The Superbook format doesn't appeal to me specifically because it's just a stack of pages without text flowing between them. It sounds like a nightmare to handle when editing.
Hi, creator of the Superbook format and founder of Bubblin here!
Reflow is bad for books. It kills referential accessibility [1] which is a critical feature of codex-style format that follows from the most common definition of a book [2].
> It sounds like a nightmare to handle when editing.
This book was programmatically generated using standard markdown manuscript and a paginator library called m2s [1]. I'd say it was pure bliss for me to edit David's book page-by-page!
Reflow is great for novels and similar narrative prose fiction, and for documents that need referential accessibility, explicit semantic outline structure (e.g., case numbering or similar) is reflow independent and superior to page numbering.
That! …but one with strong layout and referential accessibility of page numbers would be even better! I think we need page-wise referencing on novels and prose fiction too. It opens us up to the possibility of having real conversations and annotations deep inside the content of a book.
Reflow otherwise has always been a bane of _all_ digital books because it wasn't possible to scale content without leaving room for it. That's also the raison d'etre for other "non-web" formats, digital stores and even hardware to exist.
> but one with strong layout and referential accessibility of page numbers would be even better!
Page numbers are a small additional value; pagination comes for free with print so it's an easy fallback, but there's a reason why content that is most intensely dependent on referential accessibility, like codified laws, doesn't rely on it.
> I think we need page-wise referencing on novels and prose fiction too.
It's of some use for criticism, but criticism isn't the purpose of most prose fiction. It's certainly not needed for reading, and resizability does more for the primary use than page numbering. (Of course, if you really need it, “referencde page numbering” of the type used in presentations of paginated content where page number references are commonly used like the Federal Register or court decisions when reproduced in contexts with different than source (or no) pagination can be used.)
> Reflow otherwise has always been a bane of _all_ digital books because it wasn't possible to scale content without leaving room for it.
I've seen plenty of books that aren't prose fiction that work well with reflow; PragProg epubs are particularly good that way.
> That's also the raison d'etre for other "non-web" formats, digital stores and even hardware to exist.
No, it's not; even to the extent reflow is a problem for some content, PDF had total-layout-control solved before any of the alternatives, and ePubs had strict pagination and layout control as an option since 3.0; the reason for the alternatives seem to be the same reasons for any other content toolchain, distribution, and consumption ecosystem walled-garden—it's not about serving a market need, it's about capturing a market segment and creating a cost to switch out so that you can charge monopoly rents.
Not at all. Use paragraph numbers instead of page numbers. You can create an 'id' attribute for each paragraph: <p id="para45">...</p>. As you scroll the viewer can update the location bar with a suitable fragment id: chapter3.html#para45
Reflow is bad for books. It kills referential accessibility [1] which is a critical feature of codex-style format that follows from the most common definition of a book [2].
[1] https://bubblin.io/blog/referential-accessibility
[2] https://en.m.wikipedia.org/wiki/Book
> It sounds like a nightmare to handle when editing.
This book was programmatically generated using standard markdown manuscript and a paginator library called m2s [1]. I'd say it was pure bliss for me to edit David's book page-by-page!
[1] https://github.com/bookiza/m2s