Legal

This page collects the terms the garden runs on: what you can do with the writing and the code, how crawlers are asked to behave, where the images come from and what to do if something here is yours. I am not a lawyer, and this page is my plain-language reading. Where it and a licence's own text disagree, the licence wins.

The writing

The writing on this site (posts, reflections, references, artifact notes) is licensed CC BY-SA 4.0. You can read it, quote it, share it, adapt it and train on it, commercially or not. Two conditions:

• Attribution. Credit Dan Peterson, link the page you used, and name the licence.
• ShareAlike. If you adapt the writing, your adaptation carries the same licence.

ShareAlike is the trade I am making on purpose: anything built from this stays as open as what it was built from.

Before 2026-10-02 the writing was licensed CC BY-NC-SA 4.0. A Creative Commons grant cannot be taken back, so anything published before that date is available under either licence.

The licence covers my words. It does not cover the quotes, images or other material I use from other people (see below); those stay under their owners' terms.

The code

The code that builds the site is licensed MIT.

Crawlers and AI training

robots.txt carries a Content Signal: search=yes, ai-input=yes, ai-train=yes. In plain terms: index it and link to it, use it to answer questions, and train on it.

A signal is a stated preference, not a contract, and nothing enforces it. The licence is the term; the signal says the same thing to a crawler. I opened training because I think a model trained on public writing is now one of the ways knowledge reaches people. However, that only works with attribution, which is the part a training run is worst at. I have not solved that, and the licence does not solve it either.

Quoting other people

I quote other people's work in short excerpts, for commentary, criticism and review, with a link to the original (in Canada this is fair dealing). I do not republish transcripts, whole articles or long passages. If a quote of yours runs longer than you are comfortable with, write to me (see Takedowns and corrections).

Images

Every image carries a caption naming who made it. Images come from three places:

• Pexels, under the Pexels licence.
• My own photos and screenshots.
• Openly licensed or public-domain works (e.g. Wikimedia Commons, NASA, museum open-access collections). The caption names the licence and links it.

No image on the site is AI-generated. Images are not covered by the writing's licence: each keeps its own terms, as its caption says.

Photos of real people are used for commentary only. A face in a header does not mean that person has read, approved or endorsed the post.

Takedowns and corrections

If something here is yours (a photo, a quote, a passage) and you want it credited differently or taken down, write to support@latentmirror.com. Include the page, the image or passage, and what you would like done. I will take an image down while I look into it.

The same address is for anything I got wrong. Corrections are made in the open, with a note saying what changed.

Privacy

There are no accounts, no comments and no ads. Page views are counted with Cloudflare Web Analytics, which sets no cookies and stores nothing in your browser. The site has no other tracking.

When a post links to another site, that site may be sent a webmention (a standard notice that it was linked). Webmentions sent to this site are received by webmention.io. Nothing about you is shown on the site.

Email to support@latentmirror.com is forwarded to me. I keep it only to reply to it.

In short

This is not legal advice, mine or anyone else's. If you need to rely on something here, read the licence itself.