More on TechnologyTell: Gadget News | Apple News
What .epub needs
March 4, 2008 | 3:56 am
|
By David Rothman
From better CSS display consistency to improved standards for covers, Greg Schofield has listed the flaws of the IDPF‘s .epub standard as he perceives them. I’d urge you to comment on his constructive suggestions. Not everyone would agree with him on everything, such as footnote-related issues. Some might argue that footnotes aren’t even needed in E, given the possibilities available via linking.
Gripe away!
Incidentally, Greg worried whether his comments were too long. The answer is, No. You should say what needs saying, especially when it comes to .epub—pro or con. As gung-ho as I am on e-book standards, I want people to complain about .epub to enlighten those those developing it further. Jon Noring, who has played a key role in IDPF standards work, reads and contributes to this blog. And Mike Smith, the IDPF executive director, also follows us. So please, gripe away!
|
Previous
SUBSCRIBE TO RSS
Comments:
Dear David,
For me, the issue with .epub is that the Teleread blog is almost the only place where I read lots about it. On other websites there’s hardly any reference to the standard or its benefits.
I am an advocate of one (non-drm’ed) reflowable ebook standard, and I think .epub might be a good candidate to become such a standard, but it really needs more coverage elsewhere. Or maybe the format should get its own website, because when Googling for ‘epub’ I can’t find a informative website about the standard.
Additionally, Hanlin seems to be the only ebook hardware manufacturer that has hadded .epub support in their reading devices.
Finally, for me it is still not very clear why .epub is ‘better’ than (un-drm’ed) Mobipocket .prc. Any clues?
Wiebe de Jager
Ebookreaders.nl
Excellent questions, Wiebe. Answers:
1. Mobipocket has its merits but is a proprietary format controlled by Amazon, which can always manipulate things at the expense of content owners and users. A nonproprietary approach means that the evolution of a format can be discussed openly, just as we’re doing now. Long term, the format can be more responsive to users’ needs than something cooked up by just one company. Not to cast aspersions on the good people at Mobi. But an open approach is better. Just look how Amazon even tweaked DRMed Mobi to create the Kindle format.
2. .Epub is better suited for scientific, technical, medical and academic publishing.
3. It has been and is being developed in cooperation with the DAISY Consortium, which is interested in accessibility issues.
4. A nonproprietary format is better for open source software development. Meanwhile vendors can compete on interfaces, etc., rather than herding people into proprietary favorites.
5. Not sure about the exact situation re Hanlin, but via the open source FBReader, the OLPC machine and similar boxes can display .epub. More options will be available as more books become available in .epub—already in use as a distribution format at Hachette and a reader-level format by Feedbooks.
6. The IDPF site is full of information on the technical side of .epub. It could be more complete, but Mike Smith just took over as exec director—give him time.
7. I agree that .epub needs more coverage elsewhere. Both enthusiasts and journalists need to be educated about the need to think of e-book standards in the long term. Without nonproprietary standards, e-books will be less important as a serious, durable medium.
Thanks,
David
I remember when OpenReader was touted as the solution to the Tower of Babel and I see the same repeating with epub.
There is no reason to think that Mobipocket, MsLit or even Adobe’s thingies (pdf, flash)will be replaced by epub.
I am curious when we will see epub e-books for sale at any general e-book store like Books on Board
Heck Liviu, OpenReader was what led the the IDPF and publishers to get more serious about e-book standards, which the organization had neglected despite some useful work some years ago. The IDPF, better known to publishers and 1,000X better financed than OR, co-opted the cause. Hey, I can think of worse fates.
Thanks,
David
David, thanks for your explanation. What particulary interests me (since I work in the field of academic publishing) is this: “Epub is better suited for scientific, technical, medical and academic publishing.”
In what ways is epub better suited for this kind of publications? Is there already a epub editor or converter available (a Word to .epub conversion tool would be handy)?
Thank you,
Wiebe
I would agree that the current .epub readers need better CSS support, but I don’t think that’s an issue with the format itself.
As far as footnotes go, I don’t think that the IDPF has to do anything fancy with the format to support them. As stated, linking works fine for them. What I think needs to happen is that the IDPF needs to add a section to the specification with a recommended way to handle footnotes so that they all work the same way for the end user.
Also, all of the software readers need a back button. An easy way to do footnotes, both for the person creating the file, and the person reading it, would be to put all the footnotes in a separate file, and put a link to the footnotes in the text. This would work fine if there was a “Back” button to return to where you were, but some of the readers *cough* Digital Editions *cough* don’t have a back button right now.
Wiebe and Yoda:
W: Couldn’t agree with you more on the need for an .epub editor and/or converter. Jon Noring has been taking about an open source converter for a number of formats. Some funding from the right people could go a long way. And the same for other open source projects: the solution needn’t come from him in particular. I’d love to see the OpenOffice people do an .epub plug-in!
As for .epub vs. Mobi for use in upper-level publishing, one area would be treatment of images. You can use large images in Mobi, but the workflow has to be just-so, and the most convenient tools might not do the trick. Also, how about full capabilities for tables? Perhaps Mobi can correct these specific deficiencies, but their existence, after all this time, gives you a pretty good clue that Mobi was really intended for low-fi reading on a handheld.
What’s more, looking to the future, .epub’s developers are talking about all kinds of neat things such as reliable interbook linking and annotation standards to permit interactivity within books. The desirability of these capabilities would be one more argument for a nonproprietary approach and appropriate standards.
=========
Yoda: Totally agree with you re back button in DE. All the more reason for a nonproprietary format! We shouldn’t have to be at the mercy of one company. At least since we’re talking about .epub in this case, we know there’ll be alternatives to DE.
Thanks,
David
David makes some good points about epub. One thing he forgot to mention specifically is that epub is basically XHTML 1.1 and CSS 2.0 (In addition, the DAISY DTBook markup is supported for sight-impaired users). The container file is a normal Zip archive.
Using these standards not only keeps the format non-proprietary, but lets you use existing tools to generate content. I did an epub using just a text editor. We certainly do need more tools geared specifically toward epub, however.
As for footnotes, the easiest way right now is hyperlinking, as already suggested (except for the infamous no-back-button in DE that I also commented on some time ago). What I would like to see in a future version of the epub standard is markup specifically for footnotes. Something similar to the existing (but optional) oeb-page-head and oeb-page-foot. The problem with doing footnotes, as opposed to endnotes is that the reader software/device has to have some concept of paginating for the screen size and font in use.
I agree that epub needs more exposure. It is mentioned often here at TeleRead and occasionally at MobileRead (very occasionally). As for the IDPF site, I have commented to David before that the IDPF needs to do more with their web site to assist and educate publishers, content providers, authors and readers about the use of epub. Other than the specifications and a few sample files, there is not much else there. The forum is practically dead, and some posted questions have gone unanswered.
As for dedicated readers supporting epub, the Sony 505 should soon have a version of Digital Editions with epub support. Also, the Cybook Gen3 is due for epub support.
Although the epub standard is still new and there hasn’t been enough exposure yet, I think the standard holds great promise and is the best solution available to us right now for reflowable ebooks. All of us need to keep talking up epub to anyone and everyone. A single standard format that is usable on a wide variety of devices is desperately needed.
My own efforts aren’t much, but I have tried to promote epub whenever I can. I have even posted an epub version of a PD book that I converted not too long ago on my web site: http://www.zianet.com/jgray/
One additional comment: In a future epub standard, some of the “optional” elements of the existing standard should be made “required”. I understand why the first version of the standard made some things optional, but in the future, fewer optional features will create less confusion and a more even reading experience on different platforms. Considering the power of today’s embedded processors, this shouldn’t be a problem.
To answer some of the questions brought up in this article’s comments section, EPub is certainly suitable for almost any type of linear publication. For example, if the content can be represented in a single web page or in a PDF document. After all, EPub is based upon XHTML 1.1 and CSS 2, thus allowing for aesthetically-pleasing representation and presentation of most STM-type publications.
On a somewhat related topic, the one major flaw of EPub (actually OPS 2.0 and OPF 2.0 which underlie EPub) is its linearity. Despite token support for auxiliary (out-of-spine) content (content which may be presented outside of the main flow of the text if the reading system supports that – refer to Section 2.4 of the OPF 2.0 specification), the overall focus of EPub is towards linear publications. (Note that all content in an OPS 2.0 Publication has to be linearized for reading systems that choose not to support auxiliary content. That’s why I used the adjective “token” since this optional support effectively buries the powerful advantages auxiliary content can give to the ebook reading experience.)
Thus, even though EPub is based on XHTML 1.1 and CSS, it really cannot represent the “web paradigm.” I consider this not only a major flaw, but one which is very troubling since it will make it difficult to evolve EPub to handle a wider range of publications (like newspapers, magazines, etc. – and of course to be able to contain a standards-conformant “web site.”)
Why? The problem is that supporting the web paradigm and other non-linear organizations of content requires the basic architecture of the reading system to support that, and it is likely most first generation EPub reading systems will not be built looking ahead to that eventuality. Once there is an installed base of such EPub reading systems, it will be very difficult to steer the battleship in a different direction. Thus, EPub will forever be stuck in linear-land. (Imagine if the designers of HTML and the web browser took the same linear tact as EPub has?!)
One inclusion in the EPub-similar OpenReader format, which several of us developed and promoted, was the “web” option in addition to the “linear” option. Unfortunately, we could not achieve the needed critical mass of reading system developers to embrace the OpenReader format, which allowed EPub (and its doggedly linear view of content) to prevail. I now fear for the long-term of EPub as being a universal reflowable publication format because it has chosen the linear path.
We are an Ebook publisher and I would like to assure you that the Epub file system is just another for the publishing industry to attempt to justify its position as experts et al. PDF is the best the easiest and the most prevalent publishing file system there is – Epub is the Big Publishers trying to overtly ‘technologise’ and to control and monopolise the Ebook market.
The solution …
Bring on a PDF reader device from Acrobat and Microsoft joined technologies and end all the monopolising scumbags that are the big presses once and for all.
Keeping E Publishing simple is as important as allowing freedom to publish on the net itself. We the 80′s kids created the net in the 80′s and WE not all those stiff old fogeys who refuse to die with their traditions and whatnot and company handbooks.
Screw the Epub ‘standards’ rubbish … USE PDF don’t allow them to monopolise. Acrobat has done more creativity than anyone cares to even imagine – MOBI and the likes produce NO creative software or otherwise which means they have NO creative people using THEIR non existant software – so I aski you all a question why the F*** should we sell our books on their machines, and whats more struggle to get our files onto their filing systems. B******t to Epub – B******t to Mobi and B******t to Amazon and used and new.
I just hope Microsoft rings up Acrobat anmd says to them lets SCREW them all and they could in SECONDS…!
Regards
http://www.whiffyskunk.com
Is this a troll?
Seriously.
We just finished having a discussion in which majority opinion was overwhelmingly against PDF as a “universal” e-book format. It’s discriminatory against people who read e-books on small screens such as iPhones or PDAs.
If you use PDF, I won’t be reading your book. I have an iPod Touch and PDF just doesn’t work with those.