Modern "skills" and Markdown formats of the day are no different than "save the kittens". All of these practices are promoted by influencers and adopted based on wishful thinking and anecdata.
Uh, this couldn't be more false. I've implemented these from scratch at my company and rolled them out org-wide and I've yet to watch a youtube video and don't consume any influencers. Mostly by just using the tools and reading documentation - as any other technical tool.
Perhaps your blanket statement could be wrong, and I would encourage you to let your mind be a bit more open. The landscape here is not what it was 6 months ago. This is an undeniable fact that people are going to have to come to terms with pretty soon. I did not want to be in this spot, I was forced to out of necessity, because the stuff does work.
To be fair, if you have never watched a YouTube video in your life then how can you say the OP was wrong about what influencers are peddling? Side note, have you ever seen that Onion article on the man that can't stop telling people he doesn't own a TV?
Great, so how do you know this stuff works? Did you evaluate it against other approaches? How do you know it's actually reliable?
The Vercel team had some interesting findings[1]:
> In 56% of eval cases, the skill was never invoked. The agent had access to the documentation but didn't use it.
Others had different findings for commonly accepted practices[2], some you may have adopted from reading documentation, which surely didn't come from influencers.
And yet others swear by magical Markdown documents[3].
So... who is the ultimate authority on what actually works, and who is just cargo culting the trendy practice of the week? And how is any of this different from what was being done a few years ago?
Sorry, but from your first comment, I don’t particularly feel inclined to help you figure this out. I was just offering I’ve already deployed these things at a scale with success using many of the configuration options offered as documentation in the op here. this stuff isn’t some mystical blackbox, although you seem to think it is.
I measure the tooling success with a suite of small prompt tests performing repeatable tasks, measuring the success rate over time, educating the broader team, and providing my own tried and tested in the field skills that I’ve shared to similar successes to the broader teams. We’ve seen a huge increase in velocity and lower bug rate, which are also very easily measurable (and long evaluated stats) enough to put me in the position I am, which was not a reluctant one. You’re perfectly free to view my long history on this topic on this forum to see I am a complete skeptic on this topic, and wouldn’t be here unless I had to.
everyone is figuring this out still. There is no authority, I am my own authority on what I have seen work and what hasn’t. Feel free to take of that what you will. I just wanted to provide a counterpoint to your initial claim. I’m certainly not going to expose to a fine degree what has worked for my org and what hasn’t due to obvious reasons.
I’m a little unclear on the usage of the word “fake” here.
Going by article, these are real people doing actual real work, they often use stolen identities to conceal information about themselves, and they get help from outside sources to do their jobs better.
Whatever the right word is, it’s not “fake”. Maybe fraudulent? Or ulterior motives? Or deceptive? Or pretext? Or threat actor? Or foreign agents?
I agree - this is closer to bonded labor though the paying employer doesn't know it. Instead most of their earnings go to their actual employer (which is the North Korean state). "slave" maybe is more appropriate? "prisoner"?
I know we're getting deep in the meta discussion but the free will that you're describing involves basically starving to death. Sure, you can walk away but unless you're well off, we all basically live in the same society that makes sure you are ALWAYS dependent on some kind of wage. You cannot live off the land, build housing, or eat food without some kind of income in the modern world. And thus the concept of wage slave.
But wage slavery, while bad, isn't slavery still. In slavery proper, the option of walking away straight up doesn't exist. In fact, in extreme cases, even the option of dying might not be available.
It is slavery. Chattel slavery is much more severe than what we normally consider slavery. Yet “slavery” and chattel slavery are both still slavery. The reason what you’re saying is so accepted is because we are currently living under a universal liberal world order that says wage slavery id freedom.
I hope you notice I didn't mention chattel slavery. Even prior to it, all forms of slavery were about removing the agency of person and subjugating the will of the slave to the owner. That requires an active action.
Not hiring someone is a passive action. As said by many, you are not entitled to a wage. In fact, suggesting otherwise would actually require slavery. Wage slavery, instead, is a description of a particular material condition of destitution, not necessarily connected to the ethical evaluation to proper slavery.
No one says "wage slavery is freedom". What the "universal liberal world", that is, the pro-free market side says is that people should be free to associate with each other as they see fit. Being hired to provide labor in exchange for wage, the basis for wage work, is merely an extension of this. While wage work is a requirement for wage slavery, at no point economic liberals said that everyone should live under wage slavery conditions.
The common, orthodixical, sociological/economical meaning of the word "wage slavery" is about being paid, on average, barely enough to make a living, i.e. destitution in the conventional sense.
I suppose you are referring to the Marxist meaning, technically (at least as far as I know) original, meaning. First, Marxist economics are considered heterodoxical nowadays. Second, it is still about "destitution", in the sense that the working class is formally destitute of the means of production, requiring to sell their labor to have access to it. If that's the case, I hope you notice that weakens your point of "wage slavery being a form of slavery", as you lose the analogy of proper material conditions.
Sounds like having a w2 is a pretty good deal for you then.
Slavery isn’t defined by “I don’t want to talk away because the deal is too good”, it’s more like “I’m unable to walk away because I’m threatened with force if I do so”
My dad used to refer to that as the golden handcuffs when he worked for GE. Wouldn't compare it to slavery though, he just felt trapped there because nobody else would pay him that well or give him as good of benefits
The implication is that they're pretending to be legitimate employees whereas they are actually exfiltrating IP from a hostile nation state. Seems valid.
"Fake" seems fine. If I buy a fake watch, that might mean that it's a real watch that does its job of telling time, but it says "Rolex" on the front and that's a lie.
I agree that fake is an odd word to describe this. Most likely much of our IT infrastructure is extremely compromised. I wouldn't be surprised if some of the major password/healthcare/etc... leaks in the past 6 years were the result of someone "accidentally" setting a cloud bucket to public.
I actually turned down a fly-to-texas for an in person interview about a year back, but I do think in the age of the internet if we don't sacrifice some of the things we have taken for granted in the past, we're going to lose our country. Perhaps there should be a law that requires a picture of any employee standing next to their boss for continued employment - at some point in the future. (this is just an idea, not to start a flamewar, don't attack the specific idea, but attack the idea of some kind of extra checking if you don't agree with it)
Labeling the actual worker negatively seems harsh - they are probably being forced into it by the state. You might say they can willingly underperform and not be used this way - but if the alternative is a much harder life, could you blame them for playing along?
It's North Korea though and they're all eViL. Imagine a world where the U.S lifted sanctions on N.K. traded with them and stopped crying about losing a war 70 years ago. Ah well a boy can dream.
Edit:
Lol saying anything positive about North Korea on hacker news and people instantly freak out. This fucking website man. North Korea isn't what I would call a free society but it's also not the hell on earth that most liberals want you to think it is. So much of the misery that normal North Koreans have to face is because of western imposed sanctions. We've tried punishing them for 30 years now, it hasn't destroyed the regime if anything they double down. I guess it's easy for a bunch of overfed over paid tech workers to not feel any kind of solidarity for a North Korean though and insist on punishing them even more. Hell the North Korean government would even be open for this kind of agreement if we would actually guarantee their sovereignty, sadly trusting the United States of America to hold up any kind of deal you make with them is fucking impossible.
Here is a quote I came up with but is attributed to Henry Kissinger
Having the United States as your enemy is dangerous, but having them as your friend is fatal.
That old bag liked it so much he had no problem taking credit for it.
Lot's of people have tried trading with North Korea, but they're politically unreliable. China and Russia both try obviously, but so has South Korea. These cooperations usually work for a while but eventually the unreliable reality of the North Korean government wrecks it for them. If it were all America's fault, as these sort of regimes always claim, they'd be able to get on well enough with their neighbors, but they can't.
The United States plays a large role in destabilizing them I went to a lecture at my university where a South Korean professor said as much. He was hardly a fan of the North Korean regime. At this point the regime has zero interest in cooperation, I'm sorry but your government is slowly becoming an authoritarian state in its own right and is currently causing chaos at the behest of Israel a country which just commuted a genocide with the blessing of both parties in your country. Imagine trying to get along with your neighbor when they have billions of dollars of military hardware on your border. No country is to willing to cooperate with North Korea because being in the good graces of the United States is 100x more beneficial. You claim that North Korea can't get along with its neighbors please remind me which country invaded and artificialy divided Korea when they elected some one The United States didn't like.
The US has been authoritarian for a long time. What else do you call a society that keeps on humming along while doing various genocides via a culturally embedded Monroe Doctrine mentality.
Korea was divided by both United States and the Soviet Union. The Soviet Union organized elections, rigging towards a rather unpopular figure, even within the national socialist circles, for their imperialist purposes.
Also, correct me if I'm wrong, but where did OP state was american? How is their nationality even relevant here? How is the american descent to authoritarianism, which is still far from a autocratic socialist regime (at least yet), relevant to NK being distrusted even by USA's opposition (i.e. China and Russia)?
The US has fuck all to do with it. Vietnam whooped America's ass in a war which was far more socially significant for the American public (the Korean War is called the "Forgotten War" in America), still has their communist government, yet has normalized (relatively) relations with America and certainly the rest of the world and trades with everybody. North Korea is economically isolated because they refuse to be normal even by communist standards.
I don’t get what you’re saying. There is a paused civil war due to outsiders (whites, UN, etc) interfering and still occupying half the country.
Not sure what you mean what crime did the current outsider occupiers commit after they did a genocide (and not using this lightly. There aren’t that many genocides)?
Sob stories are available against/for every major human conflict. The story you linked isn’t really relevant even to the severity of what we are talking about.
None of the sides are occupied, though, neither by UN, "white" people, etc. That's literally just propaganda. At best, by a large stretch too, one could say NK is occupied by Russia and China, the only countries currently vouching for NK, and, as previously stated, that's a stretch given their reluctance towards Kim' personal kingdom. It is more accurate to say NK serves as a buffer state to them.
What genocide are you referring to? The Shichon Massacre? The one only NK sources claim outsiders commit, for the sole purpose of promoting xenophobia and ultranationalism?
Finally, I think you straight up didn't read the reference. How exactly kidnapping 2 movie directors just to produce propaganda pieces are justifiably helpful to the war effort? I hope you have the decency to realize that this isn't a war tragedy, where innocents are killed/multilated despite best efforts. This is a large private corporation kidnapping 2 civilians from the other side for strictly private purposes.
If the relevance to the discussion isn't clear, the point is that North Korea, even as a Stalinist-inspired socialist country, doesn't need to commit the crimes it does to survive. You are just trying to dismiss NK's particularly rogue behavior under the excuse that "west"(which, in this context, is just a propaganda term, could've used just USA instead) oppresses them, ignoring even other socialist countries that do not have to stoop so low.
Vietnam didn't submit to America, they kicked Americas ass. You need to get a clue. I can't help but notice you're refusing to even acknowledge the point of Vietnam, because it makes you look like an idiot.
Come to think of it, what is a leftoid like you even doing defending North Korea? You should be disowning them. You should be pointing out that North Korea is a de facto monarchy and therefore definitionally Right Wing. You should be arguing North Korea as yet another failed extremist right wing regime. Why do I have to explain your own ideology to you? Is it that you like to get dominated? Come to me, dumb slut, we can be friends. I don't discriminate.
While one could scrutinize the accuracy the of the west being a dictatorship (which will likely devolve into a discussion of semantics), I prefer to call attention to the fact that calling North Korea a dictatorship does not require to speak well of the west. Likewise, one can simultaneously criticize the west without protecting its "enemies". Such a binary, poorly critical, way of thinking is ill-suited in pursuing better material conditions for all.
Wow so you guys don't have ideological brain worms at all. I can tell you guys have never studied international relations because you refuse to try an take North Koreas concerns seriously. Do I have to remind you that a South Korean attempted a fascist coup recently and that it was left wing organisations and trade unions that mobilised to stop him?
Also which "communist dictators" are you talking about, and how do you know this? You do realise that the United States is also a one party state when it comes to foreign policy right?
> Wow so you guys don't have ideological brain worms at all.
You have to remember that the supermajority of this site are ultra AnCaps who believe that anything which infringes upon the right of companies to kill people is Communist satanism and a significant minority agrees wholeheartedly with Peter Thiel's weird brand of techno-accelerationism and actively participates in NRx movements.
Like, I understand what you want to get at and I wholeheartedly agree! Just don't be too surprised at the pushback.
Those scholars are blowhards. Trump is basically the same as any other time. As you state the US is unipolar in how it treats Others. Aka authoritarian.
> Do I have to remind you that a South Korean attempted a fascist coup recently and that it was left wing organisations and trade unions that mobilised to stop him?
Those "left wing organizations" encompass the majority of South Korea, almost 2/3 of the current parliament, and the current sitting government. And may I remind you that South Korea's current "leftist" president recently gifted Trump a golden crown to get favorable deals?
> Do I have to remind you that a South Korean attempted a fascist coup recently
By "fascist", I suppose you mean right-wing. Going by the average RW authoritarian dictatorship, that's still better than the documented NK conditions, specially given that most fall later.
> ...[...] and that it left wing organizations and trade unions that mobilized to stop him?
Nice cherry picking. Even Yoon's own party turned against him.
But even ignoring the right wing here, said left wing organizations are also in opposition to NK.
> You do realize that the United States is a one party state when it comes to foreign policy right?
That sentence doesn't make sense. There's no such thing as "one party state foreign policy". The idea of a one party state is specifically about a state that is intolerant to any other ideas other than those accepted by the one true party. If you are referring to USA's aggressiveness, may I remind you that that switches between presidents.
> If you love freedom so much shouldn't this be worrying more?
Who cares what they're called. Main concern in this case is that the result of their work poses danger to the US. Like a spies. They often do legit work and meanwhile some "extra"
I find to be true for expensive approvals as well.
If I can approve something without review, it’s instant. If it requires only immediate manager, it takes a day. Second level takes at least ten days. Third level trivially takes at least a quarter (at least two if approaching the end of the fiscal year). And the largest proposals I’ve pushed through at large companies, going up through the CEO, take over a year.
The PR won’t take 5 hours of work, but it could easily sit that long waiting for another engineer to willing to context switch from their own heads-down work.
Exaxtly. Can you get a lawyer on the phone now or do you wait ~ 5 hours. How about a doctor appt. Or a vet appt. Or a mechanic visit.
Needing full human attention on a co.plex task from a pro who can only look at your thing has a wait time. It is worse when there are only 2 or 3 such people in the world you can ask!
Exactly. Even if I hammer the erstwhile reviewer with Teams/Slack messages to get it moved to the top of the queue and finished before the 5 hours are up, then all the other reviews get pushed down. It averages out, and the review market corrects.
It just sits there, with no one touching it. Suddenly, music randomly starts and stops playing. Take it into the Apple Store, they acknowledge it’s a known hardware defect to start registering non-existent touches, and they refuse to fix it. Offer to replace it with a refurb unit for like ~$20 less than a brand new unit.
FWIW we use the minis all over our place and love'em. But we pretty much only ask basic questions or use them for airplay. Sounds quality is impressive imo.
FWIW I think “Please don't complain that a submission is inappropriate. If a story is spam or off-topic, flag it.” is different from the others.
It’s an instruction for how to use the site. It’s helpful to have it in the guidelines for when the flag feature should be used. Without it, the flag link is much more ominous.
Maybe it could be consolidated with the flag-egregious-comments rule?
Edit to add: IMHO it is not at all obvious on this site that flagging stories is meant to be roughly the equivalent of downvoting comments (and that flagging comments doesn’t have a counterpart at the story level).
By this logic, you might consider vibe coding a browser plugin that takes any HN comment less than 50 words and auto-expands it into an “insightful, well thought-out response.”
I mostly agree, although we've seen big shifts in the culture towards rule-deviating norms over time. Look at the guidelines for ideological battles or throwaway accounts, for example. And, as always:
> Comments should get more thoughtful and substantive, not less, as a topic gets more divisive.
That’s true. Fortunately, by virtue of it being added to the guidelines, quite a few folks here are prepared to reply to obviously generated comments by simply citing and linking the rule. Just search for “shallow dismissal” to see many examples.
It will take time, but eventually everyone will know about it.
> quite a few folks here are prepared to reply to obviously generated comments by simply citing and linking the rule
Note that the guidelines do explicitly say not to post about guidelines violations in comments, and to email them instead. I know this isn’t a well-loved guideline in this modern era, but duly noted: those well-intended comments are themselves breaking the guidelines.
> Please don't post insinuations about astroturfing, shilling, brigading, foreign agents, and the like. It degrades discussion and is usually mistaken. If you're worried about abuse, email hn@ycombinator.com and we'll look at the data.
If so, that seems different. If not, can you clarify?
That one, yes. “Insinuations” is a less conditional form of “Accusations”, connected by the concept of “Claims”; they’re all synonymous from a general perspective:
- I insinuate that you are a bot (often shortened to “Is this a bot?”)
- I claim that you are a bot. (often shortened to “This is a bot.”)
- I accuse you of being a bot l. (often shortened to “Are you a bot?”)
The part where I’m interpreting to include accusations of bottery and slop is “and the like. It”; the first clause, ‘the like’ refers to the generic category of accusations against posted comments, which historically were the listed examples, but is also defined to include others not listed, such as today’s popular accusations of bot or AI; the second clause, ‘It’, refers to all insinuations-class content. Without the list of examples, this reads:
’Please don’t post insinuations. It degrades discussion
Yep, this is true. Accusations, Insinuations, Claims, of bot or AI or astroturf; they all wreck discussions and I end up having to email the mods to deal with them. A lot of people use the rhetorical device of Discredit The Opposition by invoking this sort of thing, and while that’s less prevalent in ‘reads like AI’ insinuations, they still degrade the site.
With AI-assisted writing is a violation of site guidelines, and even before it was, posting of AI-assisted writing was a clear ‘abuse’ of the community’s expectations of unassisted-human discussions. Aside from expectations, I can also classically understand in Internet history that ‘violating the guidelines’ is the phrase formerly known as ‘abuse of service’, by which I interpret the above reference to abuse to refer to breaking the guideline about posting accusations.
The guidelines are not a legal contract as program code, and perhaps this one is clunky enough that it needs to be reworded slightly; thus my intent, once the flames die down here, to let the mods know about the confusion. As I’m not a mod, this is my interpretation alone; you might have to email the mods and ask them to reply here if you want a formal statement on the matter, given how many comments this thread got in a couple hours.
ps. On ’and is usually mistaken’: I’m not a mod, so I can’t judge how often accusations of AI/bot are mistaken, but I’m also an old human who learned em-dashes in composition class, so I tend to view the modern pitchfork mobs out to get anyone who can compose English as being less accurate in their judgments than they believe they are.
What constitutes “at edited”. If I throw a block of text in to an ai see if it makes sense — say a response to a post — and fold the suggestions in, is that “ai edited”?
If you look at what you wrote and can't identify what rules you've broken, how are you able to validate that the AI output doesn't change the meaning of what you wrote?
Knowing whether or not the AI changed the meaning of what you wrote is not reliant on knowing which specific rules you broke. It's only reliant on you actually reading what the AI spat out and deciding “yes, this is what I meant” or “no, this is not what I meant”.
Unless you're arguing that the rule violations are something the author intends to be part of the meaning of what one wrote?
>Knowing whether or not the AI changed the meaning of what you wrote is not reliant on knowing which specific rules you broke. It's only reliant on you actually reading what the AI spat out and deciding “yes, this is what I meant” or “no, this is not what I meant”.
That's fair.
>Unless you're arguing that the rule violations are something the author intends to be part of the meaning of what one wrote?
I think what I wanted to get at is more like this:
1. I think that they may be part of the meaning
2. I think that people would be primed to accept changes even if they change the meaning
3. I suspected that it would always correct something and wouldn't just say LGTM even if the input was fine
To check, and at the risk of this being hypocritical, I asked for a grammar correction on part of your post that I thought had no mistakes, and both in context and isolation, it corrected "spat out" to "produced." Now, this isn't a huge deal, but it is a loss of the connotation of "spat out," which is the phrasing you chose.
I think grammatical errors are low-cost, and changes in meaning and intent are high-cost, so with 2. above, running it through an LLM risks more loss than it gains.
I suspect those relative costs would be very different for someone who's not me, though. My English writing ability is much higher than my Spanish writing ability, so in some alternate universe where Hacker News was Spanish-only instead of English-only, grammatical errors would add up to be a cumulatively-higher cost than a possible change in meaning/intent. I wouldn't have the requisite knowledge to know the difference in connotation between “produjo” v. “escupió” v. the myriad other verbs Kagi Translate is suggesting to me at the moment, whereas I'd probably have a lot of cases of not just bad grammar, but outright nonsensical word choices — like Peggy Hill in a Mexican courtroom (https://www.youtube.com/watch?v=b7QCvykBXik), except with the added torment of being actually self-aware of my linguistic ineptitude.
(On that tangential note, though, I do appreciate that Kagi Translate provides multiple translations and attempts to explain their differences in connotation such that I can pick whichever one most closely matches my intent; if other LLM-assisted writing tools did that then that'd render a lot of this problem moot.)
You're absolutely right! It's not the people correcting their Grammer that are the motivation for this rule, it's the people abusing these tools and ruining every online discussion with cookie-cutter comments.
In all seriousness, if you use some tool to make sure you're using the right "there", noone will mind. Just don't generate another boring predictable comment and everything will be ok
Um, why would you do that instead of waiting for someone more knowledgable to reply, and learn from? Replies are not mandatory, and experts/insiders participating is one of the best parts of the human Internet. Let them shine.
It can catch things that I might miss or might be misinterpreted. I sometime miss simple things, like like repeated words, that an AI point out. Is a spell checker considered "AI"? Is Grammerly? Okay, maybe Grammerly from 5 years ago as opposed to today? If I'm typing on my phone and it pops up the next suggested word, is that AI edited?
And no, I don't have to reply to a post, but when I think it's a bad policy, should I just accept it without discussion? And who determines the "experts/insiders" and which voices should be allowed?
Yes, these are MY questions and feelings too. In the past, if I just HINTED at asking these kinds of questions, I was downvoted into oblivion (in other accounts. I have to say THAT specifically because some people here dive in to my account and get super anal about my age, my previous comments, my moniker, ad nauseum)
>Um, why would you do that instead of waiting for someone more knowledgable to reply, and learn from? Replies are not mandatory, and experts/insiders participating is one of the best parts of the human Internet. Let them shine.
As Isaac Asimov pointed out[0]:
“Anti-intellectualism has been a constant thread winding its way through our political and cultural life, nurtured by the false notion that democracy means that 'my ignorance is just as good as your knowledge.'”
This thread runs through many cultures and isn't just a problem on the Internet, although the Internet certainly has accelerated/worsened the problem. And it has created a distrust of experts which (as has been obvious for a long time) has made us, as a whole, dumber and less informed.
I recommend The Death of Expertise[1] by Tom Nichols for a sane and reasonable treatment of this issue. If books aren't your thing, Nichols did a book talk[2] which lays out the main points he makes in the book. During that talk, he also gives the best definition of disinformation I've heard yet.
Again, the question is who blesses the expert? There’s a difference in having a voice and your voice being taken seriously.
If someone posts a link on a a new laptop, who should respond? I am not an expert on the current laptop market, but I have options about it. Maybe my English is not the best so I run through an AI to clean it up of ambiguities or wrong wording. Maybe I say “I like to take my laptop from behind” when I meant “I lift my laptop from the back”. An AI could point out this type of error.
Yes, and it’s not like they just insta-ban every infraction.
I’ve broken the guidelines on this site before. The mods reply and say “hey, stop doing that, here is the guideline”. I stopped doing it. Life continues.
This discussion reminds me of the Paradigms of Power featured in Adiamante by L E Modisett; about consensus, power, morality and society. It’s a good read.
IMHO most of this “customize your config to be more productive” stuff will go away within a year, obsoleted by improved models and harnesses.
Just like how all the lessons for how to use LLMs in code from 1-2 years ago are already long forgotten.
reply