<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Http on Joe Bylund</title><link>http://jbylund.github.io/tags/http/</link><description>Recent content in Http on Joe Bylund</description><generator>Hugo -- gohugo.io</generator><language>en</language><copyright>© 2026 Joe Bylund</copyright><lastBuildDate>Tue, 27 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="http://jbylund.github.io/tags/http/index.xml" rel="self" type="application/rss+xml"/><item><title>We Ignore Your Accept-Encoding q= Weights and Serve Better Compression Anyway</title><link>http://jbylund.github.io/posts/00688_http-compression-negotiation/</link><pubDate>Tue, 27 Oct 2026 00:00:00 +0000</pubDate><guid>http://jbylund.github.io/posts/00688_http-compression-negotiation/</guid><description>Serving compressed HTTP responses correctly is more nuanced than calling gzip.compress(). This post walks through the CompressionMiddleware in Sylvan Librarian: Accept-Encoding negotiation, server-side algorithm priority, two distinct code paths for buffered versus streaming responses, the 200-byte skip threshold, Vary headers, and gzip mtime=0 for byte-identical repeated responses.</description></item></channel></rss>