Colophon

How it's built

It's a static site: every page is plain HTML, built ahead of time. Fonts, images, and text are all served from here. There are no analytics, no ad networks, no cookies, and no consent banner to dismiss.

I avoid outside code and outside services wherever I can — not as a slogan, but because every piece of someone else's code is something I can't vouch for, and I'd rather know exactly what runs on your machine. I'm not going to promise you a number that might stop being true next month, though. What I'll promise instead is this: when there is an exception, I'll name it, say why it's there, and tell you what it means for you.

The exceptions, as of right now:

• Music streams from a storage bucket I rent, because audio files are far too big to sit alongside the pages. It's my own bucket rather than a third-party service — nothing there watches you, profiles you, or reports back — but it is a different address from the rest of the site. Nothing is fetched from it unless you press play.
• About twenty lines of JavaScript on the music pages. It stops two tracks playing over each other, and it powers the "Play all" button so an album can run start to finish. It's my own code, it talks to nothing and stores nothing, and if you block scripts the players still work exactly as they always did — you just click each track yourself. Every other page on this site runs no JavaScript at all.

Accessibility is pass/fail, not best-effort. Every change is checked against WCAG 2.2 AA automatically before it can go live, and then I test it by ear with my screen reader — I'm blind, so if it doesn't work for me, it doesn't ship. There's more on that in the accessibility statement.

I keep it lean on purpose — as few dependencies as possible, and self-reliant wherever I can be, so I always know exactly what's running and that it stays accessible. The few tools I do lean on get credited honestly: the site is built withAstro, and set inAtkinson Hyperlegible, a typeface designed for low-vision readers.

Comments and the contact form are built to respect you. Comments are read and approved by hand before they ever appear. Messages through thecontact page come straight to me, privately — nothing you send there is ever made public. No accounts, no logins, nothing to sign up for; and if you'd like to follow along, there's an RSS feed and you don't need one.

A little honesty about how it's made, too. The vision, the judgment calls, the look and feel, and the words are mine. The code is written by Claude, an AI assistant I work with: I make the decisions, Claude does the building. I mark where AI has had a hand, here as everywhere on this site, because I'd rather you know than wonder.

Your data

Short version: I don't want it. Nothing here is ever used for tracking, advertising, sales, or profiling — there is nothing collecting it in the first place. The only thing I ever ask for is an email address when you leave a comment, and even that is entirely optional. I use it only to reply to you; it's never shown publicly and never shared. Keeping the site in my own code, with as few outside dependencies as I can manage, is exactly how I keep the whole thing clean and trackerless for everyone.

Content warnings appear on anything heavy, so you can read prepared. And if you ever have an issue, a question, or feedback of any kind, pleasereach out through the contact page — I'd genuinely like to hear from you.

Changelog

I started versioning the site properly on 22 July 2026. From here on, notable site and feature changes are logged here; new chapters and blog posts don't move the version — this is about the site itself.