s/is/has/

I agree with all that.
Underpowered studies are a bane in medicine and nutrition research is ruined the reputation of science among lay audience.

Whatâ(TM)s interesting here is that as a professional musician, this guy is a public figure and the âoeactual maliceâ standard for defamation applies â" a standard that was designed when defamation could only be done by a human being.

This requires the defendant to make a defamatory statement either (1) knowing it is untrue or (2) with reckless disregard for the truth.

Neither condition applies to the LLM itself; it has no conception of truth, only linguistic probability. But the LLM isnâ(TM)t the defendant here. Itâ(TM)s the company offering it as a service. Here the company is not even aware of the defamatory statement being made. But it is fully aware of their modelâ(TM)s capacity to hallucinate defamatory âoefactsâ.

I think that because the tort is based in the common law concept of a duty of care, we may well see the company held liable in some way for this kind of thing. But itâ(TM)s new law; it could go the other way.

The laborious, linear interface is of course another limitation of all kinds of tapes -- digital or analog. But getting rid of this also changes human behavior. People don't listen as much to long form collections; they don't even necesssarily listen to entire songs.

A mix tape is essentially a long format program manually and personally curated for you by another human being, unmediated and indeed untracked by any third corporate party. Losing the mix tape was a real cultural loss. Sure they didn't sound great, but they didn't have to.

I suppose every technological advance is potentially double edged. When people get books and literacy, verbal storytelling declines. That doesn't make books bad. the technical limitations of verbal stories -- say limited repeatbility -- are real limitations, but that doesn't mean something wasn't lost.

I disagree. I see both sides. Medicine and the computational side.

Research in medicine is full of con-founders, inadequate samples, and imperfect experiments. The confidence you bring from clean and sophisticated methods and analyses misleads you in interpreting messy evidence (I understand that you say that education research is similar). I care as much or more about where the research was conducted than what they wrote in the paper because most research in medicine is plain wrong in its conclusions.

https://en.wikipedia.org/wiki/...

The statistical methods used medical studies are relatively much simpler than in say, engineering. That's not where gotchas are. So we need robust studies, a convergence of evidence, and meta-analyses from competent centers.

> And C++? I am not even sure you can port that to Rust without essentially writing a compiler that adds all the missing OO features. Rust has limited and very non-standard OO. Which makes sense given its aims, but not when you come from C++.

GenAI will generate the code.
Rust makes most sense to C++ devs since it is essentially encoding C++ best practices into a nicer, more modern language.
People who have migrated C++ to Rust have reported good experiences for years. Microsoft had plenty of experience in the area.
AI can sort out the necessary transformations in type structure.

Not really. Rust was always built to gradually replace C++.
They don't need to do a complete rewrite of anything at once.
It will coexist in each C++ project, slowly replacing things, few types and functions at a time.

It still would have been a mammoth task in the past. But GenAI automates much of this.
Rewrites aren't as big of a deal now, compared to 3 years ago.
This is what GenAI is supposed to do. Improve agility. Better conform to best practices etc.

On the up side, it means they won't layoff devs too much, since keeping up with changes is still lots of work.

Everyone can read papers. Not everyone is qualified to understand them in proper context.

Unless you have at least a masters in the related subject, ideally a PhD, your ability to vett the papers in that domain should be assumed to be rather limited, especially when you are arriving at conclusions that the experts aren't by consensus.

In biology and medicine, evidence works differently than in tech. Controlling for confounders is much harder.

PBS is primarily (85%) privately funded. It will continue to produce shows like Masterpiece, Nova, Frontline, and Sesame Street and people in places like Boston or Philadelphia will continue to benefit from them.

What public funding does is give viewers in poorer, more rural areas access to the same information that wealthy cities enjoy. It pays for access for people who don't have it.

By opting out, Arkansas public broadcasting saves 2.5 million dollars in dues, sure. But it loses access to about $300 million dollars in privately funded programming annually.

Seriously, the idea that we know all the practically important physics there is is the kind of thing only somebody who's never done science or engineering would believe.

Industrial R&D is important, but it is in a distrant third place with respect to importance to US scientific leadership after (1) Universities operating with federal grants and (2) Federal research institutions.

It's hard to convince politicians with a zero sum mentality that the kind of public research that benefits humanity also benefits US competitiveness. The mindset shows in launching a new citizenship program for anyone who pays a million bucks while at the same time discouraging foreign graduate students from attending universtiy in the US or even continuing their university careers here. On average each talented graduate student admitted to the US to attend and elite university does way more than someone who could just buy their way in.

Republicans equate being pro-market with being pro-big-business-agenda. The assumption is that anything that is good for big business is good for the market and therefore good for consumers.

So in the Republican framing, anti-trust, since is interferes with what big business wants to do, is *necessarily* anti-market and bad for consumers, which if you accept their axioms would have to be true, even though what big business wants to do is use its economic scale and political clout to consolidate, evade competition, and lock in consumers.

That isn't economics. It's religion. And when religious dogmas are challenge, you call the people challenging them the devil -- or in current political lingo, "terrorists". A "terrorist" in that sense doesn't have to commit any actual act of terrorism. He just has to be a heathen.

no problem.

I'm actually responding to the AC above you. He is arguing that the attack wouldn't make any sense for either country to make, based on *national* interest. I'm pointing out that's not the only framework in which *regimes* make decisions.

Just put it in context: Today Russia struck the Pechenihy Reservoir dam in Kharkiv.
Russia launched the war because they thought it would be a quick and easy win, a step towards reestablishing a Russian empire and sphere of influence, because Putin thinks in 19th century terms. Russia is continuing the war, not because it's good for Russia. I'd argue that winning and then having to rebuild and pacify Ukraine would be a catastrophe. Russia is continuing the war because *losing* the war would be catastrophic for the *regime*. It's not that they want to win a smoldering ruin, it's that winning a smoldering ruin is more favorable to them and losing an intact country.