I'm a big fan of EPUB because it's a pretty neat example of the Unix philosophy "worse is better".
A proprietary or custom format might have been more suited to creating e-books (drop-caps, for example, would probably have been easier), but using HTML and CSS as the basis means it's open, easily understandable, and pretty simple to make corrections to e-books or even begin making your own.
The fact that a teenager with a little bit of interest and a small amount of knowledge can play with an e-book, then view the results on real devices, is hugely appealing. I realise that they could also be making a website, or doing `real' programming, but to some people the ability to correct typoes in _Ender's Game_ or reformat _Dune_ to their preferred layout is just so much more appealing and empowering. When I started playing making websites at age 14, I'd have killed to be able to use those skills to make an e-book.
In spanish there are communities that take paper books, digitalize them and release them in epub format. Books that probably will never see an official digital release, which surprisingly, is the fate of many classic books (it's improving but the spanish digital book market isn't as vibrant as the one Amazon created for english titles). The best community I've found has a huge manual that in essence works as an editorial blueprint of how your epub must be constructed and formatted in order for them to accept your digitalization of a book and release them under their label, which in effect makes this site a clandestine digital publisher of sorts. After you submit a title they have editors that review it, etc. and all of this is done by volunteers online. It's just awesome. There's no need to debate the legality of this site of course but if I was 14 today, with the free time I had back them, I'd definitely be geeking the hell out of that manual and digitalizing some of my many paper books. If there is a heaven, a special place must be reserved for all those guys that make that community work, if you ask me. Anyways, I haven't found similar communities in english although I assume there must be some.
I'm sure there are some Spanish-speaking people that would appreciate knowing the URL to this hub, and a few more who wouldn't mind being aware of its existence so they can reference it, etc.
EPUB is, without any doubts, my preferred ebook format. I knew it was an open format standard, but I have never taken a look at what's under its hood. After I took a look at this presentation, I'm going to dedicate this evening to exploring the format, its advantages and its disadvantages.
Unfortunately, I am not able to find any reasonably good looking Linux client for it, so I usually use ebook-convert command that comes with Calibre to convert my books from PDF to EPUB and vice versa.
If you happen to use - or plan on using - Owncloud, you might want to have a look at a few apps I made just for that purpose:
Reader [1] is an epub reader based on the Futurepress epub.js renderer. It looks close to 'native', especially when used full-screen.
OPDS catalog [2] is, as the name implies, an OPDS catalog which you can use to access publications (not limited to epub) hosted on an Owncloud instance. It allows you to access your library from any OPDS compliant client, examples of which are FBReader (Android), CoolReader (Android), Moon+ Reader (Android), Aldiko (Android), KyBook (iOS) and Marvin (iOS).
The controls are nonstandard. I haven't used it in a bit, but advancing the text with spacebar would be just peachy.
Default formatting. It's not bad, but the most annoying, especially on today's widescreen 16:9 format screens, is that while margins can be set, text width cannot. I find 35 - 45 em is about perfect for text, yet cannot set this for ePub.
Sigh.
I actually usually use Calibre or Pandoc to create HTML, then write my own CSS and serve those locally.
BookFusion can be used instead of Calibre to not only manage and organize your eBook library but also to:
1. Read ePub, PDF and other major format ebooks
2. Send your eBooks to your kindle device if you have one
3. Easily share your eBooks with friends and family the way you would a physical book.
My frustrations with poor publication curation is ... a major pain point.
I would love:
1. A tool that would present HTML content in some standardised, simple, format. E.g., Readability, Instapaper, Pocket, Firefox Reader Mode.
2. Include metadata. Perhaps a somewhat IMDB-type facility for sources which don't include good metadata. ISBN and DOI work well for traditional printed book and scientific article works, but other online content is ... less well structured.
3. Tags. I've detailed a possibly workable general tagging ontology at Ello (https://ello.co/dredmorbius/post/FTpX7LWNFjtOpYGu5xRkHg). That plus user-assignable tags, and the ability to subset/select by multiple tags would be ... nice.
4. Bibliography generation.
I've looked at, kinda, Calibre, Zotero, Mendelay (rejected on account of Elsevier ownership), and a few other tools.
Sharing between desktop (Linux) and Android other than via Kindle would be tres, tres cool.
Risking get fire for self-adv, but some time ago I made a very simple static HTML-based ePub reader (does not even support contents or pictures), but with some sync over Google Drive. I used it mostly because I have a wide diversity of platforms in my possession, and want to sync between them. Here's the link: https://h5reader.azurewebsites.net/#
Practically speaking it matters not. Calibre can do pretty lossless (practically speaking) conversions between formats. The limitations/features of ebook readers have a much greater influence on how a book looks than the format.
I love the status-quo where Amazon pretends that their DRM is not hopelessly compromised and I get to strip the DRM from the vast quantities of ebooks I purchase there. I would much rather they keep turning this blind eye than adopt epub behind a more effective DRM scheme. That would be a huge loss to me.
When my e-ink kindle can support high-resolution color images, smooth and fast LCD-like scrolling etc. then I might care about epub features, but while I don't want or expect more from my kindle than a paperback, I am quite happy.
Because money. I doubt there's any really good reason besides the fact that they don't want you to just be able to copy books directly from your Kindle and read it elsewhere.
I use a Kindle and it makes little to no difference at all. If you need you can edit a book in epub format and calibre converts it to mobi seamlessly in little to no time.
I dabbled with converting a Japanese text to an ebook and adding furigana to it. However, I couldn't find a free tool that supported generating 3.0 epubs, so I thought, it's mostly HTML, how hard can it be to just do it manually.
Very hard, I must say. The are all sorts of meta files that doesn't seem to ad anything to this format, but are somehow required.
No. I could generate / write the html, but I couldn't for the life of me figure out the xml (I think there's another type of files required too) meta data. I looked at a few books and tutorials and it just looks like gibberish.
I have a dream ... that some day there will be a standard structured document format for publishing. And ePub is very, very, very nearly it. May well be, I've only been kinda keeping tabs. And yes, I'm a huge fan.
My one biggest frustration (noted earlier here) is that the Linux ePub client, fbreader, doesn't have a way to specify a maximum content line length. Margins, sure, but not content. Result is that reading full-screen on a 16:9 display is ... kinda pessimal. Small issue, but one of those "pay attention to details" items that matters.
From the preso (nice layout there, by the way), Metadata: I'm in love. This is something that I'm familiar with via hNews and Readability. Structured information in documents would be really, really peachy.
I've got a long-standing rant on the browser world, which is entirely flipping falling down as a content presentation tool, for numerous reasons. Among the reasons for tab proliferation, I'm realising, is really piss-poor content management tools. As I'm in the midst of a large research project, with some 1800 articles curated and another thousand-plus books and documents (mostly PDF, but a mix of other styles), being able to reliably extract title, author, publisher, date, etc., information would be quite valuable.
1. A footnote / endnote / sidenote native capability. I've mucked around creating a sidenote implementation, but as a native element it'd be useful.
2. A native hierarchical threaded discussion structure. The fact that every last online Web forum (HN included) is re-implementing tin / mutt (and doing it wrong) is ... depressing.
3. A hell of a lot more thought to accessibility. From font/colour/size choices to support for text-to-speech and speech-control systems, I'm developing a growing awareness through the community I'm currently supporting. ePub with its simplified presentation may be a win here.
4. Equations. Not always needed, but support for something at least as powerful as LaTeX would be quite useful.
5. Annotations. A lightweight mode of interacting with content would also be helpful.
A proprietary or custom format might have been more suited to creating e-books (drop-caps, for example, would probably have been easier), but using HTML and CSS as the basis means it's open, easily understandable, and pretty simple to make corrections to e-books or even begin making your own.
The fact that a teenager with a little bit of interest and a small amount of knowledge can play with an e-book, then view the results on real devices, is hugely appealing. I realise that they could also be making a website, or doing `real' programming, but to some people the ability to correct typoes in _Ender's Game_ or reformat _Dune_ to their preferred layout is just so much more appealing and empowering. When I started playing making websites at age 14, I'd have killed to be able to use those skills to make an e-book.