The only people still using it are the ones that will complain about anything they can, and will never be satisfied cause they need this one special tweak that everyone else KNOWS is stupid, but fuck it, we put a VPN in there, why not support the dumbest shit for one dude on his high rocking horse. Who immediately complains about some other stupid idea. So will give him a prompt first use ...

99% of users want things to work without having to answer a bunch of stupid questions based on some developers shit head politics and need to impress his childish view points on users.

And this is why Foxbat will always be nothing better that the replacement for IE6, it will continue to be an example of why developers and their personal politics shouldn't be allowed in the git commit history.

Add another option is usually the result of not beimg able to figure out a better starting point. Its rarely solving problems, usually adding bugs, problems and complexity.

I need a browser that isnt an irrelevant PoS that has spent its entire life with "trying to be a browser" as the last thing on its list of development priorities.

XUL, email, rrs, web page editor, VPN, all sorts of political shit, literally so many options it cant help to be a slow bloated buggy pile if donkey dung.

You live in a fantasy if you think hipaa protects you from data sharing.

When you agreed to medical service and signed the hipaa notice, that you didnt read, it was you agreeing to allow them to share it for business purposes, you are not allowed to opt out and get service ... then it gets mixed in with other data and sold.

You have this false fantasy that anyone cares about hipaa beyond getting you to sign the contract.

The town can simply ban that sort of surveillance. Contract can go fuck itself if the law supersedes it, yes retroactively.

This retarded idea that a company can do whatever it wants because of a weasely contract is false.

The residents control the town, not some secretly foreign run mass surveillance company these idiots keep getting bribed by.

Thats cute. Now put it in front of the public without the person on the other end following your demo script...

All of the sudden its shit.

My company has done this as well, and it works great for the guy who made it, who talks to it with knowledge of its capabilities and nuances... but then as soon as anyone with a slightly different choice of words comes along ... its fucked.

LHS 1140b has 5x the mass, but only about 2.4x the surface gravity (since the surface is further away from the center of the planet's mass). Which is still enough to make life difficult for large animals, of course.

But even if the gravity was 5g, you could get a lot bigger than bacteria. Regular mice from Earth can survive in 7G environments for months at a time. Bacteria apparently can survive hundreds of thousands of Gs.

Though, to be fair - this really isn't an important announcement - its just an improvement developers will appreciate - and a few guys that truly understand hypervisors. The rest of us, at best will say ... 'cool ...'

While not entirely true in desktop land yet - increasingly, more and more things are containerized, then run inside virtualization ... this does matter to everyone, even if its not immediately apparent.

Fair point, I should have specified "the professors and the administrators" need to do their jobs, not just the professors.

I also get it that colleges "need customers", but... two of the universities mentioned in TFA are Brown and Princeton. Both Ivy League, both top-10 in the US, both with applicant acceptance rates of 5% or less. These universities are NOT going to run out of potential students if they start cracking down and enforcing standards. In fact, if they start cracking down and if they're very public about the crackdown, this will help their reputation immensely, and will make them more attractive to prospective students (at least the ones who don't plan to cheat).

Long before LLMs, students would find ways to cheat on assignments-- either by hiring someone to write their papers for them, or by simple plagiarism. The penalty, if you were caught, was that you flunked the course with no appeal. The penalty for repeat offenders was that you got expelled.

The only difference here is that these students aren't getting appropriately penalized. These professors need to grow a pair and do their jobs, instead of rambling on thoughtfully about "best practices".

It did become "dated" faster than any of us would have expected, reading it back in the 1980s. But "dated" doesn't mean that it's bad or that it has aged poorly. It's still one of the greatest opening lines of any novel. The reason you can tell that it's a great opening line is that you were able to recall it word for word. You probably didn't even need to look it up to remember the exact quote.

Also, $70K is the median *household* income. The median individual income in Omaha is $41.6K. As gtall said, factor in taxes, healthcare, rent etc... then factor an extra $500-$1000 to register and title your new car, another $1000 annually for vehicle maintenance (except it'll be more than that because for $20K you're buying a used car), and factor in whatever you're spending on gas...

> driving a car, industrial control, vibe coding are bottlenecked by hallucinations

No they're not, this isn't 2022 anymore. Let me reiterate, they're fairly trivial to spot and largely accounted for in serious autonomous systems. The issue is monkey paw property of the LLM - they're maliciously compliant by seeking shortcuts too greedily while skipping over subtler ontologies that aren't verbalized in corpus enough so as to differentiate into a concept ("carwash is typically somewhere you might want to drive to" wasn't solidified -> hilarious stupidity ensues). This sort of mistake isn't hallucination, but just plain stupidity of someone who's never quite understood what carwash is for, but only heard em mentioned few times in a book.

Yann LeCun does point at this, though I think he's too optimistic that DNNs could fill those gaps by having online learning experience in a closed loop. The way GPTs work is embarassing parallel SGD that you simply can't do when learning one speck of experience at a time. Would need to invent entire new optimization algorithm that scales just as SGD if not better, yet works sequentially.

I mean dumb humans are confidently wrong (hallucinating) all the time and especially humans in managerial positions he's positing this should replace.
Although LLMs hallucinations are driven by biases alien to humans.

This seems like a quick cash grab aimed at AI antis of Ed Zitron ilk, who famously insist that "AI is real only if it is never wrong and never hallucinates", but they never provide any insight why the AI needs to be always 100% correct.

Meanwhile, the real problem is that LLMs simply descend into madness and go off the rails it's not hallucination per se, but simply wandering off meaningful goals. Mere hallucinated errors are comparably simple problem and possible to correct for with checks and balances on authority, compared to more fundamental inability of LLM "intellect" not turning into a mad king (whether always correct or not, being mad is trouble).

And this isn't possible. The chances of this plan coming to pass are precisely zero. It's a fantasy. Whether it's actually a good idea or not-- that's completely irrelevant.

The real-world effect of this proposal is that it makes the Democrats look like a bunch of loopy socialists. Since we're heading into some *very* important and *very* tightly-contested elections, this is a tone-deaf and deeply irresponsible move on Bernie's part. (If it was anyone other than Bernie, I'd suspect that he was a deep-cover Republican operative trying to discredit the opposition).