Extensibility of the docs-editor collaboration layer for non-linear document models

• I have searched open and closed issues for duplicates
• This isn't a feature request
• This is not a report about my app not working as expected

Question

I've been reading through the docs-editor codebase and I'm
trying to understand how abstracted the collaboration/CRDT layer
is from the document model itself.

Specifically: is it designed in a way that could support
non-linear document models — e.g. a slide-based structure where
each "node" is a slide rather than a paragraph?

I'm evaluating the feasibility of building a community-contributed
presentation editor (docs-slides) that reuses as much of the
existing architecture as possible — shared crypto, collab, and
@proton/components — while following the conventions in the
contributing guide.

Before investing time in a proof of concept, I wanted to check:

1. Is the CRDT/collab layer abstracted enough to support this?
2. Are there internal constraints or planned refactors I should
   be aware of?
3. Is this kind of external contribution something the team would
   consider reviewing?

Start by reading the docs-editor codebase, focusing on how its collaboration/CRDT layer connects to the document model, then review the contributing guide and @proton/components usage. Determine whether a slide-based node structure is supported, identify internal constraints or planned refactors, and document whether the team would review a proof of concept.