gipp

Says who? The quadrants would just be the combination of the two applicable types

Those are the labels for each end of each axis, not for the quadrants. Believer/grifter is one axis, coaster/grinder is the other.

The first half of the title is pretty well-covered, but it seems like at no point does the author even explain what they mean by the term "infrastructure of energy," right? I assume they're not talking about electrical substations

Yeah, figuring out a remotely accurate figure for that is at least as hard as the work done here, and requires pretty different sets of expertise

> I'm assuming solutions like this are uneconomic

You kinda answered your own question already, I feel. The energy efficiency of cycling a battery (70-90% for grid scale) or pumped hydro (70-85%) is simply much, much higher than chemical storage. Here's a pretty recent one [1] showing 23% efficiency even at lab scale, and as described in the article scale is a big drain on efficiency.

[1] https://www.nature.com/articles/s41467-022-29428-9

Those are both explicitly arguments that the author made in the article, how is he missing the point?

He didn't say the lack of carbon tax acts doesn't act as a subsidy. He agreed that it does. His point was failing to distinguish between that lack, versus literal direct payments from a government budget to fossil fuel projects, is confusing and misleading. Which is true, because while both are effectively subsidies, the incentives and challenges in addressing them are wildly different.

I've only seen the posters, an early trailer, and a few paragraphs about it here and there, but I've been surprised this comparison hasn't come up more often. Looking at the poster and the plot synopsis, it's all I can think of.

And three just happen to accept such offers within a day of a major structural transition, but unrelated to it?

> Do that many people even work on AI development?

That's not the relevant figure. It's more about how much capital has been wagered by large tech cos and the major startups & VCs on a near-complete takeover of white-collar work by AI (as that's the only level of payoff that could possibly justify the mind-boggling level of investment).

The whole discourse that has exploded around this phrase, and the amount of self-seriousness and dogma about it in this sites comment sections have been really baffling to me. One overconfident billionaire writes an essay about a phrase some speaker had made up weeks before, and suddenly everyone's acting like the concept of "founder mode" is some metaphysical constant, and it's our job to understand its true nature through theological debate.

> Tech has become all Jobs and no Woz.

Distills the post to a sentence, and boy does this nail it.

A mile down is, as already stated in the article, way below any water table. There is no groundwater there to contaminate.

Seems obvious they wouldn't want to provide further visibility to it. In the interest of journalism.

Google is the one who has spent years and presumably many millions trying to make it happen, and has only recently thrown in the towel

I'd say it's more or less consistent with QFT -- it's saying essentially the same thing, just not in that exact language

Tl;Dr: "Wave-particle duality" is not the notion that matter is "sometimes" a particle and "sometimes" a wave. It is, at all times, its own separate category of thing, for which "particle" and "wave" are just metaphors that approximate its behavior.

One metaphor usually comes closer than the other depending on what system you're looking at, but it's never changing back and forth between some "particle state" and "wave state".

Let's put it this way. Opposable thumbs are such a massive fitness advantage, in so many ways that are obviously massively more impactful, that such a factor, if it even exists, would be utterly negligible.

So the only conceivable reasons to raise such a bizarre topic all involve you being a very unpleasant person to converse with -- the most generous I can think of being an unearned sense of intellectual superiority simply by the fact of raising an idea you see as taboo.

Yeah, it's not like the utility of opposable thumbs is exactly a mystery

> I'm in agreement with the author: without more detail or pushback from the Google Cloud leaders, this is a really bad look for future customers.

That's exactly the opposite of what he said:

> [...] putting out a competing statement blaming or contradicting your customer is a bad look with that customer and with all future customers.

Income -- being funneled through the unifying medium of US dollars usually, and having some other counterparty to the transaction available to verify things -- is a lot simpler to account for than wealth, and look how complex that still is, and how readily it is already circumvented. And that's with a hundred years of refining and reforming the tax code to stop it, which a wealth-based taxation scheme would not benefit from.

How are those "ridiculous scenarios?" Any attempt to base laws around the net worth of an individual is going to run into a hundred thousand such scenarios instantly. A person's "wealth" is simply not well defined. Neither ownership nor valuation of any kind of asset, other than cash itself, is nearly straightforward enough.

Cool, see you in a million years when that incredibly noisy effect converges

Yes, sure, but it's also extremely common to just use "fired" colloquially for either case (in the US at least)

? I'm not confused about anything, I'm explaining why it's a terrible decision. People aren't buying it (at least not now that it's actually out and reviewed, pre orders aside), not just because it's stand-alone hardware but that certainly isn't helping matters.

That's their reason to make it a stand-alone device. What is my reason to want it to be a stand-alone device?

One of my higher-voted comments on Reddit was in response to some thread bemoaning the decline of the site towards lowest-common-denominator meme content and recycled jokes. My comment was pointing out that people had been saying that for years, with multiple links to past, nearly identical threads.

I made that comment in 2012.

I mean... you pretty much just repeated what the article said, though. You agree with him.

If nothing else, this post is kind of conflating 2 different things: (a) do search rankings display a bias, and (b) is a "revolving door" and close relationships leading tech companies to deliberately induce such bias. I think it's important to distinguish them, because I think (a) is obviously true, and (b) is probably mostly false. Managing bias is a wildly difficult thing to do, especially with something as complex and correlated with everything as politics. There is no such thing as an "unbiased" ranking, just like there's no such thing as "unbiased" journalism.

No it's more like saying it's all just different regions of a single function. Individual electrons are just regions of that function that have particular recurring kinds of structure to them

I see this sentiment a lot on here, always with a vague intimation that it's a widespread thing. But it is just so far out of sync with the way I use search I just can't wrap my head around it.

You say 90% of you usage, but then describe the sorts of things that make up only a tiny fraction of mine. 90% of mine, maybe more, is just very quickly looking up basic facts. Was this the same actress from that other show? How many Xs to a Y? What was the exact wording of that quote and who said it? Where's that one store? I do this, I dunno, probably dozens? of times a day.

Are you saying you use it for that kind of purpose? I don't think I would trust any LLM's answers to those kinds of questions -- at least, not to exclusively handle all of them, on an ongoing basis. Even if it's only making up bullshit 5% of the time, after long enough that means I have no idea which 5% of my knowledge is wrong. Unless I'm double checking the information elsewhere, in which case what's the point?

Meanwhile Google still supplies the right answer for those simple, basic facts immediately like 99% of the time. A lot of the time right there in a snippet where I also instantly know the attribution. Even if we perfectly solved hallucination, the friction to get it from one of these chatbots is still so much higher. Harder to get to, so much slower.