change their lifestyles" as a general rule. If they did we wouldn't be talking about this new drug, would we?
> No, the symptom is being treated, not the underlying issue.
That's a semantic argument. The "issue", medically, to most people viewing this as a health problem, is excess body fat and not eating behavior.
I mean, you're not wrong, but this seems silly. YES, it would be better to have developed a cure for disordered and unrestrained eating. We didn't. And we don't really even know how. Oops, as it were. But we do have a treatment that avoids the most significant impacts of those problems.
Medicine is harder than software engineering. Not all bugs are shallow even with all the eyes in the world.
> If you can't run linux you can always run netbsd. or any *bsd.
That was NetBSD's marketing way back in the days of "I have a Sparcstation 2 in my closet, can I do anything with it?". It really doesn't apply to the systems in the linked article, all of which ran Linux very well at the time of manufacture and for which support has been really quite well maintained.
I mean, it's no surprise really, but objectively the best system in terms of coverage for ancient junk is Linux these days. Yeah, NetBSD runs on a VAX, but does it run on a 2008 Wifi router? OpenWrt probably does!
This kind of counter-intuitive/smarter-than-the-obvious pontification is a HN smell to me. Most of the time the obvious solution is correct. Driver attention management is one of the biggest (maybe the biggest) contributors to accident rate, and IMHO probably deserves some assistive technology attention. If the reverse is true, it needs actual evidence and not an attempt to reason from aviation evidence (a regime where inattention is NOT a major problem!).