Seed Documents are organized in a hierarchy with path names. Logically this behaves like a file system where each folder is a document (or has an index).

Each document is also a hierarchy of blocks, within the content field. Documents are managed by creating operations which are packed into Change blobs. The changes reference previous changes, and all the operations are walked to reach the current document state. Refs are used for naming the document and locating it, and they can reference any change.

Blocks include Paragraph and Heading which constitute the text material.

Docs can embed other Resources

Comments have a "TSID", an auto-generated ID that acts as a non-human-readable path.

They are edited by creating a new comment blob, and deleted by editing the comment to be empty.

When importing you must follow the hierarchy convention of documents with human-readable path names, and each document must be correctly organized

Imported content must include the importedAsset metadata field which references the source material that is archived to ipfs with UnixFS, so the field will contain the ipfs:// URL so that readers can view the source content.

Import Web

Importing Markdown

Importing PDF