<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: aulin</title><link>https://news.ycombinator.com/user?id=aulin</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 13 Jun 2026 10:17:28 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=aulin" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by aulin in "Harness engineering: Leveraging Codex in an agent-first world"]]></title><description><![CDATA[
<p>Dear OpenAI, the target audience of your blog or at least of this blog post understands English pretty well. Why won't you give them a simple way to disable the shitty ai translation and read the original content? Why translate it at all in the first place?<p>EDIT: found the button, all the way down in the bottom of the page... I hate this so much, give me the original content, I will decide if and when I need translation</p>
]]></description><pubDate>Sun, 07 Jun 2026 06:42:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=48432472</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=48432472</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48432472</guid></item><item><title><![CDATA[New comment by aulin in "Various LLM Smells"]]></title><description><![CDATA[
<p>Leave my consecutive short sentences alone. And jetbrains mono.</p>
]]></description><pubDate>Fri, 29 May 2026 02:19:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=48318214</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=48318214</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48318214</guid></item><item><title><![CDATA[New comment by aulin in "How Claude Code works in large codebases"]]></title><description><![CDATA[
<p>If you can't clone it it's not a repo</p>
]]></description><pubDate>Fri, 15 May 2026 05:44:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=48144970</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=48144970</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48144970</guid></item><item><title><![CDATA[New comment by aulin in "I cancelled Claude: Token issues, declining quality, and poor support"]]></title><description><![CDATA[
<p>After you see it skip reasoning so many times and saying "actually the simplest fix is" the laziest thing ever you get kind of tired of babysitting it.</p>
]]></description><pubDate>Sat, 25 Apr 2026 00:03:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=47897301</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47897301</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47897301</guid></item><item><title><![CDATA[New comment by aulin in "Highlights from Git 2.54"]]></title><description><![CDATA[
<p>Not familiar with jj and don't want to get into bike shedding, but how is describe supposed to be a good name for history rewrites?</p>
]]></description><pubDate>Thu, 23 Apr 2026 11:52:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=47874642</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47874642</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47874642</guid></item><item><title><![CDATA[New comment by aulin in "Changes to GitHub Copilot individual plans"]]></title><description><![CDATA[
<p>So far they did not change it, and none of this applies to business and enterprise accounts. My idea is that it can still be viable as most businesses will have plenty minimally used licenses with just a few power users abusing the request model.</p>
]]></description><pubDate>Wed, 22 Apr 2026 06:02:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=47859650</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47859650</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47859650</guid></item><item><title><![CDATA[New comment by aulin in "ChatGPT Images 2.0"]]></title><description><![CDATA[
<p>Well that was probably the most important general skill even before this.</p>
]]></description><pubDate>Wed, 22 Apr 2026 05:29:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=47859445</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47859445</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47859445</guid></item><item><title><![CDATA[New comment by aulin in "Measuring Claude 4.7's tokenizer costs"]]></title><description><![CDATA[
<p>Opus 4.6 also just got dumber. It's dismissive, hand-wavy, jumps to conclusions way too quickly, skips reasoning... Bubble is going to burst, either some big breakthrough comes up or we are going to see a very fast enshittificafion.</p>
]]></description><pubDate>Fri, 17 Apr 2026 21:09:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=47810589</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47810589</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47810589</guid></item><item><title><![CDATA[New comment by aulin in "Measuring Claude 4.7's tokenizer costs"]]></title><description><![CDATA[
<p>They won't. These are not "issues", it's them trying to push the models to burn less compute. It will only get worse.</p>
]]></description><pubDate>Fri, 17 Apr 2026 21:01:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=47810520</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47810520</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47810520</guid></item><item><title><![CDATA[New comment by aulin in "I still prefer MCP over skills"]]></title><description><![CDATA[
<p>What did poor copilot do now?</p>
]]></description><pubDate>Sat, 11 Apr 2026 01:14:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=47726175</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47726175</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47726175</guid></item><item><title><![CDATA[New comment by aulin in "Codex pricing to align with API token usage, instead of per-message"]]></title><description><![CDATA[
<p>> I need to try the command line version.<p>Is there any other?</p>
]]></description><pubDate>Sun, 05 Apr 2026 23:31:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=47655048</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47655048</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47655048</guid></item><item><title><![CDATA[New comment by aulin in "Codex pricing to align with API token usage, instead of per-message"]]></title><description><![CDATA[
<p>GH Copilot is still the best deal, while it lasts</p>
]]></description><pubDate>Sun, 05 Apr 2026 18:38:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=47652501</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=47652501</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47652501</guid></item><item><title><![CDATA[New comment by aulin in "GPT-5.3-Codex"]]></title><description><![CDATA[
<p>Admit I didn't follow the announcements but isn't that a matter of UI? Doesn't seem something that should be baked in the model but in the tooling around it and the instructions you give them. E.g. I've been playing with with GitHub copilot CLI (that despite the bad fame is absolutely amazing) and the same model completely changes its behavior with the prompt. You can have it answer a question promptly or send it on a multi-hour multi-agent exploration writing detailed specs with a single prompt. Or you can have it stop midway for clarification. It all depends on the instructions. Also this is particularly interesting with GitHub billing model as each prompt counts 1 request no matter how many tokens it burns.</p>
]]></description><pubDate>Fri, 06 Feb 2026 05:29:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=46909467</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=46909467</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46909467</guid></item><item><title><![CDATA[New comment by aulin in "Writing non-English languages with a QWERTY keyboard"]]></title><description><![CDATA[
<p>If you are on a us ansi keyboard and switch to a iso layout (most European layouts are iso) you have I believe two unreachable keys. And the arrangement of the others is slightly different you will have to adapt your muscle memory anyway.<p>Altgr-intl is pretty good for when you code and write English most of the time and occasionally need accented letters. If you need to write a lot in your native language it's better to get a local layout keyboard.</p>
]]></description><pubDate>Mon, 29 Dec 2025 07:36:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=46418314</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=46418314</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46418314</guid></item><item><title><![CDATA[New comment by aulin in "When a driver challenges the kernel's assumptions"]]></title><description><![CDATA[
<p>Seems to me the most expectations they had with the library was about the compression stuff and it did not include that. So in the end it was mostly rev eng. Also in this specific case you are using the library code as documentation about the hardware, the code itself has little value. I doubt it would configure as license violation.</p>
]]></description><pubDate>Fri, 26 Dec 2025 10:30:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=46390882</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=46390882</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46390882</guid></item><item><title><![CDATA[New comment by aulin in "Why is the world losing color?"]]></title><description><![CDATA[
<p>They don't get as hot when parked under the sun though.</p>
]]></description><pubDate>Thu, 03 Apr 2025 04:45:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=43564846</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=43564846</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43564846</guid></item><item><title><![CDATA[New comment by aulin in "The role of developer skills in agentic coding"]]></title><description><![CDATA[
<p>I am not wrong. You simply are not the kind of developer I am thinking about. And believe me the other kind is way more represented.</p>
]]></description><pubDate>Thu, 27 Mar 2025 12:34:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=43492931</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=43492931</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43492931</guid></item><item><title><![CDATA[New comment by aulin in "The role of developer skills in agentic coding"]]></title><description><![CDATA[
<p>In my every day experience that's pretty risky. The periphery as you call it is often an area where you lack the expertise to spot and correct AI mistakes.<p>I am thinking about build systems and shell scripts. I see people everyday going to AI before even looking at the docs and invariably failing with non-existent command line options, or worst options that break things in very subtle ways.<p>Same people that when you tell them why don't you read the f-ing man page they go to google to look it up instead of opening a terminal.<p>Same people that push through an unknown problem by trial and error instead of reading the docs first. But now they have this dumb counselor that steers them in the wrong direction most of the time and the whole process is even more error prone.</p>
]]></description><pubDate>Thu, 27 Mar 2025 05:43:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=43490665</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=43490665</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43490665</guid></item><item><title><![CDATA[New comment by aulin in "Pi-hole v6"]]></title><description><![CDATA[
<p>I use pihole for dhcp and it's extremely easy with dnsmasq. Hope their settings overhaul does not break this.<p>dhcp-option=tag:nospam,option:dns-server,x.x.x.x
dhcp-option=tag:spam,option:dns-server,y.y.y.y
dhcp-host=client1...,set:nospam
dhcp-host=client2...,set:spam</p>
]]></description><pubDate>Wed, 19 Feb 2025 06:49:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=43099323</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=43099323</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43099323</guid></item><item><title><![CDATA[New comment by aulin in "You're not a senior engineer until you've worked on a legacy project (2023)"]]></title><description><![CDATA[
<p>Self-esteem that easily turns into hubris though. I think the real seniority shows when you are able to work on a legacy codebase full of the shittiest code and not have the slightest desire to rewrite it all.</p>
]]></description><pubDate>Tue, 18 Feb 2025 05:31:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=43086388</link><dc:creator>aulin</dc:creator><comments>https://news.ycombinator.com/item?id=43086388</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43086388</guid></item></channel></rss>