<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: refsys</title><link>https://news.ycombinator.com/user?id=refsys</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 09 Apr 2026 08:18:33 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=refsys" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by refsys in "I'm helping my dog vibe code games"]]></title><description><![CDATA[
<p>Cute, but clearly Godot is doing all of the heavy lifting here.<p>"It's possible to make shitty but playable games by running random scripts through a >2MLoC game engine and iterating on errors" is interesting but not nearly as sensationalist.</p>
]]></description><pubDate>Wed, 25 Feb 2026 10:11:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=47149671</link><dc:creator>refsys</dc:creator><comments>https://news.ycombinator.com/item?id=47149671</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47149671</guid></item><item><title><![CDATA[New comment by refsys in "Ladybird adopts Rust, with help from AI"]]></title><description><![CDATA[
<p>Any terminal tool like Claude Code or Codex (I assume OpenCode too, but I haven't tried) can do it, by using as a prompt pretty much exactly what you wrote, and if it still wants to edit, just don't approve the tool calls.<p>One problem I've noticed is that both claude models and gpt-codex variants make absolutely deranged tool calls (like `cat <<'EOF' >> foo...EOF` pattern to create a file, or sed to read a couple lines), so it's sometimes hard to see what is it even trying to do.</p>
]]></description><pubDate>Mon, 23 Feb 2026 20:14:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=47128104</link><dc:creator>refsys</dc:creator><comments>https://news.ycombinator.com/item?id=47128104</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47128104</guid></item><item><title><![CDATA[New comment by refsys in "Pope tells priests to use their brains, not AI, to write homilies"]]></title><description><![CDATA[
<p>Entirely random of course. I would never reference unsavory memes about past Popes or anything like that.</p>
]]></description><pubDate>Mon, 23 Feb 2026 11:00:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=47120679</link><dc:creator>refsys</dc:creator><comments>https://news.ycombinator.com/item?id=47120679</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47120679</guid></item><item><title><![CDATA[New comment by refsys in "Google restricting Google AI Pro/Ultra subscribers for using OpenClaw"]]></title><description><![CDATA[
<p>Where does Anthropic offer an "unlimited" subscription? All of the plans mentioned on <a href="https://claude.com/pricing" rel="nofollow">https://claude.com/pricing</a> have limits, same as usage of Codex on OpenAI's ChatGPT subscription plans. If Google forgot to actually enforce a rate limit (that they do mention on <a href="https://antigravity.google/pricing" rel="nofollow">https://antigravity.google/pricing</a>) on theirs, that sounds like a huge oversight.</p>
]]></description><pubDate>Mon, 23 Feb 2026 10:48:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=47120594</link><dc:creator>refsys</dc:creator><comments>https://news.ycombinator.com/item?id=47120594</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47120594</guid></item><item><title><![CDATA[New comment by refsys in "Pope tells priests to use their brains, not AI, to write homilies"]]></title><description><![CDATA[
<p>"We value your privacy! Do you consent to sharing the contents of your confession with our 2137 partners? [ACCEPT ALL] [MAYBE LATER]"</p>
]]></description><pubDate>Mon, 23 Feb 2026 10:44:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=47120563</link><dc:creator>refsys</dc:creator><comments>https://news.ycombinator.com/item?id=47120563</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47120563</guid></item><item><title><![CDATA[New comment by refsys in "NanoClaw moved from Apple Containers to Docker"]]></title><description><![CDATA[
<p>or you pass by graffiti telling it to jump off a cliff, written in iambic pentameter (or whatever is the jailbreak meta of the month)</p>
]]></description><pubDate>Sun, 22 Feb 2026 20:59:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=47114609</link><dc:creator>refsys</dc:creator><comments>https://news.ycombinator.com/item?id=47114609</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47114609</guid></item><item><title><![CDATA[New comment by refsys in "An Unbothered Jimmy Wales Calls Grokipedia a 'Cartoon Imitation' of Wikipedia"]]></title><description><![CDATA[
<p>They also have a report form for slop sites, but none of mine got reviewed yet (I have 5 reports since November, and the help still says "We will start processing reports officially in January.")</p>
]]></description><pubDate>Sun, 22 Feb 2026 20:42:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=47114456</link><dc:creator>refsys</dc:creator><comments>https://news.ycombinator.com/item?id=47114456</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47114456</guid></item><item><title><![CDATA[New comment by refsys in "What Every Experimenter Must Know About Randomization"]]></title><description><![CDATA[
<p>Maybe people have bad memories from linear congruential generators, these could go really bad (<a href="https://en.wikipedia.org/wiki/Marsaglia%27s_theorem" rel="nofollow">https://en.wikipedia.org/wiki/Marsaglia%27s_theorem</a>)</p>
]]></description><pubDate>Wed, 18 Feb 2026 22:11:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=47067143</link><dc:creator>refsys</dc:creator><comments>https://news.ycombinator.com/item?id=47067143</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47067143</guid></item></channel></rss>