04 April 2013
July 2013 was ePub3 month. After 22 months of stalling and prevarication is something about to happen. Watch this space.
Let's start off with a little fun with this little Tumblr item Let's Talk About ePub3. This little three step graphic probably says more about ePub3 than all of the long-winded narrative following!
10 July. Setting the July 2013 ePub3 tone, my opinion rant that ePub3 was a dismal failure. This article is a continuation of that thread.
17 July. The first E-book Zero (E0) test set is prepared and available.
24 July. The AAP announcement of an ePub3 adoption initiative with a strong focus on accessibility.
25 July. A DBW article Aggressive casting doubts in a news-mannerly way on the AAP announcement.
25 July. An IDPF article is released in DBW "Why Publishers are Making a Push for ePub3 Now".
30 July. An IDPF announcement. Call for Participation: EDUPUB Workshop.
Let's get into this and try to understand the real facts and effect on ePub 3 adoption by e-retailers, reading-systems and publishers, remembering that if all three are not fully synchronized nothing much can happen.
In the spirit of ePub3 interactive content, the original article extracts are in expanding interactive expanding panels (blue) and text popups (green). Just click on the blue text to hide and reveal the article extracts associated with the comments, or click the popups to reveal the wisdom within.
First the press release by the AAP.
The Association of American Publishers supports the establishment of EPUB 3...
This initiative is potentially heartening. Especially looking at setting a baseline for education and accessibility although trade-book e-retailers will probably win the ePub3 spec. feature minimization.
If EPub3 is good for anything it is definitely education, training and learning content. It is essential that a baseline of features exists that supports these genres, not just trade books.
Unfortunately there are no names, no specifics and no details of support by any particular publisher, reading system implementor or retailer as commented by others. But here is hoping enough get on board.
Next in the July ePub3 action was a long article in Digital Book World, written by a member of the IDPF, that needs a long commentary and analysis to provide international balance. (International in this context means not U.S., Japan or Korea.)
Now I understand that this article was a PR exercise, trying to get people excited and interested in ePub3 given the AAP announcement. There are a number of swooping, soaring obvious generalizations, so we are just providing a back-to-earth landing ground. Optimism is a great thing, but never believe your own PR!
And the story starts..... now! Once upon a time, in a land far, far away...
The Association of American Publishers (AAP) announced...
Will this initiative result in accelerated ePub3 adoption. We stand poised in anticipation.
"Some people" are completely correct. Why is there such a fuss about moving to the new ePub version? EPub2 does it fine for trade and linear reading text books. Why can't ePub2 work with ePub3?
It's only a reading system thing. AZARDI does it without batting a byte; ePub2, ePub3, E0; throw it at me and I will present it my way. The specification shouldn't have tried to be backward compatible, it should have broken new ground and left version compatibility in the hands of reading systems developers.
The explanation is simple: While the long term benefits of moving...
It's difficult to understand the point that is being made here. Guess I am dumb (Damn it. Aunt Molly was right). I think it is the future history of interactive reading experiences in a paragraph, but am not quite sure.
The XHTML of ePub2 is the XTHML5 of ePub3 with a few elements added in here and there. Amazon have done well with HTML1.00 and CSS 0.5 for the past five years. There is more at play than locking on to HTML5.
EPub3 does not lower the cost of eBooks. We have delivered thousands of ePub3s over the last year and it costs a lot more money to produce a novel as an ePub3 than ePub2 for a number of reasons (that is another post).
There is nothing in the ePub3 spec about the wide range of devices, viewports and aspect ratios people are using. Why is this relevant in 2013 when the specification was written two years ago before the Phablet was invented!
To understand this it's helpful to consider what version 3 adds...
Can I just be a little small-minded and point out that it is a specification not a standard unless the IDPF became a standards body, instead of a trade organization, while I wasn't looking.
Hopefully the AAP meeting and EDUPUB group will make Fixed-Layout spine properties mandatory and introduce a few new properties for consistent textbook behaviour. Currently only AZARDI has a complete implementation of the Fixed-Layout spine properties. We demonstrated how spine properties can provide enhanced education experiences in the ePub3 concept book Famous Paintings (download). No other reading system even tries these real education concepts. It's not hard but does compete with proprietary education production applications. Don't expect a lot of changes.
In real world implementations we have found that not having a rendition:layout-scrolling option makes it impossible to create education ePub 3's of value. This is the sort of discovery that can only happen when REAL implementation testing is employed.
All of these things are fine with the exception that SVG and MathML are worth their own list number. Let's call it 5. They are very big in education content.
So while these four cheerful points sell ePub3 up, it is the unspoken Item 6 that promotes the dialogue of the anti-ePub 3 voice. What is not mentioned is the cluster of strange features such as Canonical Fragment Identifiers, trigger, switch, manifest fall-backs, images in spine and eye-watering metadata strategies to name the major items. It is these that have significantly slow adoption of ePub3. Hopefully these will be addressed and prioritized by the AAP and EDUPUB meetings.
Many people think of HTML5 (and thus EPUB3 which differs...
There are a number of very large association jumps here.
People don't think about HTML5 as rich media and interactivity. That's Flash! People think of HTML5 as being a timely and necessary correction to the XML locked-down direction the XHTML specification was heading. The same way E0 is a correction to the direction ePub3 has gone and has stalled for the same reasons.
EPub3 differs from ePub2 primarily being built on HTML5? Firstly ePub3 is built on XHTML5 which imposes a number of required limitations in the ePub packaging environment. That should be stated clearly. It means SVG and MathML have to be packaged with name-spaces for example; loosing the liberation from name-space content death HTML5 delivers. But we understand that compromise as all our production is well-formed XHTML.
EPub3 differs from ePub2: 1) starting with the metadata disaster waiting to happen, 2) the manifest packaging with un-required properties and fallbacks, 3) a spine supporting fixed-layout properties, 4) a semi-useful landmarks strategy (if it is actually used sensibly), 5) a useful if limited page navigation strategy, 6) a cover definition (but not how it is to be used); and many more differences in the details. There are a lot of DocBook ePub2 books out there!
HTML5 is not a marketing term! It is a W3C specification. It was an independent action by a group of individuals who saw the W3C direction on the modularization of XHTML was a looming disaster. HTML5 was a reality business check. It brought the W3C to it's collective knees and made it rethink, and trash what it was doing with XHTML.
This trivialization of significant events in the recent history of standardization efforts is somewhat of an understatement to the actual courage demonstrated by the WHATWG working group. Check out the history of HTML5 presented with this infographic.
In some areas EPUB3 has gone beyond the browser...
EPub3 has gone beyond the browser baseline.... What does this mean? What is a browser baseline? I haven't got the foggiest notion of a browser baseline. Educate me someone.
Global language support has been tirelessly driven by the Unicode consortium for 20 years making support of all languages in any computer platform a reality. It is one of the greatest and most marvellous technical achievements of humanity which we now grunt at - oh yeah UTF-8, whatever you say dude! Unicode is the DNA of language. It is so absolutely, magnificently....awesome.
UTF-8 and UTF-16 are now the universal encoding declarations. Helping a few Japanese Manga people get vertical Kanji rendering in their comics is not really enabling global language support.
The "broader web platform" evolves at the speed of light compared to the ePub3 specification which is locked in pre-2010 CSS and 2007 reader features suitable for eInk devices. EPub3 only partly adapts to the modern tablet era. Browsers and their rendering engines drive it. IPad/iOS, Android and FirefoxOS are all children of the "broader web platform" which have emerged since 2007.
Ebooks being sold today in the United States have been...
The IDPF seems to be only aware of two languages apparently. He doesn't know that of the six billion people on earth only 300 Million speak Engerish, Engerlish or whatever you coll it; and 125 million speak Japanese. It is expected that a "so called" international specification should at least be interested in more languages than English, Japanese and Korean.
Japan has enough money and technical resources to look after itself and contribute back to international standards. Vertical writing is NOT critical except to the Japanese and this statement is isolationist, not significant in the importance of ePub 3.
Spanish is the interesting one. 400 Million+, the second most spoken and distributed language in the world. Ahead of English. 50% of AZARDI downloads are to Spanish speaking users.
Instead, it’s the other two features of EPUB3 that are driving...
Accessibility is not only important because the government says it is important. It is important because it reflect our evolved humanity. Or not.
Making content accessible will never increase sales. That is at best a hopeful thing to say. The reason anyone should support accessiblity is straight-forward human inclusionism because we can with just a little extra effort.
It is not going to be possible to create fixed layout textbooks that are visually-challenged inclusionist without a lot of additional production effort. That is a core fact. Can it always be done for all content? No it can't.
Serious and high-quality accessibility is something that must be done to the extent possible because in 2013 we have gone beyond pitying minorities and have the will and resources to treat them not as a special part of the community, but just people.
Any blind person out there "reading" these statements would just shrug their shoulders and say "same old same old".
The overall benefit really more...
One ePub file to all? It will never happen in the U.S. but probably will in other countries around the world.
At least the one statement that has a ring of truth is "supporting some ePub3 features". However if you support the left features, and I support the right features, and they support the middle features; that is the same as either no support or dumb-down to the common denominator support. Oh yeah! That is an ePub2 replay where everything has to be dumbed down to ADE, or import to Amazon.
This is where the AAP initiative could get that dumbing down under control, and even possibly with a commitment from e-retailers and reading system developers. However please do not hold your breath!
But with many retailers and reading app developers still stuck...
No reading app developer is stuck on ePub2 except Adobe.
Get across the river! GET ACROSS THE RIVER! In the next six months! After two years some have found the river, but don't yet know how wide it is. Sadly it will not happen for North American publishers, because Apple has already proprietized ePub3 into a minimal spec that fits their business interests. The same way Adobe did with ePub2.
But that’s just about optimizing what we are already doing...
Fixed-layout is the part of the ePub3 specification that makes real sense. EPub3 has little value for linear reading at all. Not even typography is an issue really. While the cognoscenti may whine about e-book widows and orphans (without understanding how reading disruptive would look like on a 4 inch device) ePub2 has delivered the goods for those genres.
EPub3 fixed layout only has one spine-property missing in our understanding : rendition: layout-scrolling which is an essential required for text books.
What no-one in the IDPF is talking or will talk is the fixed-layout spine properties. The only reading system that implements these is AZARDI (Download 7MB the demo/test book FLO Grows Up and read it in AZARDI to understand a complete implementation of the IDPF Fixed Layout specification) . It would appear to be unlikely that the spine-properties will be supported by any other reading system any time soon or at all. Everyone will mimic the Apple proprietary implementation of ePub3. This is a shame because this is the one feature that will make textbooks possible in ePub 3.
For e-textbooks it’s going to be de rigeur to have integrated...
That is a 2012 statement of what the proprietary HTML5 packagers are doing now (iPublish, Inkling, etc). I certainly hope there will be a lot more than this.
How education content and "Manga" can be wrapped into the same paragraph is a bit strange? I am serious about this. These are not the same issues with similar dimensions. The future of 2 billion children bundled in with "Manga".
I am thrilled to discover that Manga and comics are evolving to become "motion books". And I hope history books have a lot more than boring slide-shows and video. Maybe a Manga motion history novel is just what we are all waiting for!
Stepping back even further, what EPUB is really...
I personally think what it is really about is a little bigger than that. Let's throw in HTML5 again. A magic bullet certainly never hurt Mary Poppins.
The publishing industry is moving inexorably to the Web...
In 2007 it was a little newer than that. It was more of an upgraded OEB than a fork of the Web with a focus on eInk displays as the future of reading. However ePub2 did a good job getting us seriously into digital reading systems even if it was the non-ePub hammer-blows of Amazon that really made it happen in the U.S.
PDF which is circa 1993 definitely doesn’t get us there....
Excuse me! PDF is what we use to produce print books, and it has evolved and become a highly useful international standard. Sure it is not a really suitable ebook format, but tell the hundreds of education publishers I talk to who are looking for a STRATEGY to move their print content to digital.
EPub3, more than any other specification seems to be out of step with it's potential adopters. Publishers are not really a consideration.
Once this transition is complete, publishers will have...
What is meant of course is that bits of ePub 3 could be used for distribution via your own website if you drop the OPF, manifest and crazy unrequired properties and fallback nonsense, spine, concrete paper-oriented TOC, page navigation and other navigation options, indexes that are un-ordered lists (sic), proposed dictionary XML grammars that are XML-101 child exercises, a CFI specification that is meaningless.
Sure then we can spend a zillion dollars processing brainless ePub 3 stuff to a App manifest or website. Or we could just start with E0 and spend 1% of the production cost of creating, checking and validating an ePubX.
Of course you don’t need EPUB to develop a website...
An EPub2 or 3 package would be one of the worst possible foundation constructions to build an website or native app. To actually get the navigation, spine and other information from the XML OPF is a lot of un-necessary work. Even the highly criticized WebApp manifest file does a better job.
That is why the E0 format is so important right now.
It’s a very painful thing to leap forward a dozen years...
Leaping forward a dozen years in one revision is a bit of an imagination stretch. The ePub3 specification is a "dozen years behind" problem driven primarily by the concept of backward compatibility with ePub2. This constrained the creation of something new and good.
It didn't have to. It is relatively easy for a reading system to handle a number of packages. It is 2013. It is not "do or die". We are in the age of let's do it all. There doesn't have to be a "migration" to ePub3. It just has to be an option.
In a conversation with a strong ePub3 supporter I was asked "what is the alternative to CFI". My response was there is none. Just drop it. It is digital content specification noise like JPEG compression artifacts.
The problem with the ePub3 specification is it is written like a "My Way" song. Unfortunately in digital publisher in 2013 it is not possible to have too few regrets to mention!
Kanter’s Law is that “Everything looks like a failure...
Of course there are dozens of "XYZ Laws" and corollaries that could be applied here, we could also be in the middle instead of almost at the end after two years: Murphy's Law is the obvious first candidate.Hofstadter's Law looks like it fits right in with the overall scenario, and Hindsight bias could also be applied.
Cute but ePub3 is NOT in the middle. It is 22 months old. Two years coming right up.
The requisite hard work is well underway, and thanks to...
The new AAP initiative is looking at implementing the core set of baseline features that will result in the wider acceptance of ePub3. This will be very good if it happens. I am in full support of the Radium Foundation and wish it the best of luck. A hundred little companies will probably dive on it to look good. It is a slightly desperate approach to a problem that is no ePub3, but the relentless march of technology right past ePub3.
If the Radium initiative does not handle fixed-layout spine-properties completely and as a priority (and it can't because of the way it interpretes and packages book sections), it will be a failure. THIS IS THE REAL DIFFERENCE BETWEEN EPUB2, EPUB3 AND PROPRIETARY APPLICATIONS (he said shouting). Unfortunately this river has no Siddhartha-like riverman waiting to transfer infinite wisdom. It is just a turbulent mess.
Unfortunately I don't think reinventing the book is waiting for ePub3; the publishing business has demonstrated that in 2013 it has moved on without ePub3 for the last two years.
EPub 3 and HTML5 mysteriously connected as if they were one and the same... again. The production format is not just the plumbing. It would be more correct to call HTML5 the foundation and structural framework of digital content and ePub3 a commercial container to package that structure for delivery to specific devices that understand the package. It's not poetic. But it is more accurate.
Content is King? I get the point of the statement and have used it in the past myself. But in 2013 content is a commodity for my channel which wants to be king.
It may be a philosophical point but content has never has been king. With print, audio, video, animation or digital content is the passenger. Accessiblity, readability and navigation are what matters - no regal labels. Content is the servant of these.
Don't read this summary if you are one of those tweety linkin types who don't like product and solution messages in commentary blogs. The reason we provide this commentary is because we have the experience and software solutions to allow publishers to directly address their content management issues. We have ePub3 production tools, distribution solutions and reading systems. Experience matters. Opinion doesn't. We are practitioners.
Publishers! Infogrid Pacific have created product, distribution reading systems and delivery solutions for ePub 3. We know what we are talking here.
There is a definite need to have your content ready for a dynamic digital content future. Just understand one thing: EPub 3 is just another format. Don't get sucked into the hyp of ePub 3 evangelists. It will pass faster than a fast thing.
The most important strategy any publisher, small or large can put into place in 2013 is to produce and manage your content in a future-proof manner so you can produce OEB, ePub2, ePub3, Kindle any garbage, exquisite PDF, SCORM packages, WebApps Static sites, support for legacy XML system: whatever you need. Even ePubX.
That is not addressed by some arbitrary CMS with arbitrary HTML or XML tagging.
We do not speak against the IDPF but just put everything through our production/reading system filter to sort out the chaff from the wheat.
We just want to ensure the correct information is available from real-world practitioners, who understand and can separate the clay from the gold. It's that simple.