View Single Post
Old 02-13-2008, 04:13 PM   #14
Peter Sorotokin
speaking for myself
Peter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it isPeter Sorotokin knows what time it is
 
Posts: 139
Karma: 2166
Join Date: Feb 2008
Location: San Francisco Bay Area
Device: PRS-505
Quote:
But for example, a book from Proust, is at best, divided into two parts. A single sentence can last through multiple pages.

How can the content producer deal with such a file ?
Hey, you are not making it easy for me, are you? I have to agree, arbitrarily breaking a text like that feels like a crime against literature.

All I really want to say is that it won't work all that great with the current version of our software. Moreover, it will always work best if a book which has natural breaking points is broken up into separate flows. At some point we'll figure out how to deal with books like cannot be easily broken into chapters as well, but we have to start somewhere.

Quote:
Am I missing something about processing the EPUB format? If there is a page break at the start of each chapter (say), why does it matter whether chapters are in separate flows or not. Even if they are in one flow, it should then still be ok to process each chapter separately (perhaps with a rescan to capture CSS info or other metadata). If this is not the case, that would seem to be a serious deficiency in the format.
It is possible to insert page break just by using CSS page-break-before:always, of course. However when writing software to process it, especially if your code has to work on a small device, there are a lot of differences between the page break in the flow and page breaks between the flows. You are right that it is possible to format starting at any page break, including a break inside the flow. However, CSS cascade and processing (e.g. counters) have to be done for the whole file - that's just how CSS works. That is also possible to optimize. The problem is that taking advantage of these possibilities is not all that easy. And by the nature of the format, the content which is broken in many flows will always work better than the single-flow content.

Quote:
Too bad it's required in this beta. I don't want to have an Adobe ID, so I wont test this beta ...
Sorry to hear that.

Quote:
Does this mean any epub document, even those without DRM, won't be able to be transferred to handheld devices, or other computer ?
There is are no software-enforced restictions of non-DRMed content, of course.

Peter
Peter Sorotokin is offline   Reply With Quote