EBook Creation Process 101

[BetterRed]

Maybe MS didn't like the idea of embracing something largely funded by European taxpayers - without the consent or knowledge of said taxpayers.

Besides which there would have been a hue and cry from Helsinki to Valetta, Dublin to Sofia, about MS being subsidised by the EU Commission European taxpayers - whilst not paying any tax themselves.

Where's the European IBM, MS, FB, Apple, Google, Twitter; all there is ARM and that's from the cousins.

BR

[skreutzer]

Originally Posted by BetterRed

Maybe MS didn't like the idea of embracing something largely funded by European taxpayers - without the consent or knowledge of said taxpayers.

[...]

Where's the European IBM, MS, FB, Apple, Google, Twitter; all there is ARM and that's from the cousins.

Could you please tell me what exactly is funded by European taxpayers regarding the case ODT vs. DOCX? To my knowledge, nothing like this took place. Maybe for the standardization process, but if so, I would regard it as a honorable action. Could you please explain how there should a competitor from any part of the world ever be able to emerge when confronted with the proprietary lock-in effect constructed by said companies? That precisely is the reason why the EU (especially Great Britain) wants the formats to be open, because they want to foster competition. No need to look into the accusation of unfair business practice that comes on top of it.

Just to make very clear, I'm more concerned about the waste of lifetime aspect and not so much about the economical aspects of it, since I'm not affected from the latter any more.

[dickloraine]

First: Styles are the way word is meant to be used. Of course many people don't use them, but that is hardly words fault. In the ribbon the biggest single element is the style selector. No dropdown menu anymore, but you just click on the style you want to use.
Second: The html-export is not that bad. If the document is done correctly. The problems arise if they are not done correctly.
I have no knowledge about developing a word processor or a format. And of course it may well be that microsoft does some things to make it hard for 3rd parties, and it always possible that something can be done better, but word is really complex. And it has many capabilities that go beyond simple text. Citation, formulas etc. Since I lack the knowledge I assume, they had very often good reasons to do what they do. Does odt support all the things word can do? What if you want or have to do things a certain way? Controlling your format has advantages.

[BetterRed]

There's another unspoken truth why MS might not have felt inclined to embrace ODT. OOo is/was essentially reverse engineered MS Office. I used to get the weekly digests from OOo, it was full of discussions along the lines of -- 'it doesn't work like that Word or Excel - fix it'.

So how come its OK for Europeans to 'make use of' foreign IPR, but its not OK for Californians/Kiwis/Aussies/Saffers etc to make Beaujolais, or Parmesan cheese

BR

[skreutzer]

@dickloraine: It's understandable that the format is more or less specific to a certain implementation, but for achieving more format compatibility across implementations, things have to be generalized and agreed upon, instead of inventing a conflicting standard.

@BetterRed: Now you're talking about something entirely different. Reverse engineering of formats is different from implementing features found in other, similar software, and I doubt that it is illegal to have such similarities (except patents, but the EU doesn't allow patents on software). I mean, what are developers supposed to do, invent different paper sizes, new geometric shapes to draw, menues not grouped around topics and random orthography rules? A word processor is comparatively simple when it comes to the functions that are needed, and the more advanced stuff might differ more. Of course, as people got accustomed to the way Microsoft Office is presenting their features, there's now a certain expectation about how a word processor has to look like. On the other hand, OpenOffice/LibreOffice still has manues, while Microsoft Office has a ribbon bar. Apart from the user interface, if one wants to implement serial letters, it will require to provide some fields and a dataset in one way or another.

[dickloraine]

I'm just speculating here, but OO has to be very similar to word (not menus, but how features are implemented) so that it can read word documents and work with them

[skreutzer]

Reading is a matter of format, features are a matter of function and „design“. They're not necessarily related to each other, except that some functions need particular data to be stored in one way or another, some ways more beneficial than others.

But as both Microsoft Word and OpenOffice/LibreOffice are large word processor applications, both can read/write DOCX and ODT. The point is: ODT is simpler and respects other standards. Of course people won't stop to use DOCX for pleasing developers, they'll use whatever is put in front of them. And certainly there's a price to pay for such ignorance. The authors jeffcb works with can't just push a button and a well-constructed EPUB, a print-ready PDF or a valid XHTML file get exported and published to even more sophisticated online publishing systems. Why not? It feels like 1998, but in 2015...

[dickloraine]

That implies you could do that with odt. You can't. Oh wait, toxaris add-in does this for epub. But to be clear: We are speaking of simple documents, like novels.
A print ready pdf? What do you mean by that? You can use word to make a pdf of your text. To speak of big publisher quality: They can't push just a button. You have manual work there. And specialized tools like indesign.

Finally, I understand your political concerns against docx. But that does not mean it is an unusable format or that it is crap.

[skreutzer]

I developed some primitive tools to do that for ODT, and I wish I wouldn't had to, if such would have already been around (note: my workflow differs from writer2xhtml, another alternative for this task, in being based on semantic principles, while writer2xhtml works on the exact representation through direct formatting side of things). I'm glad that I don't have to do it for DOCX, since this would inflict even more problems for no good reason other than...well, you know already. I have to apologize, “print-ready PDF” is somewhat vague, let's not see it as referring to high-quality adjustments regarding colors etc., but a decent print layout out of the box. I additionally just remembered that there are people around selling document templates for print layouts, so one could take those too into account as a „solution“, where there aren't predefined print layouts already available to everyone.

Where manual work is inevitable to produce the required output formats, I wish such has to be done for the right reasons only. I know that people are fine for investing such efforts for the wrong reasons, I also know that this won't change any time soon, so I'm only thinking about how jeffcb could be saved from some of it, especially as he does this for an university initiative, especially as it's about Creative Commons licensed publications (hopefully CC BY-SA). Well, InDesign can do what he needs, Word can be made to do it, too, but choosing between those prisons ultimately won't determine what one can do with it, but what can't be done regardless of best effort.

[Glorfindel]

Well.. I certainly hope that the OP has all the info he needs...

[DaleDe]

Originally Posted by skreutzer

Actually, DOCX is the invention of something new for the sake of doing so, because
ODT came first, is standardized, cleaner, and is also supported by Microsoft Word. Instead of embracing ODT, Microsoft made up DOCX with artificial quirks in it, persuading it to get it accepted as a different standard conflicting with ODT. As the European Union forced Microsoft to make their format more open, DOCX is at least an improvement over DOC, but it seems to be designed to still make it hard to make use of it without Microsoft Word.

Regarding the only software that can fully deal with DOCX, in my opinion, it would be of great help if there would be some kind of “book mode”, which would disable direct formatting and present styles as the main mechanism for markup. Per default, Word's GUI is configured for composing office documents, which is unfortunate for writing long form texts. And, let's not forget about the fact, that Microsoft Word is still in the year of the Lord 2015 unable to export well-formed XHTML (not even talking about valid output), so that people like Toxaris, Hitch and now jeffcb too have to invest valuable lifetime for compensating the lack of cooperation on the side of Microsoft.

I don't object to Microsoft per se, I object to what they do and don't.

Atlantis can read DOCX and can create an ePub directly. It does not require styles (it will use them) but builds CSS entries on the fly by comparing paragraphs and collecting formatting so even if you didn't use styles it will create one CSS entry for all of the paragraphs that are formatted the same. The ePub that is created will conform to XHTML and pass epubcheck.

Dale

[hidden.platypus]

First of all I want to say Thank you to @hitch for directing me to here:
https://wiki.mobileread.com/wiki/EPub_Tutorial

I (erroneously believed that) have been serious about editing my ebooks for awhile.

I'm sort of crazy, as I suppose are some of the posters here. I prefer to muck about with the code, as opposed to with an editor. No offense to the guy who created/maintains Sigil, (and not just because I know he's likely to be around here somewhere) but WYSIWYG editors like that drive me crazy. It won't do what I ask it to do. Simple things like change the fonts. Part of it is my fault. I just don't know enough about epub, and I haven't read the Sigil manual.

But from a user perspective (admittedly crazy-person user) Sigil should be intuitive enough to tell me how to change the fonts even if that means it needs to tell me that fonts don't matter in epub unless you embed them, and then show me how to do that if that's what I want to do.

I spent some time using the Google last night, searching for css and epub and this forum did not come up.

Desperation drove me to it this morning. I was going to ask BetterRed for help, and I like to work before asking for help so I trolled around the forums and as luck would have it... here we are.

I'm not an advanced user. I'm the guy everyone else in the office thinks is advanced because I know a tiny bit about coding, but I know enough to know that I know nothing AND I've met BetterRed so I know what a guy who's advanced would know, and I know I don't know an eighth of what that guy knows. I'm the kind of guy who knows what Regex is but not how to use it. Not advanced, but not a beginner either.

I know html, I know Word (about the only thing it seems that I knew before BetterRed is Shift+F7 and I figure that's because people used to steal my mouse when I was first starting to use the computer so I memorized every Word shortcut, and they haven't really changed since DOS). I don't understand why people prefer to tab all over the place rather than indent. And I don't understand how it's easier for other folk to change all the formatting individually, rather than just create a style but I DO know they do that.

But I'm kind of a bull in a china shop. The little I know is dangerous to me, and my ebooks. I don't know where to start since I've been out here off the reservation for so long. I don't know what is good conduct, from what is appropriate, much less best practices.

So how does an intermediate user get to the point where he can produce decent ebooks? I tried indesign, but that wasn't for me. I'm not graphically inclined.

I use Word 2013 and convert the docx into epub via calibre.

That works for some books, but others do strange things. Case in point:

I did a find and replace in a Word doc replacing periods with one space after them, with periods with two spaces after them. From ". " to ". "

Now after conversion Calibre gives me this code at the end of almost every sentence:

<span class="calibre6">.  </span>

In Aldiko, that turns a sentence that should read:

Spoiler:

Mobileread forums are the best place to learn ebooks, and get help if you need it.

into:

Spoiler:

Mobileread forums are the best place to learn ebooks, and get help if you need it .

There is whitespace before the period now because of the span tag.

Next problem. Moon+ Reader shows numbered lists as bulleted lists. Does epub not like numbered lists from Word?

Aldiko shows them as numbered lists but all the numbers are 1. So it's like a bulleted list with 1. as the bullet.

Where does a guy like me learn to fix his errors and clean up after himself?

I'm going through Pablo's tutorial. Thank you again for that suggestion, and the other info in this thread even though much of it went over my head.

Thank you all also for any help in advance.

[Toxaris]

In all fairness, the WYSIWYG modus of Sigil is not used that much by the 'hardcore' members. Almost all of the 'serious' e-bookmakers work in codeview. Using the bookview will get you into issues sooner than later.

That being said, the only way to progress is by doing it and studying how other books have been made.

I will not say anything about converting docx to ePUB via Calibre, the changes are big that I say something that isn't true or taken the wrong way. I don't use it anyway. I use my add-in to create an ePUB directly from Word.

Lists in ePUB can be a hit and a miss. Unordered lists should be fine, ordered lists are a different thing. Not everything will then work as thought. Now, I do not know how Calibre converts the list of a Word document to HTML, but I do know that it can be difficult to make the distinction from a docx document. So, it could a conversion issue.

[hidden.platypus]

So this is pretty cool.

I'm actually using your tool on my ebook right now.

Thanks for responding.

So this might be sort of circular, but where can I find a beginners guide to advanced formatting?

[dickloraine]

I don't know why you changed the dots to have two spaces after them, but I guess word or you somehow gave the replaced dot a different style, which calibre correctly assigns. If you use the style inspector, you can see if the guess is right.