jonas21

The Walmart greeter also isn't paying for the bulk of their healthcare expenses because Walmart provides subsidized health insurance to all employees who work at least 30 hours per week. All US employers with at least 50 employees are required to do so under the ACA. If the greeter worked fewer than 30 hr/wk, they wouldn't get insurance through Walmart, but they would likely qualify for an ACA subsidy that covered close to the entire cost of a health insurance plan on the marketplace.

The statement, "The US spends ~$14,570 per person on healthcare. Japan spends ~$5,790" is about the average amount that the country as a whole is spending per person on healthcare, not what any given individual is paying. Per-capita GDP (i.e. the average economic output per person) is the most relevant comparison.

They are substitute goods. A common failure mode is not realizing this until it's too late.

I wish the article had more technical details. Obviously, 2048 being a power of 2 stands out as being possibly related.

> I think this is fine if it also then means that obtaining a qualifying ID is treated as a no-cost and highly-accessible right for all citizens.

This is essentially what the Supreme Court said when they upheld Indiana's Voter ID law in 2008 [1]:

> The burdens that are relevant to the issue before us are those imposed on persons who are eligible to vote but do not possess a current photo identification that complies with the requirements of SEA 483. The fact that most voters already possess a valid driver’s license, or some other form of acceptable identification, would not save the statute under our reasoning in Harper, if the State required voters to pay a tax or a fee to obtain a new photo identification. But just as other States provide free voter registration cards, the photo identification cards issued by Indiana’s BMV are also free. For most voters who need them, the inconvenience of making a trip to the BMV, gathering the required documents, and posing for a photograph surely does not qualify as a substantial burden on the right to vote, or even represent a significant increase over the usual burdens of voting.

> Both evidence in the record and facts of which we may take judicial notice, however, indicate that a somewhat heavier burden may be placed on a limited number of persons. They include elderly persons born out-of-state, who may have difficulty obtaining a birth certificate; persons who because of economic or other personal limitations may find it difficult either to secure a copy of their birth certificate or to assemble the other required documentation to obtain a state-issued identification; homeless persons; and persons with a religious objection to being photographed. If we assume, as the evidence suggests, that some members of these classes were registered voters when SEA 483 was enacted, the new identification requirement may have imposed a special burden on their right to vote.

> The severity of that burden is, of course, mitigated by the fact that, if eligible, voters without photo identification may cast provisional ballots that will ultimately be counted. To do so, however, they must travel to the circuit court clerk’s office within 10 days to execute the required affidavit. It is unlikely that such a requirement would pose a constitutional problem unless it is wholly unjustified.

[1] https://supreme.justia.com/cases/federal/us/553/181/

2007 was 19 years ago. If you step back another 19 years, you'll find that the major tech companies of the era had huge defense contracts: IBM, HP, Oracle, SGI, Texas Instruments, etc. Not only that, the development of many technologies we take for granted today -- like integrated circuits, the Internet, even Postgres -- were directly funded by the DoD. Much of the growth of Silicon Valley in the early days was a direct consequence of working with the military.

That location information is not available to apps or ad networks without user consent. The government can access it from the carrier with a warrant, but that's not what we're discussing here.

No, it wasn't chosen at random -- it had to be a question that any reasonable person would immediately recognize as harmless, but where the old model would inject a bunch of safety caveats and the new model would not.

It's basic physics, the sort of example you might find in a high school textbook.

I assume that is the point of having a Greenpeace USA -- to shield Greenpeace International and other Greenpeace organizations from liability. And it seems to have mostly worked.

Google Maps' high-resolution "satellite" imagery is actually captured from planes.

Antarctica is huge (1.5x the area of the US), it would be a dangerous logistical nightmare to fly the sorts of patterns you need to capture aerial imagery there, and it's almost entirely covered in non-descript ice -- what would be the value of having high-resolution imagery there?

This is also a low effort comment, despite the word count.

In contrast, shubhamjain found Meta's earnings release for the specified time period, quoted numbers that appear to contradict the claim, and provided a link to the release. This adds to the conversation, while a comment that says "Source?" or a few paragraphs that can be reduced to "Source?" do not.

I mean, if you go outside and everyone else is carrying an umbrella, it's probably going to rain.

If you don't meet the financial requirement ($200K annual income or $1M net worth), you can also qualify as an accredited investor by passing the Series 65 exam and filing a form with the SEC.

So you have to prove that either you can afford to lose some money or you have enough investing knowledge to know what you're getting into. Seems fair.

Upvote the conversations that you find to be more interesting. If enough people do the same, they too will make it to the top.

If you say: "Generate a strong password", then Claude will do what's reported in the article.

If you say: "Generate a strong password using Python", then Claude will write code using the `secrets` module, execute it, and report the result, and you'll actually get a strong password.

To get good results out of an LLM, it's helpful to spend a few minutes understanding how they (currently) work. This is a good example because it's so simple.

> a human brain operates on 12 to 25 watts

Yeah, but a human brain without the human attached to it is pretty useless. In the US, it averages out to around 2 kW per person for residential energy usage, or 9 kW if you include transportation and other primary energy usage too.

Here's Claude's translation of the PDF in the repo:

https://claude.ai/public/artifacts/0c40c3f8-16de-4947-93c1-3...

I couldn't verify it, and a human translation would be preferable -- but it's probably good enough to get an idea of the story if you want to read some right now.

Why? Very few nonprofits contain that language in their mission statements. It's certainly not required to be there.

There are dozens of spam and security settings that admins can change in the Google Workspace console, presumably because different businesses have different requirements. So in practice, there's not just two sets of rules in gmail -- there's probably thousands or millions (however many combinations of settings are actually in use).

What do AI or GLP-1s have to do with a casino economy?

> If I could destroy these things - as the Luddites tried - I would do so, but that's obviously impossible.

Certainly, you must realize how much worse life would be for all of us had the Luddites succeeded.

It's correct in the same way as saying ads in the New York Times don't change the articles. Seems fair.

So that you can be using the current frontier model for the next 8 months instead of twiddling your thumbs waiting for the next one to come out?

I think you (and others) might be misunderstanding his statement a bit. He's not saying that using an old model is harmful in the sense that it outputs bad code -- he's saying it's harmful because some of the lessons you learn will be out of date and not apply to the latest models.

So yes, if you use current frontier models, you'll need to recalibrate and unlearn a few things when the next generation comes out. But in the meantime, you will have gotten 8 months (or however long it takes) of value out of the current generation.

> There is also a solenoid inside the enclosure, but since the keyboard is not powered up, this was not being activated — and these are seemingly particularly loud! But why? The teletypewriters, paper terminals and card punches which preceded such early “glass teletypes” would have been very noisy indeed, hence it was decided that terminals like the IBM 3278 would need to provide much more positive feedback than simply that provided by a clicky switch module.

Reading this, I thought it could not possibly be true -- the keys are so loud and clicky already. But it turns out it actually was the case, as demonstrated here:

https://www.youtube.com/watch?v=ZT9CHub9Cxs

Yeah. Waymo described how this works a couple of years ago:

https://waymo.com/blog/2024/05/fleet-response/

The top-level comment was edited. It originally said that it was chillingly familiar to us or something along those lines.

Isn't Little Snitch exactly the sort of application they're worried about?

> I'm willing to accept robotaxis, and accidents in robotaxis, but there needs to be some solid figures showing they are way _way_ safer than human drivers.

Do you mean like this?

https://waymo.com/safety/impact/

You owe the same amount in income taxes regardless of whether you're paid in stock or cash.

I think you're confusing this with the case where founders or early employees own large amounts of stock that they received early on in the company. They paid little to no taxes when they received it because it was worth basically nothing at the time. Later, the company has grown and it's worth a lot, so they have a large unrealized gain that they can use as collateral for loans. But that certainly wasn't guaranteed to happen. Most startups fail.

> At this point it seems pretty clear that LLMs are having a major impact on how software is built, but for almost every other industry the practical effects are mostly incremental.

Even just a year ago, most people thought the practical effects in software engineering were incremental too. It took another generation of models and tooling to get to the point where it could start having a large impact.

What makes you think the same will not happen in other knowledge-based fields after another iteration or two?