<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>obliv.log</title>
    <link>https://personal-blog-red-ten.vercel.app</link>
    <description>Notes on private information retrieval, lattice cryptography, and protocols that forget what you asked for.</description>
    <language>en</language>
    <atom:link href="https://personal-blog-red-ten.vercel.app/feed.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>Why every PIR scheme has to touch every record</title>
      <link>https://personal-blog-red-ten.vercel.app/posts/pir-must-touch-every-record</link>
      <guid isPermaLink="true">https://personal-blog-red-ten.vercel.app/posts/pir-must-touch-every-record</guid>
      <pubDate>Tue, 14 Jul 2026 00:00:00 GMT</pubDate>
      <description>A server that skips a row has already told you something. Working through the lower bound from first principles, and why it shapes every practical design.</description>
    </item>
    <item>
      <title>SimplePIR, and what a preprocessed hint really costs</title>
      <link>https://personal-blog-red-ten.vercel.app/posts/simplepir-hint-cost</link>
      <guid isPermaLink="true">https://personal-blog-red-ten.vercel.app/posts/simplepir-hint-cost</guid>
      <pubDate>Sun, 28 Jun 2026 00:00:00 GMT</pubDate>
      <description>Throughput looks excellent until you account for the client-side state. Measuring where the hint stops being worth it.</description>
    </item>
    <item>
      <title>Noise growth in RLWE is a budget you spend once</title>
      <link>https://personal-blog-red-ten.vercel.app/posts/rlwe-noise-budget</link>
      <guid isPermaLink="true">https://personal-blog-red-ten.vercel.app/posts/rlwe-noise-budget</guid>
      <pubDate>Tue, 09 Jun 2026 00:00:00 GMT</pubDate>
      <description>Every homomorphic operation draws down the same account. A practical way to reason about depth before you write any code.</description>
    </item>
  </channel>
</rss>