<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: carsareok</title><link>https://news.ycombinator.com/user?id=carsareok</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Tue, 26 May 2026 18:20:45 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=carsareok" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by carsareok in "The User Is Visibly Frustrated"]]></title><description><![CDATA[
<p>Oh yes, the part where actual crucial non-derivable information was left out certainly comes into play as well. I suspect many people underestimate the sheer magnitude of implicit cultural and organizational knowledge they and their colleagues carry with them. It's always part of their context in everything they do, so to speak and they expect the "bots" to carry this as well (without giving it to them).</p>
]]></description><pubDate>Tue, 26 May 2026 08:36:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=48276861</link><dc:creator>carsareok</dc:creator><comments>https://news.ycombinator.com/item?id=48276861</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48276861</guid></item><item><title><![CDATA[New comment by carsareok in "The User Is Visibly Frustrated"]]></title><description><![CDATA[
<p>Because it's a completion engine and has no notion of "signals".<p>Swearing was in the texts they were trained on to complete token by token. I suspect it weren't texts with a lot of high-quality reasoning.</p>
]]></description><pubDate>Tue, 26 May 2026 08:28:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=48276802</link><dc:creator>carsareok</dc:creator><comments>https://news.ycombinator.com/item?id=48276802</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48276802</guid></item><item><title><![CDATA[New comment by carsareok in "The User Is Visibly Frustrated"]]></title><description><![CDATA[
<p>Yes, I have definitely witnessed this as well.<p>I think, I hope, this will be fixable to some degree, but at this moment I believe it's best to communicate in Queen's English and try to maintain the level of clarity of thought you expect of them in return.<p>My pet theory is that actual real conversations they were trained on with bad grammar and spelling are in general relatively starved of proper reasoning. By talking to them in this fashion you activate their lowbrow patterns and while it may not be catastrophic I can't imagine it helps.</p>
]]></description><pubDate>Tue, 26 May 2026 08:22:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=48276745</link><dc:creator>carsareok</dc:creator><comments>https://news.ycombinator.com/item?id=48276745</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48276745</guid></item><item><title><![CDATA[New comment by carsareok in "The User Is Visibly Frustrated"]]></title><description><![CDATA[
<p>Instead of reacting directly to the issue at hand I suggest you ponder what failure mode is being activated and why.<p>They are fundamentally not able to tell truth from fiction, but this also means they don't make errors like we do. They definitely create output we recognize as errors, but that's very different from our failure modes and you have to get used to it.<p>In my opinion it's better to branch off with an altered context that somehow avoids or mitigates the issue you're running into. Let's say they miss the mark. If you tell them "Don't do that" in the "conversation" this means the error is now and forever part of the context (assuming you stay within context limits and no compaction). Depending on their training this may or may not be detrimental to the quality of the rest of the conversation. You are now entering a section of their training where "error + someone swearing at them"-conversations have happened. I can't tell for sure, but my gut says this is not an advantageous place to be.<p>They are as I'm sure we all know completion engines and are in a very real way constantly cosplaying being productive "agents". They don't know if they are part of some type of modern Shakespearean play where sitting behind computers is part of the story or if they are in what we call "reality". By training on "conversations" they have become more likely to complete their input in a way that mimics what we call having a back and forth with some degree of technical accuracy.<p>In the extreme case you have a context that starts like "Please make all junior mistakes in this assignment. Make the code unreadable and be sure to include massive gotchas in subtle parts of the logic.". The results of this context won't be pretty. The other way around is not saying "Please make no errors", it's explaining in detail what you think is the right way. Coding style, if you care, architecture, etc. it all needs to be part of the context if you suspect it will substantially impact the completion. You have to imagine what real-life conversations have started with "Please make no errors". Again, I have no proof of course, but I have a strong feeling that human conversations that started with clearly and properly articulated specifications are qualitatively different from human conversations that started with "make no errors". In one you can see the pointy-haired boss and the other a seasoned engineer. Try to stay on the engineer side of their training.<p>I completely agree that they should be trained (or instructed) to react in a robotic tone stripped of all human pretense. We are trying to get at useful, general reasoning patterns latent in the data they trained on and, I regret to say, not the "human" parts which are usually a masterclass in cognitive biases and failures to reason.<p>Edit: the last sentence should be read in the voice of the Matrix's Architect.</p>
]]></description><pubDate>Tue, 26 May 2026 08:14:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=48276689</link><dc:creator>carsareok</dc:creator><comments>https://news.ycombinator.com/item?id=48276689</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48276689</guid></item><item><title><![CDATA[New comment by carsareok in "Ferrari Luce"]]></title><description><![CDATA[
<p>I'm one of those people that doesn't care for cars. They are equipment to me. I like "getting places", yes. But I don't like "personality" in my tools. Cattle, not pets. I don't want to drive around looking smug in my 650k shit bucket. Cars are an enormously wasteful, idiotic drain on the world, but the calculus is such that I am "forced" to own one. I find the idea that each of us is owning and maintaining our very own special little box that exudes "personality" preposterous and I'll bet the farm that future generations will think we were mental.<p>This is not apathy in my opinion. This is rational. Cars are just tools. Metal boxes to enable mobility. Car people have turned them into this cult of personality that I think is batshit insane. It's not just cars mind you, we do this with watches, shoes, you name it and it's all very peculiar, but cars are my pet peeve because they are so obviously wasteful and dangerous. Not just directly like killing 40k per year in the US alone, but also through obvious geopolitics.<p>People want to move around and they want to smile smugly and think they are better than others. Those two things are pretty much universal. I say we separate those issues. You can move around all you want but smiling smugly you do in some other way than in your "car". We'll have really good public transport and you'll assert your dominance in some other fashion. I personally recommend we reintroduce dueling to the death.<p>By the way I don't know anybody that would buy a new car every two year to keep up with the Joneses and I live in a pretty "Jonesy" place. That's a bit hyperbolic at least in my neck of the woods (Netherlands). Most people here keep their cars until they become unreliable.</p>
]]></description><pubDate>Tue, 26 May 2026 07:27:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=48276307</link><dc:creator>carsareok</dc:creator><comments>https://news.ycombinator.com/item?id=48276307</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48276307</guid></item></channel></rss>