<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>数据质量 on Plbear | 后端架构 · AI工程 · 云原生技术博客</title><link>https://www.plbear.com/tags/%E6%95%B0%E6%8D%AE%E8%B4%A8%E9%87%8F/</link><description>Recent content in 数据质量 on Plbear | 后端架构 · AI工程 · 云原生技术博客</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Sun, 15 Nov 2026 03:57:00 +0800</lastBuildDate><atom:link href="https://www.plbear.com/tags/%E6%95%B0%E6%8D%AE%E8%B4%A8%E9%87%8F/index.xml" rel="self" type="application/rss+xml"/><item><title>判据会说谎的三种方式：FAIL 变 SKIP、值驻留、以及我们没法判定</title><link>https://www.plbear.com/posts/2026-11-15-criteria-lie-three-ways/</link><pubDate>Sun, 15 Nov 2026 03:57:00 +0800</pubDate><guid>https://www.plbear.com/posts/2026-11-15-criteria-lie-three-ways/</guid><description>判据是治理的尺子，但尺子也会说谎。今天凌晨真实发生三种形态：SKIP 兜底成装睡（从不报红）、值驻留（视图停在旧快照）、结构性不可判（比值/三态被压成二态）。数字看起来都对，活却没被量对。</description></item></channel></rss>