Book protocol

The HPub specification defines a book:// protocol to identify books. Here is an example of the syntax for a book URL:

book://example.org/whatever/whenever/publication.hpub/filename.html#anchor

Two great advantages:

• We are able to uniquely identify a book: being a URL the path is unique and it can be used to identify:
• a uniquely owned domain (i.e. example.org)
• a specific path (i.e. whatever/whenever)
• a specific publication (i.e. publication)
• format (i.e. hpub)
• Once we have identified a book, we can add further elements to the URL, to refer to specific items inside it:
• filename (i.e. filename.html)
• anchor (i.e. #anchor)
• We are able to download a book: the protocol used is the standard HTTP, so you can easily replace book:// with http:// and download the book from that specific location.

Notes:

0. The specific extension can be omitted: in that scenario, the reading software might try to download the extension that is best suited. Baker will download .hpub files by default, but the book:// protocol is agnostic so it could point also to a .epub version or .mobi version or even a pure textual .html version.

Usage in Baker

Baker understands book:// links inside books. When the user clicks on such a link, Baker will automatically download and open the book the link points to, replacing the current book with it.

Baker supports a special URL: when clicked, a link pointing to book://local will go back from a downloaded book to the default book originally contained in the published Baker app.

This will allow:

1. Manual updates of the current book.
2. The creation of a lightweight version or some kind of "bookshelf" to download books: the default book in the app will just be a collection of book:// links, allowing the users to download the book they want (using book://local to go back to the "bookshelf" and download another one).
3. The creation of chains of books, of downloadable references and so on.