<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: JohnMakin</title><link>https://news.ycombinator.com/user?id=JohnMakin</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sun, 23 Aug 2026 01:09:00 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=JohnMakin" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by JohnMakin in "Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces (2025)"]]></title><description><![CDATA[
<p>code changes change the compiler output.</p>
]]></description><pubDate>Sat, 22 Aug 2026 00:23:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=49395356</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49395356</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49395356</guid></item><item><title><![CDATA[New comment by JohnMakin in "Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces (2025)"]]></title><description><![CDATA[
<p>I guess I will expand on what I meant why I react to claude memory acting mechanically or logically anything like human memory, is because it isn’t how memory in the brain works, they’re not comparable.<p>The layman’s understanding I have of memory, as someone that has dealt with memory issues much of my life, is that memory formation is heavily tied to emotions. emotions are triggered by input which sends a complex set of signals throughout the brain - you’re not just finding where in your head to store this, your brain is deciding how important it is, and what else to correlate it with - so it can tie them to other related memories. then on top of all this, much of the sensory experience you intake is subconsciously compared against high priority memory impressions and deciding what to pay attention to.<p>you could, argue that the sensory input is the simple md files and the emotional mechanism is the same effect as to how attention mechanisms work in llm’s. Ok, I can almost buy that, but these tools lack a fundamental ability to decide how important things are.<p>an analogy. you tell a person “if you pick a daisy in the next five years, an assassin will come to kill you” and they hold a knife to your throat while they say it, your brain whether you like it or not is going to say “THIS IS AN IMPORTANT MEMORY I NEVER MUST FORGET” and you’ll see something that looks like a daisy and have a panic attack 3 years later. that memory is never fo tell me claude or other tool harnesses using  memory harnesses can prioritize memories the way that human would, instead they forget even when reminded, because the human brain is just so much better at it</p>
]]></description><pubDate>Fri, 21 Aug 2026 00:32:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=49382179</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49382179</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49382179</guid></item><item><title><![CDATA[New comment by JohnMakin in "Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces"]]></title><description><![CDATA[
<p>Just a heads up - this person in gp comment, their original post in a way that is much much different than the one I was replying to originally both in tone and content.</p>
]]></description><pubDate>Fri, 21 Aug 2026 00:21:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=49382089</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49382089</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49382089</guid></item><item><title><![CDATA[New comment by JohnMakin in "Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces"]]></title><description><![CDATA[
<p>I just don’t find it really relevant to the argument presented I guess. I disable auto memory and have my own mechanisms and infrastructure with how my agentic system “knows” and “remembers” things which is roughly an automated, sometimes self-correcting working index on the file system. It behaves much better than claude’s automated “memory” system, so I use that, but digging into how that worked and making something of my own just makes me really dismissive of comparing it to something like actual memory, so I apologize if it came off dismissive.</p>
]]></description><pubDate>Thu, 20 Aug 2026 23:45:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=49381830</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49381830</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49381830</guid></item><item><title><![CDATA[New comment by JohnMakin in "Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces (2025)"]]></title><description><![CDATA[
<p>Even claude with “memory” enabled isn’t really “remembering” anything. It just injects it into the context and you hope it happens to find it relevant in its attention mechanisms, and then remembers to actually act on it. Anthropic’s own documentation states claude can and will ignore/truncate these. It’s a context trick, nothing approaching actual “memory,” and in fact, arguing with it will make a bunch of memory files, sometimes contradictory, and clutter up the context and act even worse.</p>
]]></description><pubDate>Thu, 20 Aug 2026 21:06:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=49380281</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49380281</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49380281</guid></item><item><title><![CDATA[New comment by JohnMakin in "Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces"]]></title><description><![CDATA[
<p>I don't think anyone in this conversation is saying this behavior is anything but the fault of the user not understanding how these tools work? This is a weirdly aggressive post.</p>
]]></description><pubDate>Thu, 20 Aug 2026 17:38:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=49377715</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49377715</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49377715</guid></item><item><title><![CDATA[New comment by JohnMakin in "Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces (2025)"]]></title><description><![CDATA[
<p>Do you? I’m not sure what you’re getting at.</p>
]]></description><pubDate>Thu, 20 Aug 2026 17:26:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49377574</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49377574</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49377574</guid></item><item><title><![CDATA[New comment by JohnMakin in "Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces (2025)"]]></title><description><![CDATA[
<p>Because plenty of people, even ones that should know better, really believe it's a conscious, thinking entity, not just some turn of phrase.<p>I have a coworker that spends at least 10 hours a week arguing with his like you would with a conscious person. I've gently tried to explain it's like arguing with your compiler for giving you an incoherent error message - it's pointless. It doesn't understand, it can't understand, and even if it could, you arguing with it isn't going to make it "learn" or act differently.</p>
]]></description><pubDate>Thu, 20 Aug 2026 16:35:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=49376937</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49376937</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49376937</guid></item><item><title><![CDATA[New comment by JohnMakin in "Feature Request: Support AGENTS.md"]]></title><description><![CDATA[
<p>this is so easy to work around though. it’s a file name. it doesnt even care about symlinks. if you want this badly it is so easy to do yourself,  It’s the weirdest thing to complain about as someone who uses a multi provider system locally.</p>
]]></description><pubDate>Thu, 20 Aug 2026 03:11:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=49369950</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49369950</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49369950</guid></item><item><title><![CDATA[New comment by JohnMakin in "Feature Request: Support AGENTS.md"]]></title><description><![CDATA[
<p>cp AGENTS.md CLAUDE.md<p>what am I missing?</p>
]]></description><pubDate>Wed, 19 Aug 2026 23:25:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=49368499</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49368499</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49368499</guid></item><item><title><![CDATA[New comment by JohnMakin in "Remote workers report the highest well-being in study of 7,700 employees"]]></title><description><![CDATA[
<p>> lack of visual boundaries between work and home life,<p>This was the major one for me, in addition to the other stuff you mentioned. The "always at work" feeling was reinforced by the management style of remote work being like, we expect you to be generally available to respond at any time in a 12 hour window, sometimes more (because everyone works their own schedules across multiple timezones). In an office this separation is clean, and if I'm responding to messages or working at home, it's generally accepted this is outside the norm and I'm perfectly free to just shut the laptop off until the next morning 99% of the time unless I am on-call. It's nice.<p>The first thing I did when I went back fully in-office was tear down my home office and turn it into a game room. I generally do not even use a computer in my home. Sometimes we have enforced WFH days due to various things or events, but I established the understanding with management that generally I will not be expected to work full days from home outside of emergencies and I take those days off.<p>I understand some people love it, and I have a good office environment which helps, but the downsides (general office annoyances, commuting, lunch expenses, clothing expenses, etc) way outweigh the downsides of working remote for me. I got severely depressed and burnt out very quickly.</p>
]]></description><pubDate>Wed, 19 Aug 2026 18:01:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=49364936</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49364936</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49364936</guid></item><item><title><![CDATA[New comment by JohnMakin in "Superpowers, Not Superintelligence"]]></title><description><![CDATA[
<p>paying other people to optimize their life so they don’t have to bother themselves with cooking for their children, yes, I fully believe that, and I believe that’s why this current world is so fucked, so we are not in disagreement at all, in a very odd way.</p>
]]></description><pubDate>Wed, 19 Aug 2026 01:27:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=49355399</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49355399</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49355399</guid></item><item><title><![CDATA[New comment by JohnMakin in "Superpowers, Not Superintelligence"]]></title><description><![CDATA[
<p>Many, many people have crappy relationships with their daughters. Put yourself in the pov of the daughter - “daddy thinks this isn’t worth his time but the ai says i’ll like it and i do” and the dad that wakes his kid up early to give them a special breakfast he spent time and energy on perfecting for the impossible child diet.<p>You think the former is outsourcable and preferable to the latter? I could make terrible judgments on you I won’t post because they’re cheap, or maybe you’re childless and young and have a cynical view of the world, but holy crap, in the context of this broader convo, and site, is a real “wtf” moment I can’t really do anything but digest in real time.</p>
]]></description><pubDate>Wed, 19 Aug 2026 01:26:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=49355386</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49355386</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49355386</guid></item><item><title><![CDATA[New comment by JohnMakin in "Superpowers, Not Superintelligence"]]></title><description><![CDATA[
<p>" It plans his daughter’s recipes and orders the ingredients."</p>
]]></description><pubDate>Tue, 18 Aug 2026 17:38:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=49349391</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49349391</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49349391</guid></item><item><title><![CDATA[New comment by JohnMakin in "Superpowers, Not Superintelligence"]]></title><description><![CDATA[
<p>A use case provided in this very article is "plans daughter's recipes and orders the ingredients." The whole act of doing that without automating it is what makes it meaningful.<p>What a miserable mindset.</p>
]]></description><pubDate>Tue, 18 Aug 2026 17:38:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=49349379</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49349379</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49349379</guid></item><item><title><![CDATA[New comment by JohnMakin in "Superpowers, Not Superintelligence"]]></title><description><![CDATA[
<p>You can tell the people pioneering this are utterly miserable and soulless. "Let's build something to automate all the parts of your life that are actually your life."<p>What person that was even mildly satisfied with living their life would want this? What's next, agents that play video games and watch tv for me? What's the point?</p>
]]></description><pubDate>Tue, 18 Aug 2026 16:59:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=49348709</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49348709</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49348709</guid></item><item><title><![CDATA[New comment by JohnMakin in "On AI regulation and messaging"]]></title><description><![CDATA[
<p>This is why this hype is so annoying - they <i>have</i> to make claims and goals like this to justify the current level of investment. It couldn't be like "ok, we have this cool tool that can sorta code and do computer stuff and automate things" and be a $50 billion dollar company, no, we have to shoot much, much higher, so high that I think the only inevitability is this all collapses in on itself in a financial black hole, which is a shame.</p>
]]></description><pubDate>Mon, 17 Aug 2026 19:44:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=49336547</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49336547</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49336547</guid></item><item><title><![CDATA[New comment by JohnMakin in "When Genius Fails: The Intellectual Arrogance of the AI Labs"]]></title><description><![CDATA[
<p>I would describe having billions under management and not understanding really basic kelly criterion as having wrong fundamentals, but I think you were talking about his beliefs in AI.<p>But thats sort of the whole problem, believing you know a lot about the future of AI, even if you’re correct, does not mean you have the expertise to manage a fund around that belief, and it’s absolutely shocking to me he was able to get that much money</p>
]]></description><pubDate>Fri, 14 Aug 2026 22:33:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=49305443</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49305443</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49305443</guid></item><item><title><![CDATA[New comment by JohnMakin in "Why does Opus 5 feel worse to work with?"]]></title><description><![CDATA[
<p>Confidently guessing, and building an entirely new set of assumptions based on a confident guess is the single source of the most frustration I experience using these. "Intelligence" does not look like that, and all the prompt hacking or clever hooks in the world doesn't seem to stop them from doing it.<p>I don't know there is really a solution. Some models get better at this for a while, then regress. It's like whack a mole. Until this is 'solved' these will never be fully human out of the loop, but I am suspecting this is a fundamental nature kind of thing with them.</p>
]]></description><pubDate>Fri, 14 Aug 2026 16:13:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=49300781</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49300781</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49300781</guid></item><item><title><![CDATA[New comment by JohnMakin in "Incident with Pull Requests and Issues – Resolved"]]></title><description><![CDATA[
<p>2-3 9's of availability on most services shown on their status page (and these are always generous). For a platform that wants to be <i>the</i> industry standard CI/CD platform, there is no way you can justify that.</p>
]]></description><pubDate>Wed, 12 Aug 2026 19:49:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=49277668</link><dc:creator>JohnMakin</dc:creator><comments>https://news.ycombinator.com/item?id=49277668</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49277668</guid></item></channel></rss>