stupendousyappi

I don't blame them for using a different config language, k8s style YAML sucks, which is why half of the k8s ecosystem is tools to generate it for you. And I believe Agent and Alloy are meant to be installable outside k8s as well. I do think it would have been less confusing to just call this product Agent 3.0 instead of Alloy, but I'm not too concerned about naming decisions.

Wow, I wish I'd discovered this about 12 hours ago. I just built a new PC yesterday and started installing and configuring Silverblue today, and basically everything I spent this afternoon on is automated by this project. Getting VSCode devcontainers working with Silverblue + Distrobox in particular would have saved me some time. Since I'm not particularly invested in my Silverblue setup yet, I'm installing this now.

This is how I feel about Vlad's statement on their Discord that he wants to provide email services. That strikes me as a big project in terms of both development and operations, somewhat contrary to their privacy focus, and not something that their users are asking for (on Discord, I'd say that most commenters said they were unlikely to use it any time soon, switching would be too much of a hassle). And while I subscribe to Kagi and love the search product, they're still a tiny company with no long term track record, and far from what I would feel comfortable trusting with my email, which IMO is much more sensitive than my search history. Meanwhile, although Kagi Search already reliably beats Google IMO, there's still a ton of stuff they could do to improve their main product.

Bamboo grows faster in compatible climates.

The benefits described sound contradictory to me. If you're removing enough heat to reduce the likelihood of Yellowstone erupting, you're removing enough to reduce the power output of the plant. Perhaps there's way to balance those two outcomes, but I know that other geothermal plants have sucked their heat reservoirs dry over decades of use. I suppose you could do a better job of matching the rate of heat extraction of the rate of natural regeneration, but I imagine that would enormously reduce the potential output of the plant, and that's not an issue this paper addresses.

Only if the employees are already accurately sorted by value.

I'm reminded of George RR Martin's quote (spoken by Varys) that "Power resides where men believe it resides." Over time technology makes it easier for small numbers of people to control larger numbers, and eventually I expect AI to bring that principle to its logical conclusion, but for now, Varys's observation remains the foundation of modern society in most places.

Orcas are almost certainly several different species as well, but they remain officially classified as one until scientists agree on how to divide them up.

I've read that the the Hawaiian islands disrupted ocean swells in ways that Polynesians could detect even 1000+ miles away. In addition, the Hawaiian islands are huge and lush, and dumped a large amount of plant debris into the water, some of which Polynesian sailors would occasionally notice while at sea. I'm skeptical that those signs would have been noticable all the way from the Marquessas islands, but the Marquessas Polynesians were likely ranging 1000+ into the ocean themselves, which evidently was enough for them to deduce the existence of large islands in the general direction of Hawaii long before they discovered it. Evidently according to oral tradition, the discoverers of Hawaii arrived equipped with plants, animals and other gear to establish a new settlement- they knew the islands were there before anyone set foot on them.

I assume these are all currently non-standard APIs that Firefox will nonetheless be expected to implement? Does Firefox ever force work on the Blink team, or it is it only ever the other way around?

IMO, if that's true, it's horrifying. I think there's a great risk that AGI will lead to the extinction or near extinction of the human race.

They may be doing it to stun the animal.

What is the browser extension incident you're referring to? Can you provide a link to an article or previous HN discussion?

I believe they switch to Hebrew letters.

That's based on two very cautious geoengineering approaches that have virtually no negative side effects. If governments were willing to consider other geoengineering approaches, I think that number could be driven down significantly. My favorite candidate is ocean-wave-based olivine weathering, as proposed by Project Vesta. They think that, at scale, that approach could get costs down to $21/ton, plus reduce ocean acidification more effectively too. But it's more complicated and could have some negative effects, such as putting a lot of poisonous heavy metals like nickel in the ocean. But because politicians hate ever having to say that they're knowingly causing any problem, those avenues get starved of research funding, as we see here. But I think that, within 50 years, we're probably going to find a scalable CO2 removal approach for less than half the target cost of this program, and that basically could solve global warming, but because of political cowardice it's discovered 30 years later than it could have been.

Everyone wants to do emission reduction first and delay geoengineering as long as possible, when we should be doing the opposite. Even if the financial cost appears much smaller, it's clear now that large scale emission reduction is politically very expensive. Emission reduction is the clean, ideal solution that we don't have the ability to scalably implement yet. Geoengineering should be temporary quick and dirty approach we use to buy time, creating some problems that last decades in exchange for time to implement a solution for a problem that lasts for millennia. Assuming that cheap geoengineering techniques whose negative side effects are bounded in space and time can be found, but I think that they can.

Google certainly didn't support Amazon forking Android into FireOS. I believe that was just about when Google began migrating tons of APIs from the Android OS to Google Play Services.

Over the very long term, I see no reason to think that higher CO2 levels are fundamentally worse for life than pre-industrial levels. In general, plants are the foundation of the food chain, and higher CO2 levels are better for them. In addition, life depends on liquid water, and given that surface temperatures are generally closer to the freezing point of water than the boiling point, I could see how higher temperatures might increase the amount of land and ocean surface hospitable to life. The problem is the transition period, during which most things go extinct until evolution can slowly regenerate the lost diversity over 10+ million years.

There are several others. I'm not a climate scientist, but my understanding is that the most significant potential negative feedback loop by far is the potential for increased cloud cover from increased humidity. It's evidently the most difficult aspect of global warming to model, although climate scientists are making continuous progress. In addition, the marginal radiative forcing of CO2 decreases with its concentration, which means both that increases in emissions do less harm as things get worse, and that reductions in emissions do less good.

Noah Smith has an article on Substack arguing that economists have largely failed to produce useful research on climate change, in part because of a preoccupation with carbon taxes to the exclusion of other research topics. The article is subscription only (edit: not true, see reply below), but his four big points are these:

* First, academic economists have basically ignored the topic. A survey of top economics journals found that, out of 77,000 total articles published, only 57 (0.074%) were about climate change.

* Many of the most cited papers have turned out to be crap, with basic calculation and data coding errors the authors have had to publish corrections for. Smith argues that this is partly because economists don't collaborate with other researchers much, especially natural sciences like climate change.

* The most important model for integrating climate change into macroeconomic analysis is the DICE model created by Richard Nordhaus. But the DICE model has big problems, the most important of which is that, due to discounting of future economic effects, it basically ignores the welfare of future generations. The DICE model also assumes that preventing climate change will be very expensive, and hasn't adjusted for recent technology advances on that front.

* Last, economists have been obsessed with carbon taxes, and haven't dealt with how politically unpopular they are, especially in international negotiations. This is, IMO, similar to how economists love to promote free trade, saying that the losers can be compensated via money from the overall economic gains, leaving everyone better off, and ignoring that this never, ever actually happens.

I'm in the camp that thinks that economists haven't earned a ton of credibility on the specific topic of climate change. Our best shot, IMO, is massive R&D efforts and CO2-removal geoengineering (especially wave-powered olivine weathering, as promoted by Project Vesta). Accompanied by crippling taxes on heavy emissions industries, but I think people will have an easier time accepting those if it's clear that governments are pursuing alternatives to taxes too.

Then hope the UK government approves the ARM acquisition.

I mentally tune out anyone who uses the word "elitist". To me, it's one of the clearest signs that they're arguing using prejudices rather than evidence.

Breyer was a prominent copyright law scholar before he became a judge. Between that, and the fact that he hasn't gotten many opportunities to write majority opinions in his 27 years on the court, and the fact that he's likely about to retire in a few months, giving him the opinion was the obvious and collegial thing for Roberts to do.

I'm biased, but I don't think the technical elements of this case were challenging to an intelligent layman. I think most people smart enough to become judges can understand the concept of an interface that is independent from an implementation, which is all this case really required, along with quantitative estimates of the amount of code involved. So I doubt that he needed clerks to understand any of it.

The line between "sharing his dream world" and "committing securities fraud" is evidently in the eye of the beholder. I admit that I'm surprised he doesn't get sued more for his blatant lies.

Filibuster.

Do you have any specific criticisms of that index, or just generalized ideological grievances? Moreover, are you saying that the UK is not meaningfully more democratic than China, and therefore that index is inaccurate?

I think public opinion is pretty clear at this point that no one cares about the specifics of the legal arguments made in any court cases other than lawyers and activists trying to gin up arguments for their side, and perhaps a very small number of people who care about the rule of law for its own sake. Everyone else just cares about the outcome of the case. No credibility is at stake, because 95% of the population has no interest in the contents of legal briefs.

Just to use their most recent scandal, what about their claim that Richard Hopkins, a postal worker in Pennsylvania, was told to backdate a ballot postmark? After which the post office inspector general investigated and reported that Hopkins had denied that claim?

I'm guessing you'll claim that Project Veritas was truthfully reporting what Hopkins told them, and it's not their fault that he changed his story once he talked to law enforcement. Even though he also told the investigators that his initial affadavit was written by Project Veritas.

Not much of a sacrifice, the best OLEDs are currently made by LG, Samsung and Vizio.

This question is equivalent to "Why doesn't the Istanbul government spay and neuter stray cats?", a question the article doesn't address. If there really are no systematic efforts to control the stray cat population, I have a hard time believing that nature isn't performing the same function. Cat populations grow explosively unchecked- if this phenomenon has been happening for hundreds of years, there's no way that the cat population hasn't been butting up against various limits by now, with all the suffering that that implies.

This, except for geoengineering. Developing the infrastructure to apply it at planetary scale will take years, and then actually applying it will take trillions of dollars. If by some miracle it turns out to not be necessary, we have plenty of time to cancel it, but now (really, 10 years ago) is the time to start spending 100s of billions of dollars preparing (and 100s of billions reducing emissions, of course).

FWIW, my preferred approach is ocean-powered olivine weathering, as in Project Vesta. But we'll need multiple approaches.