I am particularly interested not only in selling books but in promoting reading, I think one of the great thing the iPad provide is precisely not to allow tow programs running at the same time. I disagree with you when you compare it with the Android.

My greatest surprise is that this unique devise really make reading a different experience. It allow you to think as does the paper book. That is not the case with Kindle and or any other form of screen reading.

It is a game changer, great for books, authors and publishers.

Two big surpises for me are the quality of the sound (WOW, esp. vs the iPhone) and the fact that GPS is enabled. Would have waited for the 3G model except for the MiFi app that turns iPhone into a WiFi hotspot. One disappointment is that Apple has crippled the Bluetooth so that my Bluetooth headset does not work. Am using the Line2 app that turns iPad into a full-featured phone but would like to be able to use a BT headset.

ENJOY

Perhaps you know that EPUB is coming up for an update. Maybe someone should suggest that an official testing suite be developed as well so that any file that passes the test suite is “good”. That way publishers are not left to an unofficial party’s work to validate their files. In a previous job I worked with geospatial data and in that environment there are a set of standards that have reference implementations attached to them that are the “Good Housekeeping Seal of Approval” for each standard. If your tool/data passes the test, it’s guaranteed to be valid to the spec. Also any changes in the test require a new version of the spec.

I would hope that in this case, publishers (and conversion houses) will learn that they need to be aware of the requirements of the spec, in order to do their own acceptance QA on finished EPUB file, whether they generated them or someone else did. Perhaps they should negotiate a maintenance/correction clause into their contracts to cover instances like this in the future.

As far as where the consumers stand in all this: if someone buys a book from Sony, it will work on his Sony reader; the same with Apple. Because of the DRM wrappings the files aren’t transferable anyway. The version of epubcheck being used really doesn’t matter in this case.

The real issue would be if Sony decided that they were going to do the 1.0.5 validation and then did it in such as way that all the old files they distributed are now invalid. It is the vendors who previously distributed data at the 1.0.3 level that need to figure out how to handle the later versions in a backwards compatible manner. Another instance might be when one vendor (e.g., Sony) moves and another (e.g., Kobo) doesn’t but they both use the same DRM scheme (Adobe). In that case we might see incompatibility issues.

It is hardly the fault of publishers if a previous version of epubcheck failed to validate properly. We can only go by what the standards bodies release to us. And of course, the vast majority of publishers do not create their own epub files, outsourcing to businesses that do the conversions, big and small. But I don’t need to tell someone from Aptara that!

So it’s perhaps a little disingenuous to claim that “publishers allowed sloppy files to be released”. Publishers have mostly bought in a specialist service and then relied upon the standards bodies for a validator which passes those files. Publishers have pinned their QA to standards (be they de facto in the case of epubcheck, or de jure in the case of epub itself). Are publishers to blame for every new iteration of epubcheck then which requires more upgrade work? Surely you can see that’s not sustainable? If publishers get their fingers burned every time a new validator rejects files which were validated in a previous iteration and therefore correctly (at the time) released into the digital supply chain, then having to go through the process of recalling and upgrading, duplicating their work and costs, what do you think that might do to the still-nascent ebook market?

In the case of Apple, I agree now that it is not their fault. But it looks like there should be questions asked about how so many previously-ratified epub files are now potentially unfit for purpose. The industry looked towards epub as a solution to the highly complex product variety out there for customers. But now we see epub on Sony (wrapped in ACS4) as incompatible with Apple epub (wrapped in Apple DRM) and vice versa. We see old epub files that were validated with a de facto standard tool (epubcheck) now being reclassified as invalid. And where is the end user in all this? Still just as clueless as to compatibility, interoperability, and choice. It would be great to have you write a separate blog entry on epub and perhaps start a new debate here.

IDPF didn’t release a new version of the EPUB spec (which is currently v2.0). The developers (mostly at Adobe) of a validation tool called epubcheck released a new version of their software (1.0.5) that Apple has used to validate their EPUB files (and maybe included that validation in their iBook/iBookstore process). I don’t see any irresponsibility on Apple’s part for using the latest version of the software. Neither is IDPF to be held accountable for anything since they didn’t change the spec. To the best of my knowledge the software has no official standing within the IDPF.

The main changes between 1.0.3 and 1.0.5, as I understand them, are improved validation of some items that weren’t be validated before. So technically, files validated using 1.0.3 might not always be valid EPUB files. Publishers who would like to display on the iPad should upgrade to 1.0.5 in order to pass the validation. In doing so they are producing better EPUB files that should still work on previous eReaders. Another change is to flags any files included in the EPUB file that are not listed in the manifest. If these files are not being used in the eBook, then having them in the file is wasted space and wasted bandwidth for those organizations which support the download and distribution of the files.

If Google is going to use epubcheck to validate the Editions files, I would image that they would use the latest version of the software as well. Why would they want to release files that may not be completely correct? Will any reader they develop allow files that are valid to 1.0.3 to be read/distributed? That’s their call.

So Apple was not slack/sloppy. If anything they might have hurt themselves by limiting the existing base of files that can be read by their reader and sold in their store. If you view validation as a silly obstacle, so be it. The publishers allowed sloppy files to be released and the updated tool is now identifying problems that crept through.

Now, IDPF released v1.0.5 on 22 March, not even three weeks ago. So this effectively means 99.9% of epub files owned by publishers out there need to be reconverted to the latest version. Good news for the likes of Aptara but not for publishers!

How can Apple be so slack? Why would they neglect to make the iBookstore compatible with ALL versions of epub? Will this mean Apple (or other new entrants in future) will keep placing a resource-load onto publishers for reconversion every time IDPF releases a new version of epub?

Ultimately, Apple may lose out. Because Google Editions is just around the corner folks, and you can bet your bottom dollar they ain’t gonna be placing any silly obstacles in publishers’ way. Google Editions will likely have, therefore, a wider selection as many publishers, I imagine, will produce new titles as 1.0.5 but may not spend precious cash reconverting their entire backlist just because Apple got sloppy.