Quote:
Originally Posted by DaleDe
But of course the paper book could be changed so that doesn't really work either.
|
I forget where I discussed this (probably the Amazon Page Numbers topic), but physical editions don't get tweaked too often, they only get updated when a sufficient amount of major revisions come along, and/or they have methods to keep the page/line-breaks EXACTLY the same (kerning, tracking, etc.).
Print typically only has to deal with 1-3 versions (Hardcover, Softcover, Large Print).
With Print, they also only care about the LOOKS of the final output, who cares about the underlying mess.
With ebooks, the underlying code
matters, and it is easy to fix any typos/mistakes. As you can see above, the amount of minor variants can quickly balloon out of control.
Quote:
Originally Posted by DaleDe
However, generally speaking most authors don't revisit a book the way you said with no reason except to just change things that result in the same presentation.
|
Sure, I could see where a lot of self-published authors won't be going back through to touch up older books (although as Hitch mentioned, a lot of the work at her shop is redoing older conversions, and a majority of her clients = self-published).
But these ebooks will have to be updated
eventually, especially when different/better formats/devices come in the future.
(For example, tiny thumbnail-sized images may have been considered "okay" many years ago, but that just won't fly any more on high-def Kindles. So a 2017 reader may return the old ebook to Amazon, citing formatting/usability issues.)
And on the publisher side of things, I would still say many are going through and redoing a lot of their backlog.
See article in Post #24 discussing backlog of a BPH. I suspect the same thing is occurring at mid- + small- + academic-publishers as well.
That article resonated with my experiences exactly, and I am just a single one-man conversion show, working in my little niche of books. :P
Accessibility
A lot more Libraries+Publishers are making that a higher priority, and are trying to go back through their old books and bringing them up to date. Example,
"Creating a Roadmap for Accessibility" speech given at ebookcraft 2017.
I would recommend watching a lot of the other speeches from that conference.
The Digital Reader wrote an article about it when it happened. I also wrote summaries of my thoughts on each speech via PM to a few friends... if anyone is interested in that, let me know.
EPUB2->EPUB3 Transition
I suspect backlogs will also slowly update from EPUB2->EPUB3, where a lot of semantics will be wrapped around things like footnotes:
EPUB2:
EPUB3:
(28 character difference * # footnotes = a lot of bytes)
InDesign->EPUB
I haven't even brought this behemoth up yet...
A ginormous amount of books are produced using InDesign, and we all know the spaghetti code from an InDesign EPUB.
Side Note: I shudder to even think about all the hideous spans.
In this post, I showed an example of HTML of a
single paragraph from InDesign going from 891 characters -> 440 characters (manually cleaned).
A publisher might decide to go back to their InDesign file and push the "EPUB3 button" instead.
Maybe later InDesigns will output cleaner code, maybe the designer can push some buttons and tell InDesign not to output as much crap, but I bet your bottom the final size of the raw HTML will be vastly different... throwing off the
Byte Methods.
And who knows how much HTML+CSS differences there would be between an output from InDesign CS5 -> InDesign CC 2015 -> InDesign CC 2018. That would be something interesting to test.
Side Note: Just today, I read an
article on the updates in InDesign CC 2018:
Quote:
Export HTML Improvements
If you’ve ever tried to export HTML from InDesign, you’ve probably noticed that it loves to add class and id tags to almost every paragraph and object—it can be very overwhelming and a pain to clean up. Fortunately, a new item in the Advanced panel of the HTML Export dialog box—labeled Don’t Include Classes in HTML—allows you to exclude all these extraneous materials. It removes the class and id attributes present in the tag, and even many redundant div tags present in the HTML, to create clean-looking HTML and remove unnecessary grouping of content.
|
Quote:
Originally Posted by Hitch
Pssssst: don't tell Wolfie! 
|
We all know:
- Kobos are the only valid ereader.
- ADE is the only page numbers.
- EPUB2 is the only ebook format. (EPUB3 can go shove it.)
- Clean code is the only valid code.
- All unused CSS classes need to be thrown away.
- JSWolf's Charis SIL variant is the only font you should embed!
- iPad =/= iBooks.