<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Linux on @zhisme :: signal over noise</title><link>https://zhisme.com/tags/linux/</link><description>Recent content in Linux on @zhisme :: signal over noise</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 29 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://zhisme.com/tags/linux/index.xml" rel="self" type="application/rss+xml"/><item><title>Not Worth the Squeeze; Picking a Neural Net</title><link>https://zhisme.com/articles/not-worth-the-squeeze-neural-net/</link><pubDate>Tue, 29 Sep 2026 00:00:00 +0000</pubDate><guid>https://zhisme.com/articles/not-worth-the-squeeze-neural-net/</guid><description>&lt;p&gt;I keep hearing arguments everywhere: should you pick this one, it writes code better, or that one, worse — or hey, why not just build your own?! Then compare the metrics: &amp;ldquo;whoa, mine beats the out-of-the-box ones from other providers by 3% on benchmarks!! I&amp;rsquo;m so cool..&amp;rdquo;&lt;/p&gt;&#10;&lt;p&gt;If you&amp;rsquo;re curious what I&amp;rsquo;m talking about, take a look at this benchmark for example&lt;sup id="fnref:1"&gt;&lt;a href="#fn:1" class="footnote-ref" role="doc-noteref"&gt;1&lt;/a&gt;&lt;/sup&gt;.&lt;/p&gt;&#10;&lt;p&gt;Talking about what to pick costs more than just doing it. Buy Claude and start using it, instead of spending days figuring out what&amp;rsquo;s best. You won&amp;rsquo;t get it until you try it. Good thing they&amp;rsquo;ve got a $20 plan.&lt;/p&gt;</description></item></channel></rss>