California is in the middle of a huge fire insurance crisis. It started with the intentional housing supply restrictions that drove up property prices and rents. In suburban areas, rebuilding costs were mostly increased indirectly through higher wages (as tradespeople and laborers have to make rent.) This sent insurance rates through the roof and caused a wave of policy cancellations. Many insurance companies exited the market altogether [1].
Climate change is also to blame. The firestorms of 2017, 2018 and 2020 broke all records, and were insanely expensive to rebuild after. The typical trigger is a katabatic wind event [2] after a long dry spell. This massively reduces relative humidity (often to 5-10%,) making ignition much easier. Once a fire starts, the wind spreads it extremely quickly. Sustained wind speeds of 50-60mph are not uncommon near mountain peaks.
In 2017/2018/2020, the precipitating events were so intense that the initial responses focused exclusively on helping the residents out. By the time the actual firefighting began, the fires were already enormous.
It's surprising to me that we haven't seriously looked into large-scale sprinkler systems, such as this one deployed in Spain [3]. These could take a major bite out of the initial uncontrolled stage. They could either be deployed in the wild along naturally defensible lines, or at the perimeters of inhabited areas.
They're expensive upfront, but not as expensive as the alternative. They might also reduce the need for prescribed burns.
It's really an amazing study in political science. Progressives are supposed to be all about helping the poor, but some were sold a camouflaged dystopia.
To reach utopia - the fantasy goes - we must first travel through a dystopia in which the poor are enserfed by the housing barons, paying upwards of 70% of their income in rent. Unfortunately the dystopia never ends in practice.
Then there's the perplexing question of how real estate taxation got to this point. As the saying goes, a real estate family that pays any income tax at all needs to fire their tax advisor.
Indeed artificial housing supply restrictions are about 100000000% more important. Alas, a portion of the progressive left was co-opted by the housing barons many decades ago. I believe California voters would be more progressive otherwise.
Here in SF, a good example is the leading progressive mayoral candidate, Peskin. He's basically a housing subversive. He'll pay lip service to it, then sabotage YIMBY efforts. Earlier this year he sponsored an ordinance blocking higher density in parts of his district. The current mayor vetoed it [1] but he got the board to override the veto [2].
The barons originally sold suburban supply restrictions as anti-sprawl measures, co-opting the greener factions of the left. Then they sold density restrictions as anti-traffic measures.
No suburbs + no density = no new housing.
Luckily this unholy coalition has started to crumble a few years ago. A large majority of Democrats is now in favor of more housing.
True. The counterargument is that if you care about quality, you're already better off with Continuity Camera, so maybe you already view the built-in as just a backup.
A MagSafe charging spot on the back of the monitor would be a nice touch. Charge your phone there and its camera is clipped in place by default.
You can't pick up a rock in Italy without technically disturbing an ancient ruin of some kind. It's unsurprising that this might be the case right off the coast too. Still amazing to look at when the water is clear in aerial pictures.
As an aside, historical preservation was used as a pretext for artificial housing supply restrictions in Europe much earlier than in the US. Eventually US property owners caught on. Now any old 20th century box is revered like a Haussmannian mansion in Paris.
1/20 is less than my guess would have been. It's interesting that Wikipedia is heavily weighted in LLM training corpuses. Eventually there will be a feedback loop not unlike the one among meat popsicles.
How many covers are better than the original? Would we expect that to be possible with AI?
One might assume that a house.gov URL would imply some degree of credibility. This particular link, however, is closer to something that would be found on the ground of said bat cave.
> "Long Covid" took quite a while to be demonstrated and we were monitoring everything like a hawk. "Long flu" has no such monitoring.
It wasn't demonstrated because of the monitoring. It was demonstrated by the droves of desperate patients who kept banging on the clinics' doors and wouldn't take no for an answer. Claiming that 'long flu' is the same thing is (IMO) just a different variety of Covid denialism.
I'm excited at the prospect of LLMs being deployed here. An attacker only needs to find one weak link in the chain. Eliminating weak links is hard, and might be NP-complete.
However, throw resources at it and you might just make weak links much rarer. Throw a variety of heterogenous LLMs at it, and you might be looking at a large force multiplier.
This is true. However, we've had zoonotic spillovers since before the dawn of civilization, and the Covid pandemic stands out in modern times in the ways I listed.
There's no reason to suspect the next pandemic will stand out in the same ways, or as much in general, particularly if it's Influenza.
If you can't help it, this is called Non-24-Hour Sleep-Wake Disorder (N24SWD) [1]. If it's just because you're not exposed to natural light a whole lot, circadian drift is "normal," though my recollection is that ~25 hours is more common. I guess it gets diagnosed if and when you complain to a doctor that you can't help it.
Covid was "special" in a lot of ways (though of course it wasn't a lab leak.) Flu pandemics can be extremely deadly, however:
- 'Long flu' is not a thing (it's actually possible but it's rare, mild and short)
- Our baseline immune competence would probably be much higher (even though neutralizing antibodies to H5N1 are low in the population.) This should be particularly true for people whose first flu as children was with Influenza A.
- Unlike Covid, no reason to suspect H5N1 will cause diabetes, heart disease, or various other kinds of permanent damage. It's probably also unlikely to cross the blood-brain barrier.
- There are various vaccine efforts already in the pipeline.
I wouldn't expect it to be like Covid in 2020 but maybe like in 2023, after it went from the 3rd leading cause of death to the 10th.
Some use cases require a small memory footprint, e.g. parallel inferences. I suppose there are also dark patterns like tracking, where you don't want the load to stand out.
Assuming you're right, the vertical platform vendors should consider having a public Sherlocking policy. If you're anywhere near Apple's stuff, the Pixelmator outcome is a large fraction of your upside.
Much larger than it should be for the ecosystem's sake. Excessive cannibalism isn't probably in Apple's interest even.