Hacker Newsnew | past | comments | ask | show | jobs | submit | Miraltar's commentslogin

Instead or in addition to following blogs, what I'd love to have is a way to filter out those I don't like.

Local keyword exclusions (to keep the server requirement minimal) might be pretty high impact.

Well, if you're supposed to administrate shocks to teach or test someone's memory, asking the question while they're screaming isn't just about protocol, it does break down the purpose of these shocks. Saying that participants did administrate shocks because they trusted the legitimacy of what they thought they were doing doesn't hold up under these circumstances.

No, because you'd have to show that the participants thought there was a breakdown of the procedure and purpose, and that they continued despite that.

If they think the procedure is to read the next question when the previous one has been completed, and they do, even if the other person is screaming, they think they're "following rules". They're not the ones who came up with the procedure.

Which is the whole point: the participants were trying to follow rules, even if they made mistakes in following those rules. The idea that there was a total "breakdown" of the rules doesn't seem supported at all.


Fair point, but there's a logical relationship between 'testing someone' and 'following a set of instructions that don't achieve that effect'.

Your point is fair, but what is really nuanced is that the people who 'stopped' were the best ones at following the rules.

This seems interesting to me - they were conscientious about 'what was happening' - not just blithly following orders.

The 'rule followers' maybe were conscientiously applying the 'spirit of the test' and quit when they realized it was not reasonable.

The others were 'pressing buttons'.

Even then, it's subject to interpretation. There's a perfectly rational reason why people might subject to 'following the rules' if that's what they've been asked to do and have a sense of 'dutiful civic conduct' and 'trust in institutions'.


This reevaluation postulates that the participants didn't deviate by mistake, but deliberately. The participant could have waited for the respondent to be in a state in which they could answer. (Reminder: the exercise was officially about answering questions, not enduring shocks).

Instead, most participants rushed through, most likely to end their own negative experience. Which is much more nuanced that "gosh, they told me to do it."


If I'm not mistaken, they were told the point of the experiment was supposed to be about "memory and learning". If a teacher was doing a "commission" as they put it, they aren't really following the purpose of the experiment any longer.

Context is important. Maybe that was told in the first 3 minutes of the briefing, and them came 30 minutes about the shocks. I would not assume the briefing was so thorough.

> HAL is decidedly second-tier. Given the option, everyone would pick arXiv over HAL.

Can you elaborate on that?


Guidelines say "please use the original title, unless it is misleading or linkbait" in that case I'd argue that it's linkbait so changing it is justified


> Looks like it is happening - AI bringing the end of theory publications

To keep the original, with added clarity from the opening paragraph. Could use an emdash for irony.


Create some sort of score that goes up when a "child" misbehaves. The further the child the lower the increase but at some point you get banned anyway


Your detector did not work well on an AI-collab writing fiction project I did a while ago, tagged it as 100% human even with high confidence for the most part. But to be fair, most detectors weren't significantly better, although this one gave a justification that made sense https://aidetector.com/


Stopped reading there: "This model (Moshi) could [...] recite an original poem in a French accent (research shows poems sound better this way)."


Is it still true today?


Are you asking if a drug and sex fueled lifestyle is still pursued today by musicians? The answer sadly is yes. There are lots of musicians that don’t go full throttle but most of the highly famous ones are running the red line. This is the OP’s point. Whether it’s the person amplified by the lifestyle that makes them famous or it’s the fame that enables the lifestyle that allows them to destroy themselves.

The road is littered with smashed guitars.


Speaking of roads, everyone points out lifestyle choices, but the lifestyle of popular bands/musicians is also countrywide or worldwide tours. It doesn't look like an easy life, so I wonder to what extend those excesses are related to being on the road maybe half of the year? I think this means no true social life for extended periods of time; not having people you value telling you that you're past that red line is one less safety.

Also, artists in general are a peculiar profile I think. It's not only famous singers that take drugs, commit suicide etc. One can easily find many writers and painters, some of them even only became famous postmortem.


I'm guessing they used the same argument that was made for calculators, printing... By lessening the burden on your brain you weaken it.


I think it's fine when you're just working on your own but if you have to review code or work in a codebase you don't own that's much more effort.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: