---
title: "EPUB is not that scary"
date: 2026-08-21
tags: post
---

In a prior life, I worked on Kindle — specifically on content pipelines and data processing. This either excuses the fact that I had no real idea what EPUB was from a data standpoint, or makes it all the more embarrassing, since Amazon has, quite notoriously, long waged a war against EPUB and the other open e-reader formats.

I will submit as my defense a certain kind of subconscious aversion to anything under that curfew. In much the same way that I have largely exited the world of fintech after Stripe — with the notable exception of fervently checking for updates about the growth and progress of FedNow.

All of this is a lead-up to say: I had reason at my day job, thanks to Steph, to try generating an EPUB file from Markdown source. An idea to which I said, "Yes, how hard can it be?"

Usually, when I say that, it is a prelude to disaster. But in this case, EPUB turns out to be pretty nice and simple. So I am going to drastically sand off the edges of some of the complexity, because I think the core bit is what's important:

**EPUB is a zip file that contains some metadata files and a bunch of XHTML. That is all.**

This was delightful as a discovery and disastrous as a derailment, since now I'm going to use that as license to spend some time — just like I did when I did my deep dive into ICS a while back [[ics]] — to daydream about some features that might be frivolous and unjustifiable, but nonetheless fun to take advantage of.