<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Elixir on jed.codes</title><link>https://jed.codes/tags/elixir/</link><description>Recent content in Elixir on jed.codes</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 01 Jul 2026 21:41:52 -0700</lastBuildDate><atom:link href="https://jed.codes/tags/elixir/index.xml" rel="self" type="application/rss+xml"/><item><title>TDD is the lab, telemetry is the field</title><link>https://jed.codes/posts/betzen-telemetry-driven-development/</link><pubDate>Wed, 01 Jul 2026 21:41:52 -0700</pubDate><guid>https://jed.codes/posts/betzen-telemetry-driven-development/</guid><description>&lt;p&gt;&lt;img src="https://jed.codes/images/telemetry-driven-development.png" alt="Split illustration: on the left, an engineer in a lab coat working at a workbench with red/green/refactor monitors labeled &amp;ldquo;MODEL: TDD — Predefined Behavior&amp;rdquo;; on the right, a field-scientist figure walking through a landscape of telemetry span waterfalls and flame graphs, alongside a placard reading &amp;ldquo;The purpose of a system is what it does — Stafford Beer&amp;rdquo; and &amp;ldquo;FIELD SCIENCE: TeDD — Actual Behavior&amp;rdquo;."&gt;&lt;/p&gt;
&lt;p&gt;Noah Betzen&amp;rsquo;s &lt;a href="https://www.youtube.com/watch?v=irQicdafnyM"&gt;&amp;ldquo;Telemetry Driven Development&amp;rdquo;&lt;/a&gt; talk from ElixirConf gave me a reason to write down a conviction I&amp;rsquo;ve been carrying for a while: for most of us, Test Driven Development (TDD) is a substitute for understanding how the system behaves. A more honest description is that it &lt;em&gt;models&lt;/em&gt; how the system behaves against a set of predefined behaviors and expectations. &lt;strong&gt;TDD is a model of behavior — not the behavior of the system at large.&lt;/strong&gt;&lt;/p&gt;</description></item></channel></rss>