On 11/18/2013 4:11 PM, Keith J. Schultz wrote:

Hi Hans,


Am 18.11.2013 um 13:21 schrieb Hans Hagen <pragma@wxs.nl>:

On 11/18/2013 10:00 AM, Keith J. Schultz wrote:

2) Now, what a EPub-READER must implement to handle is very
     little. There are HARDLY ANY provisions that a certified EPuB-READER has
             to implement any particular engine or features therein to display/render
     the information contain in the EPub-file/wrapper.

right, and I'm not going to waste time on it till i have a decent ebook reader that behaves well

The point you are missing is that the ereaders are behaving well. They are following the epub
         standard, and that to the letter of the standard. The problem is that the standard does not
enforce any particular implementation. If you look at the slow progress of the standard that
actually requires a full implementation of the HTML5 standard. That  wait will very long.

sure, and every time i see an epub novel i realize that for something like that one really can stick to rather dumb html ... the point is that one cannot expect context to output simple everywhere accepted html from complex rendered input ...

Furthermore, ereaders are made by companies more interested in profits than spending a few Euros
more to put decent HTML engines into their readers. Why they do not do that is beyond me!

3. Modify the way in which ConTeXt generates the XML files. Ideally, I should be able to write something like

Would be nice if there where commands in ConTeXt or a module for defining what should go into the CSS and a
mode "epub" where the ConTeXt commands are converted to suitible HTML5 structures that are suitiable for
most ereaders.
Features:
                       1) margins in percentages
                       2) font sizes based on em
               3) a new file for every chapter optional for sections user defined
Just a few. Lots more can be found in any decent documentation on writing ebooks.

context outputs xml and as a bonus provides a css too ... one can always convert that xml to his/her ebooks liking .. maybe at some point the mtx-epub script will do that

I always to like to look at programming as modular and would think that a epub/ebook module would be nice that maps
there are commands for layingout ebooks. these commands can then be mapped back to standard context commands.

in that case code in xml and either processit by context or transform it into something ebooks can render

For some interested in producing a epub then can use the conventions for producing ebooks and ConTeXt can provide the
math conversions to regular page dimensions used in PDFs for proofing or creating a printed version. It would also make the
creation of EPubs from ConTeXt a simple parsing exercise.

so far i had no projects where epub was needes so it has a low priority and i still read paper books (or when i would have ebooks i wouldn't need to render them) ... pdfs views quite well on e.g. nexus 7 devices and i assume the upcoming sony high res ebook will also do pdf well