My current thinking depends on something in the standard.site document spec. Implementers are supposed to include a well-known tag on their documents that exposes the AT URI for the actual record behind the page.

If every standard.site document reliably includes its AT URI in a predictable way, then it gives us a stable identifier to work with.

Detecting standard.site documents

In a highlighting service like Margin (the one I use), we could detect when the page we’re on is a standard.site document. If we can find that well-known tag and extract the AT URI, then we could embed that URI directly into the records for the highlights and annotations we create. So instead of a highlight just pointing to a URL, it would also carry the canonical AT URI for the document’s underlying record.

It would at least serve as a useful data point for anyone who wants to integrate annotation data more deeply into their site. Even if nothing else happens, that metadata would be there.

What this might enable

Where this starts to feel interesting is when I think about downstream uses.

For example, something like Leaflet might be able to include Margin annotations as comments, or something adjacent to comments. If highlights are tied to the document’s AT URI, then it becomes possible to associate them more directly with the record itself instead of just the page URL.

Maybe bookmarks could show up as a lightweight “people liked this” signal. Leaflet already has recommendations, so this wouldn’t be conceptually out of place. I’m not totally sure what the right shape of that would be, but it feels like there’s something there.

I don’t know. It seems interesting, at least worth exploring.