MobileRead Forums - View Single Post - Article: The publishing industry has a problem, and EPUB is not the solution

gmw · 09-01-2013, 10:07 AM

I don't know much about EPUB3 - other than it seems to be biting off more than might be good for it, but I see EPUB2 as pretty much the same sort of thing as ODF (Open Document Format) used by OpenOffice etc, but dedicated for ebooks (which is good). EPUB solves a lot of problems for which HTML has no elegant solution. ODF and EPUB allow a complex document to be both modular (an advantage for software, and for those debugging content problems), and a single file (an advantage for the user and distributor). It's a simple and elegant solution (or it could have been, had it not been designed by committee).

And, thanks to the wide acceptance of EPUB these days, there are effectively only two standards to worry about any more, and Amazon provides the tool to go from EPUB to Kindle so I really only have to care about EPUB. I can't speak to the vagaries of different EPUB2 support, my requirements so far have been fairly simple. So far so good.

Where I do agree with the article is that the emphasis on the EPUB standard seems to be focused on obscure things or publishing industry things, rather than what is actually useful to the users of EPUB files (readers, publishers, distributors etc.) - including the software. For example no one can agree on what exactly should be inside the description field of the metadata (eg: HTML or other formatting capability - even just simple paragraphing - is not defined). As a result, it appears that the metadata of an EPUB is mostly ignored (despite the many articles telling you how important it is), with resellers getting their information from the publishers/distributors directly (and almost as unreliably), rather than from the EPUB. As far as I can tell EPUB metadata has become the purview of end-users with software like Calibre.

The article makes great claims for the wonders of using HTML files directly, but it's taken years for it get to where it is now - and it's far from perfect. EPUB is as close as we have to a standard for ebooks now, I don't see it going away. EPUB3 may turn out to be a mistake, in which case some other variation of EPUB will take over or it will be adapted until it works. It will all work out in the wash, it just takes time.

09-01-2013, 10:07 AM	#3
gmw cacoethes scribendi Posts: 5,818 Karma: 137770742 Join Date: Nov 2010 Location: Australia Device: Kobo Aura One & H2Ov2, Sony PRS-650	I don't know much about EPUB3 - other than it seems to be biting off more than might be good for it, but I see EPUB2 as pretty much the same sort of thing as ODF (Open Document Format) used by OpenOffice etc, but dedicated for ebooks (which is good). EPUB solves a lot of problems for which HTML has no elegant solution. ODF and EPUB allow a complex document to be both modular (an advantage for software, and for those debugging content problems), and a single file (an advantage for the user and distributor). It's a simple and elegant solution (or it could have been, had it not been designed by committee). And, thanks to the wide acceptance of EPUB these days, there are effectively only two standards to worry about any more, and Amazon provides the tool to go from EPUB to Kindle so I really only have to care about EPUB. I can't speak to the vagaries of different EPUB2 support, my requirements so far have been fairly simple. So far so good. Where I do agree with the article is that the emphasis on the EPUB standard seems to be focused on obscure things or publishing industry things, rather than what is actually useful to the users of EPUB files (readers, publishers, distributors etc.) - including the software. For example no one can agree on what exactly should be inside the description field of the metadata (eg: HTML or other formatting capability - even just simple paragraphing - is not defined). As a result, it appears that the metadata of an EPUB is mostly ignored (despite the many articles telling you how important it is), with resellers getting their information from the publishers/distributors directly (and almost as unreliably), rather than from the EPUB. As far as I can tell EPUB metadata has become the purview of end-users with software like Calibre. The article makes great claims for the wonders of using HTML files directly, but it's taken years for it get to where it is now - and it's far from perfect. EPUB is as close as we have to a standard for ebooks now, I don't see it going away. EPUB3 may turn out to be a mistake, in which case some other variation of EPUB will take over or it will be adapted until it works. It will all work out in the wash, it just takes time. Last edited by gmw; 09-01-2013 at 10:09 AM.