Annotations Integration: Deep Linking, Saving, & Sharable Book Clippings

[mekarpeles]

Context

Is your feature request related to a problem? What is the motivation for implementing this feature?

Deep Linking & Sharing Book Clippings: This [see bluesky post attached] is what Books Team is planning to make easy -- sharing of clippings.

This is a product extension of #1214. See also "Snippets + Annotations + Hypothes.is"

People are currently doing it manually. We have it on our 2026 roadmap

A few months back @brewsterkahle asked:

can we/ should we support the highlighting feature I see in the browser: https://electrek.co/2025/10/07/tesla-releases-new-more-affordable-model-y-3-2k-more-than-last-week/#:~:text=Tesla%20Model%20Y-,Tesla%20releases%20'more%20affordable'%20Model%203/Y%20that%20costs,CEO%20Elon%20Musk%20pushed%20for.

On Archive.org, there's already an affordance to save "Web Archives".

Proposal & Constraints

What is the proposed solution / implementation? Is there a precedent for this approach succeeding elsewhere?

This feature proposes we also be able to save clipping regions (or snippets or whatever we want to call them) -- a type of annotation. This is much the same type of service that https://github.com/hypothesis provides.

Which suggestions or requirements should be considered for how this feature needs to appear or be implemented?

@cdrini already has the majority of annotations integrated into BookReader already as per #1214. It's proposed that the individual snippets would live in archive.org patron items (json of the book, page-range, region, optional comment). There would be an annotations server that indexes all snippets across patrons as to make them queriably by/per book item or in aggregate.

Product Considerations

• The BookReader would continue to use the hypothesis UI integration to select and act on a BookReader region.
• Instead of (or in addition to) hypothes.is as a backend, raw annotations would be saved to the patron's archive.org item publicly. The technical implementation and policy considerations need to be reviewed.
  • How [files, format, schema] would annotations be stored on a patron item?
• Similar to https://github.com/archivelabs/pragma.archivelab.org, a simple Internet Archive web Annotations server/service would be hosted on the prod cluster that indexes patron annotations and makes them searchable (a) in aggregate [e.g. newest clippings], (b) per patron [to see your saved annotations], (c) per patron per item [to fetch patron's annotations on an item], (d) by/per item [to fetch all annotations for an item as a mode to explore community clippings].
• An Internet Archive Annotation Service would require that clippings/annotations in the patron's items are in some way queriable, so the annotation server knows globally what annotations exist so they may be reindexed.
• We'll also need a service that is capable of rendering snippets with the right protections and considerations for social previews.

Additional Resources

Add any other screenshots or assets related to the feature request.

Success Metrics

How will you know if this change is successful? What specific metrics are relevant and how do you expect them to change?

An increase in number of patrons who are clipping quotes.

Stakeholders

@-tag stakeholders of this feature
@bfalling @brewsterkahle @cdrini @mekarpeles @ximm