I agree with your broader point, but I think you are parsing your quoted excerpts from California and FL legal statutes incorrectly.

The CA statute does not state that ambulances can exceed the speed *limit* by 15 MPH when ignoring a stop sign or stop signal; it states very plainly that the ambulance cannot exceed the speed of 15 MPH when doing so. In other words, the ambulance has to slow down (typically) to 15 MPH when passing through such intersections.

Similarly, the FL statute states that the absolute speed limit for ambulances is 40 MPH, except in cases of "extreme emergency". It does not state that ambulances can exceed the normal speed *limit* by 40 MPH.

The e-mail thread in question is about an LLM-based code reviewer (effectively a fancy static analyzer), and not LLM-generated code. I think the copyright concerns are significantly different between those two use cases. I tend to agree that new code contributed by an LLM is likely entangled with copyright concerns depending on the origins of the LLM's training data sets; plus the hallucination and error risks are far less impactful if confined to code reviews and bugfinding. Also, at this point, I imagine an LLM trained exclusively on open source data sets could be quite capable.

Ted Tso, a senior linux contributor and maintainer, appears to agree; he noted on the thread that there are multiple use cases for LLMs and conflating all of them is not helpful.

You're a very poor liar, every bit if your nonsense is easily falsifiable.

Even back in 1990, when the models and computations were significantly less mature, the IPCC estimated ~0.3C of warming per decade in the 21st century.

Are you capable of the mental arithmetic to project to 2050 from 1990?

Measured temperatures have demonstrated acceleration beyond those initial projections (I'm surprised the current administration/ignorance cult hasn't deleted the NOAA page).

You're a toddler with oppositional defiance disorder, in a world governed by the hard rules of physics and chemistry. You and your doomsday ignorance cults are irredeemable, but those of us who are real engineers and scientists will try to pick up the pieces.

This is entirely incorrect. Infants can acquire Hep B from mothers, and the virus can survive outside the body for a week.

The first study above (at least) doesn't seem to support the claim. To quote:

Overall, the study finds that consistent conservatives: Are tightly clustered around a single news source, far more than any other group in the survey, with 47% citing Fox News as their main source for news about government and politics.

...

By contrast, those with consistently liberal views: Are less unified in their media loyalty; they rely on a greater range of news outlets, including some – like NPR and the New York Times– that others use far less.

And this study dates back to 2014. It seems logical that the advent of Trump, OANN and such have only worsened the phenomenon.

I consider myself a fiscal conservative and social liberal.

From that perspective, I recognize that Fox News, Breitbart and friends are generally more divorced from reality than the NYT or WaPo.

The Republicans, with a few exceptions like Massie, only pay lip service to the deficit when convenient. Objectively the economy has generally performed better under Democratic administrations following fiscal crises induced by irresponsible levels of GOP deregulation. The GOP also has a tendency to actively sabotage the government with poison pilled legislation, or the utterly irresponsible ongoing slash and burn of the federal government. Optimizing the government calls for a scalpel, not whatever idiocy Musk and co. are engaging in.

On the other hand, some Democrats believe in fantasies such as Modern Monetary theory and student loan forgiveness without tuition reform; that said, the student loan forgiveness attempt was a better use of funds than (largely fraudulent) PPP loans or a significant fraction of defense spending, carried interest, agricultural subsidies and other loopholes.

Don't quite agree. Quoting

The hackers broke into this first set of victims by brute-forcing accounts with passwords that were known to be associated with the targeted customers, a technique known as credential stuffing.

Did 23andme have no safeguards against brute-force attacks at scale? Or identifying logins from multiple accounts across a small set of IPs etc.? Or identifying logins from a location entirely different from the customary geolocation of the user, prompting an e-mail verification?

Gmail etc. all implement similar safeguards.

Yes, password hygiene is important, but blaming an 80 year old grandma for reusing a password is ridiculous, when a giant corporation could easily add safeguards.

Not to worry, they'll claim that the polio vaccine given to their great-grandparents manifests epigenetically unto the Nth generation.

(The latest conspiracy theory appears to be that polio elimination was due to DDT elimination, and not due to the vaccine...alas for humanity).

The twit oligarch has apparently also blocked Threads on the in-vehicle browser on Teslas, citing "security reasons".

He's almost as unconvincing a liar as Putin, though in the abstract neither Threads nor Twitter are sites for the privacy conscious.

They're effectively a (very opaque) credit bureau, but they seem to habitually violate the FCRA, and make it ridiculously difficult to get a copy of your report, opt-out, or correct misleading information.

Setting aside this issue, they've faced multiple class action action lawsuits regarding incorrect data supplied to lenders.

++, though he may have unskilled grifter tendencies--he previously flogged a "sword fighting game" kickstarter which failed.

In life, as with his (otherwise engrossing) fiction, Stephenson could use an editor.

Will try to respond in more detail later, but at least the second link purports to list age-adjusted mortality rates, to quote:

The new study, conducted by researchers in Texas, Missouri, Massachusetts and Pakistan, covers the years 2001 through 2019 and examines age-adjusted mortality rates—the number of deaths per 100,000 people each year—from the top 10 leading causes of death, as recorded in 2019. These include heart disease, cancer, lung disease, unintentional injuries and suicide. The researchers then analyzed county-level results in each of the five presidential elections that took place during their study period, identifying counties as Republican or Democratic for the subsequent four years. They found the gap in mortality rates between Republican and Democratic counties increased for nine out of 10 causes of death. (The gap for cerebrovascular disease, which includes stroke and aneurysms, remained but narrowed.) Political environment, the authors suggest in the paper, is a “core determinant of health.”

I agree there's probably some correlation to wealth (both in reducing susceptibility in the first place due to diet/lifestyle opportunities as well as better healthcare once sick), and there are probably many other sensitivity factors.

However, w.r.t. COVID, I notice you don't mention vaccination rates, which are probably the biggest influence--the vaccines are not sterilizing, but they definitely reduce death rates. Anti-vax correlates both with extreme left and right political views, but I think given the rhetoric from the likes of Carlson and various GOP politicians, I believe it is currently a much higher correlation with Trump-republicanism. Here's a study on that, to quote:

A U.S. News analysis shows that as of Jan. 26, counties where former President Donald Trump received the most votes in 2020 had a full vaccination rate of 52%, while counties that went to President Joe Biden had a rate of 66% – a difference of 14 percentage points.

I agree that state-level analyses are too coarse-grained.

Here's data showing that Trump/republican-voting counties had higher rates of hospitalization and death relative to blue counties, and it's not just COVID-correlated.

To quote:

Those living in counties that voted 60% or higher for Trump in November 2020 had 2.26 times the death rate of those that went by the same margin for Biden. Counties with a higher share of Trump votes had even higher mortality rates.

Another report indicates the red/blue county death rate delta predates COVID, although the magnitude increased during COVID.

To quote:

In a study published in June in The BMJ, Warraich and his colleagues showed that over the two decades prior to the pandemic, there was a growing gap in mortality rates for residents of Republican and Democratic counties across the U.S. In 2001, the study’s starting point, the risk of death among red and blue counties (as defined by the results of presidential elections) was similar. Overall, the U.S. mortality rate has decreased in the nearly two decades since then (albeit not as much as in most other high-income countries). But the improvement for those living in Republican counties by 2019 was half that of those in Democratic counties—11 percent lower versus 22 percent lower.