<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Claims on Programmer.ie: Modern AI programming</title>
    <link>http://programmer.ie/tags/claims/</link>
    <description>Recent content in Claims on Programmer.ie: Modern AI programming</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Mon, 14 Sep 2026 05:00:18 +0100</lastBuildDate>
    <atom:link href="http://programmer.ie/tags/claims/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Claims, Evidence, and Decisions</title>
      <link>http://programmer.ie/books/applied-ai/18-chapter/</link>
      <pubDate>Mon, 14 Sep 2026 05:00:18 +0100</pubDate>
      <guid>http://programmer.ie/books/applied-ai/18-chapter/</guid>
      <description>&lt;p&gt;&lt;em&gt;Part 3 — Give Intelligence a Runtime&lt;/em&gt;&lt;/p&gt;&#xA;&lt;h2 id=&#34;three-sentences-and-a-merge&#34;&gt;Three sentences and a merge&lt;/h2&gt;&#xA;&lt;p&gt;A review of a cache change says three things:&lt;/p&gt;&#xA;&lt;div class=&#34;highlight&#34;&gt;&lt;pre tabindex=&#34;0&#34; style=&#34;color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;&#34;&gt;&lt;code class=&#34;language-text&#34; data-lang=&#34;text&#34;&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;The cache TTL is set to 60 seconds in config/cache.toml. The retry test passes.&#xA;&lt;/span&gt;&lt;/span&gt;&lt;span style=&#34;display:flex;&#34;&gt;&lt;span&gt;p99 latency stays under 200 ms.&#xA;&lt;/span&gt;&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;p&gt;Someone reads it and merges the change. The next day, the deployed configuration turns out to say 600 seconds.&lt;/p&gt;&#xA;&lt;p&gt;The obvious question is whether the review was wrong. The more useful question is narrower: &lt;strong&gt;which part of the decision rested on which sentence, and what had anyone actually checked?&lt;/strong&gt; One sentence could be confirmed against a file. One could be tested. One had no evidence at all. The merge used two of them and knowingly left the third open, or it didn&amp;rsquo;t. The merge record lists no basis, so the basis cannot be reconstructed.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
