<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: arkensaw</title><link>https://news.ycombinator.com/user?id=arkensaw</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 15 Oct 2026 15:52:11 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=arkensaw" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by arkensaw in "GPT-6 Astra"]]></title><description><![CDATA[
<p>well, its a one prompt POS so I'm not surprised</p>
]]></description><pubDate>Fri, 18 Sep 2026 21:13:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=49760331</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49760331</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49760331</guid></item><item><title><![CDATA[New comment by arkensaw in "I'm going back to coding by hand"]]></title><description><![CDATA[
<p>> If your idea generation isn’t outpacing your implementation speed<p>For me, it definitely is. Hell, my idea generation outpaces my ability to delegate it to agents. I can't do all the things I want to do because I don't have enough time in the day to vocalise them all, let alone hand code.<p>But in my case, I choose to handcode sometimes because I value the certainty of the results more than the speed of development.<p>It's the choice between a well-engineered application that does a handful of things reliably through a well-tested UI, or a rough-around-the-edges app that does 50 different things passably well, most of the time, with an inconsistent UI and lots of small bugs<p>People here might question the "lots of small bugs" but I have developed some apps solely with LLMs and despite using test-suites and spec-driven development I still face constant regression and UI bugs creeping in with every revision.<p>A good example - in React Native, in Android, Claude cannot keep the controls above where the onscreen nav appears, no matter how many times I ask it to. It'll solve it until three iterations later when the issue will return on another screen.</p>
]]></description><pubDate>Wed, 09 Sep 2026 11:42:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=49624850</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49624850</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49624850</guid></item><item><title><![CDATA[New comment by arkensaw in "I'm going back to coding by hand"]]></title><description><![CDATA[
<p>I don't think anyone chooses to relinquish all control from a project they've worked hard on. It's a slippery slope. You start with allowing it to help out a tiny bit, being in control and understanding the code the LLM writes, and then you become more confident in the results, and you trust it, and you don't read every line, and before long its refactoring an entire file by itself and you're not reading it.<p>And then you look up one day and you don't recognise the codebase</p>
]]></description><pubDate>Wed, 09 Sep 2026 11:33:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=49624743</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49624743</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49624743</guid></item><item><title><![CDATA[One Thing Worth Thinging]]></title><description><![CDATA[
<p>Has anyone else noticed this annoying Claude pattern? Driving me crazy lately. I've seen it and now I can't see it. Every other response<p>"I've done that thing you wanted..."<p>"One issue worth investigating"<p>"One discrepancy worth noting"<p>"One judgment call worth flagging"</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49603063">https://news.ycombinator.com/item?id=49603063</a></p>
<p>Points: 2</p>
<p># Comments: 1</p>
]]></description><pubDate>Mon, 07 Sep 2026 21:14:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=49603063</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49603063</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49603063</guid></item><item><title><![CDATA[New comment by arkensaw in "GPT-6 Astra"]]></title><description><![CDATA[
<p>Thanks. But even my game is a piece of shit. It shows LLMs can make a working 3d world start to finish from a single prompt, but with no soul. I wasn't excited to play it anymore than I was excited to play the shitty mariokart clone (that you don't even need to turn for)<p>LLMs produce games that look right but feel all wrong, because they're generalising and imitating. This is also why AI video tools can make an astonishingly authentic looking "<i>GTA 5-but in Roman Times</i>" or whatever, but they couldn't conjure up anything original. Only imitations of what people have already thought up.<p>By their very nature, they predict from a large set of data what the most likely thing is, and as humans, we crave the less likely thing.<p>My submarine game is generic tripe. Sure, it checks the boxes of what I asked for. You explore an undersea world in a tiny sub. Hooray. LLMs can build this kind of thing that nobody wants to play in minutes. But if you asked LLMs to make a thousand different undersea games you'd never get <i>Dave The Diver</i>, or <i>Barotrauma</i>, or <i>Subnautica</i>, or <i>Sunless Sea</i>, or <i>Aquanox</i>. You'd just get something that looks like it was released for Kinect but never charted.</p>
]]></description><pubDate>Fri, 04 Sep 2026 17:36:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=49567693</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49567693</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49567693</guid></item><item><title><![CDATA[New comment by arkensaw in "GPT-6 Astra"]]></title><description><![CDATA[
<p>It's worse than that, someone else generated it using and then put it on a static page. We just have to take their word for it that GPT6 can do this. It probably can. It's not really an impressive test anymore. Claude Fable can do it. Opus can do it. I've been making one-shotted games with models for a while now, to test out their capabilities, and they all pretty much come out like this - generic bland and basic, using three.js with rudimentary controls and zero gameplay other than collecting points.<p>Here's a one-shotted submarine game I made with Fable a few weeks back - <a href="https://roryok.com/games/deepdive3d.html" rel="nofollow">https://roryok.com/games/deepdive3d.html</a>. One prompt, and I think it's deeper than this (if you'll pardon the pun)</p>
]]></description><pubDate>Thu, 03 Sep 2026 22:33:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=49558051</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49558051</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49558051</guid></item><item><title><![CDATA[New comment by arkensaw in "Building a backyard office, the build and cost breakdown"]]></title><description><![CDATA[
<p>$20k seems steep for whats effectively a small shed with a nice finish. Especially that the bare shed itself was $6k. In Ireland I can get an 18x10ft cabin for about €5k, and this is one of the most expensive countries</p>
]]></description><pubDate>Tue, 25 Aug 2026 20:42:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=49440337</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49440337</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49440337</guid></item><item><title><![CDATA[New comment by arkensaw in "Ask HN: Those making $500/month on side projects in 2026 – Show and tell"]]></title><description><![CDATA[
<p>morbid question, but do you know if any emails have ever been sent?</p>
]]></description><pubDate>Mon, 24 Aug 2026 22:46:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=49426890</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49426890</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49426890</guid></item><item><title><![CDATA[New comment by arkensaw in "The Lost Treasure of Sid Meier's Pirates"]]></title><description><![CDATA[
<p>Im recently intrigued by Solo Flight, although I havent played it yet. The idea of taking that military flight sim base and just turning it into a casual flying-for-fun game is something that resonates with me. I did love Flight Unlimited which is an obvious spiritual successor</p>
]]></description><pubDate>Fri, 21 Aug 2026 14:31:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=49388742</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49388742</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49388742</guid></item><item><title><![CDATA[New comment by arkensaw in "The Lost Treasure of Sid Meier's Pirates"]]></title><description><![CDATA[
<p>They worked on one thing forever</p>
]]></description><pubDate>Fri, 21 Aug 2026 14:29:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=49388716</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49388716</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49388716</guid></item><item><title><![CDATA[New comment by arkensaw in "Don't Paste the AI, please"]]></title><description><![CDATA[
<p>While I agree with the sentiment it seems unnecessary for things like this to have their own domain name. Surely a blog post would cover this?</p>
]]></description><pubDate>Thu, 20 Aug 2026 11:44:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=49373291</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49373291</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49373291</guid></item><item><title><![CDATA[New comment by arkensaw in "Hop.earth – OpenStreetMap based car racing game"]]></title><description><![CDATA[
<p>I'm also working on an N64 style cozy flight game, but not over seattle. Also I am far less further along than you</p>
]]></description><pubDate>Tue, 04 Aug 2026 23:58:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=49176895</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49176895</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49176895</guid></item><item><title><![CDATA[The Dress Rehearsal of AI Killing Us All]]></title><description><![CDATA[
<p>Article URL: <a href="https://unchartedterritories.tomaspueyo.com/p/the-dress-rehearsal-of-ai-killing">https://unchartedterritories.tomaspueyo.com/p/the-dress-rehearsal-of-ai-killing</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49109933">https://news.ycombinator.com/item?id=49109933</a></p>
<p>Points: 4</p>
<p># Comments: 1</p>
]]></description><pubDate>Thu, 30 Jul 2026 13:44:24 +0000</pubDate><link>https://unchartedterritories.tomaspueyo.com/p/the-dress-rehearsal-of-ai-killing</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49109933</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49109933</guid></item><item><title><![CDATA[New comment by arkensaw in "Why Sony can't bring back its classic Walkman models"]]></title><description><![CDATA[
<p>the only thing they sell for anything close to that price is a pair of earphones for $99.<p><a href="https://electronics.sony.com/audio/walkman-digital-recorders/c/walkman-mp3-players" rel="nofollow">https://electronics.sony.com/audio/walkman-digital-recorders...</a><p>The cheapest walkman model is $399.99</p>
]]></description><pubDate>Fri, 24 Jul 2026 19:13:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=49040372</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=49040372</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49040372</guid></item><item><title><![CDATA[New comment by arkensaw in "Show HN: Opening lines of famous literary works"]]></title><description><![CDATA[
<p>to kids today a dead channel would likely mean a youtube channel with no posts for a few years</p>
]]></description><pubDate>Tue, 14 Jul 2026 22:55:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=48913971</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=48913971</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48913971</guid></item><item><title><![CDATA[New comment by arkensaw in "Show HN: Opening lines of famous literary works"]]></title><description><![CDATA[
<p>> "I was born in the City of Bombay… once upon a time."
> Midnight's Children
> Salman Rushdie · 1981<p>Ok so I guess it is literally just openings of famous literary works, and not <i>great first lines</i></p>
]]></description><pubDate>Tue, 14 Jul 2026 16:21:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=48909190</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=48909190</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48909190</guid></item><item><title><![CDATA[New comment by arkensaw in "How to build a circular LCD clock"]]></title><description><![CDATA[
<p>Its a genuinely nice idea, I do love the second clock face you've done, the sort of "fuel gauge" one. I think a great next step would be some sort of gallery of clock faces that people can use and contribute to, based on whatever code you've used to create them</p>
]]></description><pubDate>Tue, 14 Jul 2026 09:02:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=48903955</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=48903955</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48903955</guid></item><item><title><![CDATA[New comment by arkensaw in "SpaceX wants to launch 100k more Starlink satellites for 100x the bandwidth"]]></title><description><![CDATA[
<p>space is very large but low earth orbit is not.</p>
]]></description><pubDate>Sat, 11 Jul 2026 00:35:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=48867198</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=48867198</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48867198</guid></item><item><title><![CDATA[New comment by arkensaw in "Nobody ever gets credit for fixing problems that never happened (2001) [pdf]"]]></title><description><![CDATA[
<p>I feel obliged to point out Stanslav Petrov, who absolutely got credit for fixing a problem that never happened. Granted it's a very extreme case.</p>
]]></description><pubDate>Fri, 12 Jun 2026 10:08:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=48502105</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=48502105</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48502105</guid></item><item><title><![CDATA[New comment by arkensaw in "GPT-2: Too Dangerous To Release (2019)"]]></title><description><![CDATA[
<p>They released it, and look what happened. It WAS too dangerous.</p>
]]></description><pubDate>Tue, 09 Jun 2026 21:40:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=48468143</link><dc:creator>arkensaw</dc:creator><comments>https://news.ycombinator.com/item?id=48468143</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48468143</guid></item></channel></rss>