Ask an e-book creator...(how do they do it)?

[nrapallo]

Ever wonder what to do so as to create an e-book for mass consumption; or even for your own use?

Just ask an e-book creator (see I present to you....the e-book uploaders of Mobileread.com for who they are)!

Need quick & dirty advice from an e-book creator or even a walk through of their work flow, well then, that is the reason for this thread; a psuedo-FAQ for e-book creation.

Pertinent Mobileread threads:
- Why do you upload books? Why don't you? by mtravellerh
- How to create "proper" Mobi books! by HarryT
- Mobipocket Creator and "The Mystery of the Missing Link" by mtravellerh
- Book Designer Hints and Tips by Patricia
- Book Designer Hints and Tips by crutledge
- Creating multiple ebook formats from same source files! by nrapallo
- What tool set do you use? by crutledge


So, what's on your mind...?

Questions that could be asked (and answered by e-book uploaders):
- Where do you find content (text / html / pdf scans)?
- What technical knowledge would be helpful i.e. (X)HTML, CSS, Python, Perl)?
- What software tools do you prefer to prepare your e-books?
- What are the final steps to create e-book in a specific format and/or muliple formats?
- How do you prepare you posting of that e-book here?

[nrapallo]

Originally Posted by nrapallo

Questions that could be asked (and answered by e-book uploaders):
Where do you find content (text / html / pdf scans)?

I work mainly with .html from websites or Project Gutenberg (the former usually copyrighted and for my own personal use; the latter, of course, is in the Public Domain). The biggest hurdle is adapting someone else's HTML code and CSS to work properly on small screen (6") ebook readers. Also, properly building the "index" is crucial to the ease of navigation within the ebook. A TOC (Table of Contents) is a must!

After many previews or early builds, I weed out (fix) what works and what doesn't; mostly via experimentation. Then, I may add overriding CSS like minimal left/right margins and full text justification, but that's about it. I like to preserve the original's look-and-feel as long as it's not anti-ebook!

What technical knowledge would be helpful i.e. (X)HTML, CSS, Python, Perl)?

For me, a working knowledge of HTML and CSS has been most helpful. I often check the www.w3schools.com website for tutorials and syntax.

What software tools do you prefer to prepare your e-books?

I use a powerful text editior, TextPad for my HTML/CSS editing, as well as Mobipocket Creator (for .mobi/.prc), eBook Publisher (for .imp) and calibre (for .lrf/.epub/.lit/ereader .pdb).

What are the final steps to create e-book in a specific format and/or muliple formats?

I use the GUI eBook Publisher and GUI Mobipocket Creator, but the command line 'ebook-convert' from calibre.

How do you prepare you posting of that e-book here?

For multiple e-book postings, I just type the first post and then copy that text for any other e-book format I upload. It's tedious, but also the most rewarding in that more users get to use that e-book!

[zelda_pinwheel]

excellent idea, nick. thread stuck.

[naguz]

I have a question it seems to be difficult to find an answer to with my ample googling skills at least: I have a book, like in a bundle of pages made out of dead trees with letters printed on them. What is the best way to make an ebook (epub naturally) out of this?

I will be using a SkyPix TSN410 (a handheld scanner that saves the images to an sd card) as it was a cheap and seemingly good alternative for book scanning. Also, it's main purpose, beside ebook making hopefully, will be to scan a few pages here and there on the campus library. I believe contrast ratios etc. will be quite OK. (Not received it yet).

But apart from the scanning, I would, naturally, want the job to be as easy as possible, without compromising on formatting and quality.

Especially the formatting part seems difficult. The free OCR alternatives out there seems to mainly rip the text out of there without worrying abut formatting. It'd be nice to have as much of the formatting as possible intact, as long as it doesn't decrease the quality of the epub.

Most of what I find, however, concerns already OCR'd text, and how to format this properly. I'd like to keep as much of the footnotes, chapter headings, etc. as original as possible.

Is there any OCR apps out there that are good at this? I'd guess it narrows the alternatives down a bit by the fact that it should preferably run under linux and be free... But as the .jpgs from the scanner is portable, I *could* (but would rather not have to) reboot my desktop pc into Win.

Apart from that, if anyone would like to point to a begnners guide they think are epsecially good, concerning also the process after OCR, that is of course most welome.

From what I have read, people seem to recommend a free web-service which allows you to convert only 30 pages a day if you want to keep formatting from image files. But I hope someone here has a better idea.

[DaleDe]

Please check the wiki for tips on creating your own eBook, OCR and similar topics.

[FunkeXMix]

Hi,

I would really appreciate if you could give me an estimate for how long it takes to convert a PDF to a fully functioning Epub file?

[DaleDe]

Originally Posted by FunkeXMix

Hi,

I would really appreciate if you could give me an estimate for how long it takes to convert a PDF to a fully functioning Epub file?

Anywhere from 5 minutes to months.

[Elfwreck]

Originally Posted by FunkeXMix

Hi,

I would really appreciate if you could give me an estimate for how long it takes to convert a PDF to a fully functioning Epub file?

Like DaleDe said. It depends on how the PDF was made, and what you're using to convert it.

PDF is less a "filetype" and more a wrapper for content that can be viewed on many different machines. It was designed to be printable--everything else came later, and support for "convert *away* from PDF" is erratic. If the PDF was made from an XHTML file with conversion in mind, switching to something else is relatively easy. If the PDF was made from an older version of Word, then every line break may be a hard return, and take extensive time to convert to useful ebook formats. If the PDF is scanned images, they'll need to be OCR'd & corrected first--and those images might look fine on the screen but still be low-resolution enough to cause swarms of OCR errors.

For a lot of commercially-released PDFs, usually exported from InDesign or Word, conversion to epub can be as little as an hour or two--if you're comfortable with all the steps, and willing to accept a few formatting errors (like extra paragraph or page breaks). If not, then you need a page-by-page check, possibly a line-by-line check, and how long that takes depends on your reading speed and familiarity with the program(s) you're using.

[DaleDe]

And to make matters worse PDF can be edited with Acrobat so it may not even be linear with edits linked elsewhere. Words can be in different places in the text, items can be marked deleted but still there. It can be as bad as an assortment of letters with on page coordinates defining their location. It can also be just a bunch of images that look like pages or it can also be a mix.

[GeorgeHB]

Does anyone know how to create a mobi file from an InDesign book (.indb) file? I've tried making a PDF from it, but when I put it through Calibre it jumbles the TOC and the pages run together.

[DaleDe]

? Use the export to mobi option available from the tool

[VandaSue]

Hi,

My name is VandaSue. I have finally jumped most of the hurdles of my first ebook; solving two of the three issues of the upload/validation process. Now, I receive this error:

ERROR ITMS-9000: "OEBPS/content.opf(17): element "metadata" incomplete; missing required element "dc:language"" at Book (MZItmspBookPac

Of course, this is incomplete error message, but all that I can copy right now. I'm hoping someone can give me an idea of what is going on?? The thought that came to me is that this is a 32 page book of Illustrator files; embedded into ID. (btw, congrats for whomever but in the spell check here) So, there is no text! Not a single line of text. It is all images, or as stated AI files. Is this the issue? Producer cannot find any English, thus stumbling to find some? Perhaps I can put a small period somewhere on my first page?? That might satisfy? Or is that the problem?

Any help would be appreciated. Thanks,

VandaSue

[DaleDe]

It is complaining about an entry in the content.opf file, not any of your data files. ePub requires that you declare the language of the ePub file and this is missing in your file. Add that entry and it should be ok. If there is no language in the file then declare the language of the title.

Dale

[Jellby]

Originally Posted by DaleDe

If there is no language in the file then declare the language of the title.

Or use one of these:

und (for undetermined) is used in situations in which a language or languages must be indicated but the language cannot be identified.
zxx is listed in the code list as "no linguistic content"

[VandaSue]

Thanks for the earlier answers. OK. New to this. I have not used code in years. However, where do I find the code that is created by InDesign when it is being validated by the iTunes Producer? I get my error message and I would like to correct it. However, I cannot find the code, other than the code that is visible in the box provided by iTunes in the validation box with the error message.

I have seen others suggest that I create my own files folders, such as: meta-inf, and contents. However, I believed that using the export in ID was going to do this when it created the ePub and when I uploaded into iTunes? Pipedream here??

I simply need to correct the language in my code, or so my error message states. Hopefully, I have managed to get by with the other messages that first appeared and correct them. I would like to upload my book by tomorrow.

Sorry for my lack of knowledge here. I will be reading up on code, you can believe me. I have several other books to place here, so have to do this on my own. Tired of publishers.

Thanks,

VandaSue