[llasram]

Originally Posted by acidzebra

While you are obviously more well-versed in the whole backend/format side of things, when you say "better" do you mean better because publishers appear to accept it, or better because it is 'technically' better?

I don’t know much more versed I actually am, but I meant that EPUB is the “best” e-book format publishers are actually producing books in. I see at the best for three reasons:

• Technical superiority. And specifically, technical superiority through existing standards. The current state of XHTML+CSS represents decades of experience with the largest collection of “reflowable content” information ever assembled by the human species. HTML may not be perfect, but the odds that a competing “book-specific” format will do a better job are pretty slim. Adding vector graphics with SVG just cinches the deal.
• Openness. Sans the keying aspects of DRM, all of EPUB is openly documented. Anyone has all the information they need to produce tools which both produce and consume EPUB documents, making possible much richer ecosystem surround EPUB content. Moreover, because EPUB leverages existing open standards, many tools to work with and produce e.g. the XHTML+CSS of the OPS already exist.
• Existing tooling. An EPUB book is just OPS+OPF in an OCF container (a ZIP archive with some metadata). Most commercial e-book generation tools can/primarily produce the formats publishers sell from OPS+OPF (or the older OEBPS) as source. I don’t have direct experience here, but my anecdotal understanding is that publishers’ e-book production pipelines thus generally involve producing OPS+OPF then converting that to the various saleable forms. Because EPUB ultimately is the same format as what the publishers are converting from, it has the highest fidelity to what the publisher is producing in the first place.

Originally Posted by acidzebra

When I was looking into what makes BBeB, I stumbled over a blog post by Bill McCoy, the General Manager of ePublishing Business with Adobe Systems Incorporated.

It had this interesting tidbit:

...

He goes on to predict "in the long run I believe the momentum behind interoperable XML-based formats is unstoppable." but still, I thought his views on BBeB were interesting.

Yeah... I’m not sure what exactly he’s getting at there. I’m not sure how interesting the choice of archive format and “compiling” XML into a binary representation really is, beyond the fact that it doesn’t seem to save much/any space vs. just using DEFLATE. LIT’s markup-encoding at least amounts to little more than attempted compression and perhaps some limitted amount of pre-parsing – I must confess I haven’t stared at the BBeB internals enough to know how similar it is.

As for capabilities, I think what he says next in that post is spot on:

Assume the industry successfully establishes an XHTML-based reflowable document format based on the evolution of OEB, with an associated single-file container package with pluggable DRM, then I see no strong raison d'etre for Mobi, BBeB, or any of the other OEB-derivative eBook formats to hang around forever.