dqv

I'm frowning upon your use of a good experience that one person had in a completely different context to overshadow a bad use of AI that affects hundreds of thousands of people. They both happened in healthcare, but other than that, they are different situations with different outcomes.

I made a single presumption: that the PCP's use of AI for this situation is not mandatory. Is it mandatory or is it not?

In 2024 when the AI empathy snakeoil machine was inflicted on the nurses, the nurses had no choice in the matter. And management has already established that they may do something like this again.

Your situation is completely different. It's a doctor overseeing their own tool. If the AI makes a mistake, the doctor is responsible and would presumably take responsibility. Who is responsible if an AI tool makes a mistake in a performance evaluation of an employee? The employee! The employee is guilty until proven innocent. But the employee gets no choice in whether that AI tool is used against them (absent a union agreement which dictates how AI tools can be used).

The reason I fear the misuse of AI is because it is so useful.

They already did a trial which used the trendy buzzword to surveil the nurses. The nurses want to implement controls on that trendy buzzword before it is further misused by management to cause harm to patients (by way of making the nurses focus on meeting some arbitrary performance metric rather than focusing on patient care).

... They're aware that it's individuals in management, which is why they want to use their union contract to dictate how management can use AI.

You're seriously saying it's surprising that your PCP had a different response to AI that is presumably not mandatory for him to use? Cmon dude.

No way to know what data sources NielsenIQ is using, but for what it lacks in data from retailers, it probably makes up for by way of consumers' receipts.

I don't know how to estimate the impact tbh. The same article links to https://www.bain.com/insights/weight-loss-drug-users-spend-l...

How much does 7% of the population spending 4% less on groceries impact the numbers in the original article?

A product advertised as having a 1 year warranty and then the seller not honoring the warranty is the product not being sold as described.

But yes, it's my mistake for assuming the purchase protection would... protect my purchase.

And now I won't make that mistake again. The value isn't there, neither with their aggregation service nor their credit product.

I can just use Google Pay with a credit card, both of which are more convenient and offer better protection than PayPal does respectively.

Nah I thought I liked PayPal as a customer until I found out that they favor large merchants against the consumer. I had a problem with a product and the company was playing games with the warranty. I only had it for 3 months and it was already falling apart. I made a complaint with PP and it was auto denied.

I will never use PP for a large purchase again.

Yes, at the lower end of those salaries, you have to do things (and this is one single example btw) like have roommates, which drastically increases your housing instability. Why? Because you have to rely on your roommates' ability to pay rent. If they don't, what happens? You lose your PrIvIlIgE to live there! Most places in the US do not have protections for this. So you can try really hard to find another roommate, but you have a very limited amount of time to do so (basically until the next rent payment). So then you get evicted. Now your ability to find a new place to live is damaged because you have a permanent "evicted" mark on your credit/tenant report. Oh, and of course, the landlord can go to court and try to get the court to force you to pay the remainder of the rent or at least continue to pay the rent until they lazily find a new tenant. So now you're financially damaged too. There are so many footguns like this in the US that people, on HN especially, are ignorant to. Imagine trying to keep a job while also dealing with all these shenanigans. Can you even get time off work to go to the court hearing? If you don't, the landlord can get a default judgement.

Oh and sorry because this is HN and people will automatically assume I mean owning the property and renting rooms out to others, I mean both people are renters in this case.

So the stereotypes are true. My friend in elementary school told an amusing story about his friend, in a Swedish family, who moved from our nice suburb to a suburb in another state, one like the grandparent is describing. They didn't understand why there was such a vast difference between the two. It seemed obvious to me, but I guess it's not obvious to foreigners.

You ran right past the GP's statement too:

yet with terrible distribution

applies to the quality of a suburb too. Obviously. You wouldn't catch me dead in a Texas suburb or Florida suburb.

Yes, this is where the disagreement between the two camps arises. Some people view reading holistically while others are only concerned with the end result. If only the end result matters, it doesn't matter much how you got there.

This isn't a perfect comparison (since I believe there may be some underlying mechanical difference), but it's like how it's possible to get your nutritional needs met with liquid nutrition or eating whole foods. They have the same end result nutritionally, but some people find liquid nutrition to be horrible.

I think it's especially the case for harder-to-read books that I prefer the textual version. Some books are highly referential. They may take place in a country I'm not familiar with. Take The Stranger by Camus. I have never been to Algiers (or Algeria for that matter), so I want to get a feel for the beach in Algiers. Sometimes the author will refer to a flower I've never heard of, so I will look it up. Stuff like this is important to me for experiencing the story, but for others, it's not important. They use context clues to determine "it's a flower" and just fill in whatever flower they feel like putting there.

It's not a maybe. People's brains work differently.

For one thing, in the written version, it's way easier to scan backwards precisely to a point in the text to clarify missed details.

Sometimes I like to stop reading for a moment and take in the scenery. With an audiobook, the narrator has already decided to blurt out the next line. Or I'd have to quickly pause it and lose my train of thought. With reading I'm fully in control of when the next bit of information will be introduced.

Audiobooks are great for dialog-heavy stories, but sometimes I want to spend a little bit longer on a particular page when the author is building imagery... what's the equivalent of that in an audiobook, slowing the speech down? Except it's not about speed, it's about soaking in the particular meaning of a word (or string of words) before going onto the next.

I don't think this even scratches the surface of how different they are as mediums of storytelling.

https://www.openculture.com/freeaudiobooks

As others have pointed out, libraries often have Libby access which can have pretty huge selections of audiobooks. There's a discovery feature that lets you search by vibe, which I am finding useful.

"this absolutely positively must be a group effort" (eg, the military).

Which, by extension, means education absolutely positively must be a group effort, even if taxpayers can't understand the n-order effects of severing tax money from education. If society needs the military, then it definitely wants soldiers who understand figurative language so that those soldiers don't shoot themselves in the foot. Being able to communicate well, critically think, understand history - these all come from the humanities (although the learner must go beyond the entry level courses to benefit).

It goes

Subject: Information about Your Automatic Renewal

This is an automatically generated email from Nintendo for customers who have a Nintendo Switch Online + Expansion Pack individual 12‑month (365‑day) membership set to renew automatically.

Dear [user],

Your Nintendo Switch Online + Expansion Pack individual 12‑month (365‑day) membership will automatically renew soon.

...

Deadline to turn off automatic renewal: [1 month from now]

It also does this right when you first sign up for automatic renewal except the deadline is [1 year from now].

Which algorithm is the hot garbage one to you? SuperMemo 2?

Finding out about this program was such a relief. I haven't worried about the prospect of having to call 911 since signing up.

That's not true. We have a lot of generalist bees (and other pollinators too! bees aren't the only ones that pollinate!) that can pollinate the Eurasian crops. This is also true for any other crop on any other continent. And I am dubious as to your claim about major crops. Soybeans, for example, are self-pollinating, so it would seem any generalist bee can pollinate soybean crops for better yields.

This is an issue of biodiversity. If we rely on a mono-species for pollination and that mono-species goes extinct or its population plummets, then our crop yields are down 20% until we can build up other bee populations.

This is about giving back to Nature so that She may continue to cradle us. And She does tend to punish us for "optimizing" toward one specific species.

We do enough already taking crops from one part of the world and putting them elsewhere. In exchange for that, we really ought to be trying to support the native wildlife as much as possible. The Europeans can have their honey, and I will have my maple syrup. On occasion, of course, so as not to upset Nature.

I suppose honey bees are not native in North America pretty much the same way as the human species?

No? Well not in a way that wouldn't be stretching an owl over a globe. But Carolina Jessamine is toxic to honeybees and not natives (or at least there exist native bees who have adapted to not slurp on it if it is toxic to them). That doesn't stop people from spreading the lie that Carolina Jessamine "hurts bees". It hurts some species of bees. To transfer this concept to the human population, you'd have to start arguing that there are different species of humans or, again, construct a stretching-an-owl-over-a-globe argument.

And people can't mention every caveat in every discussion, sorry. You've really just constructed a strawman.

In a 40-minute discussion with someone like Doug Tallamy, both the issue of invasive honeybees and pesticides will come up. The venn diagram of people who care about both things is very close to a circle.

Also, as to your edit - that honeybees rely on humans doesn't change their impact on native bee populations, which is they outcompete native bees.

There's nothing weird about correcting the popular ignorant assumption that the only pollinator that matters is honeybees.

You know, I thought, surely, by now, capitalists would be calling for reform to capitalism to prevent the weird ponzy scheme from shutting down, but it's the same clichés! Not even an interrogation of why people are souring to capitalism, just the same old tired response.

Make sure you do 5 prayers to Bezos and another 5 to Thiel on your capitalism rosary ($14.99/month subscription). If you've run out of prayers for the month, they actually have a sale for the 4th of July, $8.99 gives you 12 more prayers! Golly, I could use those two extra prayers to pray to Musk.

Precursor to age verification gateway.

In the future, an AGEnt will attest that you are old enough to access the resource.

But that just goes back to the banks again. Why did the Washington politicians encourage easy mortgage lending standards? Lobbying by big banks. https://scholarship.law.gwu.edu/faculty_publications/1301/

In a similar vein with CAFE standards: why were those loop holes introduced? Lobbying by the auto industry.

Yeah, but I have no idea how good it is or how expensive it is: https://www.dnb.com/en-us/smb.html

They are acquaintances and friends, although not close friends. It's also unclear what their biases are since I can't read their minds, so there is no way of knowing if their biases are harmful for their lives. If I knew what their biases were, I would probably have had better understanding of what the output meant to them; hence the conclusion that we do not share the same biases.

I see emotional intelligence as a tool, not some virtue to uphold. Here it is identifying the value of a relationship and avoiding potentially breaking that relationship apart simply because one aspect of it is not optimal. Not every relationship needs to have perfect alignment with my beliefs.

But more importantly, I don't have enough information about them to properly navigate a conversation about their use of the LLM for advice.

I just really don't like the idea of pretending or lying to people with their whole lives ahead and going counter to what you believe to be the truth for some social comfort or emotional intelligence aspect.

Fine, but you're also doing that for your own comfort, and not necessarily for the benefit of the other person. It can (and often does) cause the friendship to end before you've actually helped them understand what you want them to understand. I've lost friendships this way.

I'd rather keep the relationship intact and revisit the conversation when the time is right than risk abruptly ending the relationship with the same outcome where they go through their whole life with malformed or harmful ideas. You incorrectly assume that they would "know" the alternative views just by communicating, but much like their biases allowed them to "understand" the output of the LLM, their biases may disallow them from understanding the alternate views you are trying to impart.

Emotional intelligence. These interactions aren't people sharing points of view, they are sharing something they see as meaningful, and no matter how politely I convey it, it's still going to come off as "that sounds stupid and meaningless". It's not like I have better advice to give. Not worth dying on that hill.

This is my read on how people react to LLM responses, especially about personal/social issues too - that many are fooled by the LLM. On occasion, someone has shown me LLM output, believing it to be a profound and illuminating response to their prompt.

Even being able to see the original prompt, the response just seemed kind of empty (blocking out the advertisingesque magnificent language leaves little substance behind), but somehow they were convinced that it was a coherent answer to their prompt and even peeved with me for not sharing their astonishment at how well it addressed their concerns.

It's a kind of symbiosis between the LLM and the prompter, where the prompter's own cognitive biases fill in what the LLM can not. Since I don't share their same biases, the response is nearly meaningless.

Very uncomfortable experience having to pretend to agree with how profound it is to not make their prompter upset.

Pentagon spokesperson Sean Parnell said that the Pentagon had granted exceptions to Hegseth’s optional flu shot policy to the Army, Navy, Air Force, National Security Agency, and the Defense Health Agency.

So which ones are still exempt from the vaccines? Space Force, USMC, Coast Guard, who else?

"overengineering"

    UPDATE appointments
    SET starts_at_utc = local_time AT TIME ZONE timezone_name
    WHERE timezone_name = 'America/Vancouver'
      AND starts_at_utc > now();

In all likelihood this will never happen in a particular timezone.

You sure about that?

https://lists.iana.org/hyperkitty/list/tz-announce@iana.org/...

2026b - changes to future timestamps

2026a - changes to past and future timestamps

2025c - changes to past timestamps

2025b - changes to past timestamps

2025a - changes to future timestamps

2024b - changes to past timestamps

2024a - changes to future timestamps

2023d - changes to past and future timestamps

2023c - changes that changed future timestamps reverted

2023b - changes to future timestamps

I definitely prefer the... fragile? approach for this problem

It's just a different kind of complexity and one that requires having a library that can arbitrarily load different versions of tzdata.

I've been thinking about it for a while though - a time zone conversion library that also accepts an additional "tzdata_version" argument.