<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><atom:link href="https://rhostone.co.uk/feed.xml" rel="self" type="application/rss+xml"/><title>RhoStone — Lab book</title><language>en-gb</language><link>https://rhostone.co.uk/lab-book/</link><description>RhoStone’s lab book: entries on models, contracts, data and the decisions that depend on them.</description><item><title>Deciding what a model can ignore</title><link>https://rhostone.co.uk/lab-book/complex-systems.html</link><guid>https://rhostone.co.uk/lab-book/complex-systems.html</guid><description>Every model leaves most of the world out. The work is choosing what to drop, and noticing when something you dropped starts driving the result.</description><pubDate>Mon, 28 Sep 2026 00:00:00 GMT</pubDate></item><item><title>What a notice clause looks like to an engineer</title><link>https://rhostone.co.uk/lab-book/between-disciplines.html</link><guid>https://rhostone.co.uk/lab-book/between-disciplines.html</guid><description>Moving a clause from contract language into a state machine, then into a financial model, exposes gaps each vocabulary hides. It also loses things.</description><pubDate>Wed, 30 Sep 2026 00:00:00 GMT</pubDate></item><item><title>What happens after the score</title><link>https://rhostone.co.uk/lab-book/models-to-decisions.html</link><guid>https://rhostone.co.uk/lab-book/models-to-decisions.html</guid><description>Most of the effort in a risk model goes into the prediction. Most of the trouble turns up between the number and the person who has to act on it.</description><pubDate>Thu, 01 Oct 2026 00:00:00 GMT</pubDate></item></channel></rss>
