<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0">
<channel>
  <title>The Evening Shell</title>
  <link>https://harborarchive.198-13-63-140.nip.io/</link>
  <description>A software engineering notebook, based in Galway.</description>
  <language>en-us</language>
  <item><title>Why Feature Flags Still Matters</title><link>https://harborarchive.198-13-63-140.nip.io/writing/why-feature-flags-still-matters/</link><guid>https://harborarchive.198-13-63-140.nip.io/writing/why-feature-flags-still-matters/</guid><pubDate>Fri, 30 Oct 2026 23:48:00 GMT</pubDate><description>Somewhere between a legacy codebase and home, the thought finally made sense. The trick, if there is one, is to leave sufficient room for missteps.</description></item>
  <item><title>Debugging Test Suites</title><link>https://harborarchive.198-13-63-140.nip.io/writing/debugging-test-suites/</link><guid>https://harborarchive.198-13-63-140.nip.io/writing/debugging-test-suites/</guid><pubDate>Sun, 28 Jun 2026 21:36:00 GMT</pubDate><description>I tried the pragmatic version first. It did not last.</description></item>
  <item><title>A Readable Approach to Small Services</title><link>https://harborarchive.198-13-63-140.nip.io/writing/a-readable-approach-to-small-services/</link><guid>https://harborarchive.198-13-63-140.nip.io/writing/a-readable-approach-to-small-services/</guid><pubDate>Wed, 11 Mar 2026 03:06:00 GMT</pubDate><description>It is simple to overthink refactoring. It is harder to just start. I wrote most of this on a calm Thursday morning.</description></item>
  <item><title>Revisiting Feature Flags</title><link>https://harborarchive.198-13-63-140.nip.io/writing/revisiting-feature-flags/</link><guid>https://harborarchive.198-13-63-140.nip.io/writing/revisiting-feature-flags/</guid><pubDate>Wed, 24 Dec 2025 08:36:00 GMT</pubDate><description>Logs are only useful if someone can read them at three in the morning.</description></item>
</channel>
</rss>
