mobi without ePub ...

[Sella174]

No, not a clue. Why would the code be a mess?

[JSWolf]

Originally Posted by Sella174

No, not a clue. Why would the code be a mess?

Because chances are they didn't properly use styles and used the toolbar, tabs, returns, fonts, etc.

[Sella174]

Which is why I import it into LibreOffice: so that I can clean it up and reformat it the way I want it. Then when I export it back to whatever version of MSWord the author is using, my styles are retained in the document that they will be editing, as well as all my notes.

[JSWolf]

Originally Posted by Sella174

Which is why I import it into LibreOffice: so that I can clean it up and reformat it the way I want it. Then when I export it back to whatever version of MSWord the author is using, my styles are retained in the document that they will be editing, as well as all my notes.

Sometimes it's actually easier to clean up the mess after it's converted to ePub because a lot of it can be fixed with regex and/or search/replace

[Sella174]

LibreOffice understands regular expressions just fine.

[JSWolf]

Originally Posted by Sella174

LibreOffice understands regular expressions just fine.

But you don't see the ePub coding. That's what you need to see in order to fix it.

[Sella174]

Why do I need to see the ePub code?

Once the odt has been cleaned and properly styled, one of the things that gets done to it, is it gets exported to HTML. This HTML file is cleaned (by a little script that heavily uses regular expressions), eye-balled and uploaded to the website. Thus I already have super-clean HTML in my possession.

Please understand that the authors do not understand HTML, the ePub format or for the most part anything beyond the basics of their word processing program. Any revision of the text is done by them using their word processor. So editing the HTML or an ePub as a source document is seriously out.

The readers are asking for mobi files, because they have Kindles. I was therefor looking for a way to generate the mobi from what I already had, namely HTML, odt or PDF (derived from the odt). But it seems I need to package the HTML into an ePub before I can create a mobi. Silly, but an expected reality in this age of lazy, point-&-click computing.

So I'm writing a program that will package the HTML in a valid ePub, along with added metadata, without trying to be intelligent and messing up my neat HTML files.

[JSWolf]

Because sometimes it's easier to clean up the mess when it's an ePub than if it's a word processing document.

[JSWolf]

Given that they have Kindles, you'd be best off having Mobi and KF8. Newer Kindles that can run firmware 5.8.11 will benefit from KF8.

[Sella174]

Originally Posted by JSWolf

Given that they have Kindles, you'd be best off having Mobi and KF8. Newer Kindles that can run firmware 5.8.11 will benefit from KF8.

Obviously.

And that is why how to get to the mobi and KF8 without first making a redundant ePub was the original question.

[DaleDe]

Originally Posted by JSWolf

ePub 2 came from MS Reader format.

Actually it came from Open eBook as shown in our wiki. OEB was actually used as the source format for LIT (MS Reader), IMP and MOBI.

Dale

[JSWolf]

Originally Posted by DaleDe

Actually it came from Open eBook as shown in our wiki. OEB was actually used as the source format for LIT (MS Reader), IMP and MOBI.

Dale

But, ePub came from MS Reader's version.

[Hitch]

Originally Posted by Sella174

Obviously.

And that is why how to get to the mobi and KF8 without first making a redundant ePub was the original question.

Which is why I said, in my post: just build an OPF and an NCX, for your superclean HTML. Then drop the OPF onto Kindlegen or KindlePreviewer, and it will build a dual-MOBI file. Voila, no ePUB and you have what you need.

This has nothing to do with Gutenberg. It's simply what happened, when various groups had differing opinions about the "best" format, when eBooks were uber-young. Just as Betamax and VHS were the battling formats, in the early days of video. {shrug}.

I will say this--I've used LO and OO and Word and Pages and all kinds of word processors, over the last decades. If OO or LO works for you, great, but the cruft output by any of them is pretty much the same. What needs cleaning is always pretty much the same, if, like most of us, you have utterly untrained (and untrainable, typically) authors. I can clean Word the fastest, of all the word-processing solutions. Just offered FWIW.

So: just build an OPF and an NCX. From what you've said, that's obviously well within the scope of your abilities. Place it in the correct relative position, with your HTML files directory, and drop that OPF on Kindlegen or Kindle Previewer. You'll have a dual mobi without any problems--and without an ePUB.

Good luck to you. I don't disagree with you, btw, about the era of lazy point and click computing, but after working with thousands of authors, in the last decade--of which we've received, MAYBE, 20 files in which the author actually used Styles or Headings--which obviously translate pretty perfectly to HTML and thus to ePUB/MOBI--I don't believe that you can train them. I used to think I could--but the truth is, you can't. Right-brainers in particular are resistant to learning things like Heading and Styles, even though it would be to their own massive benefit to use them. Even our repeat clients--nope. (n.b.: most of ours can't even use Track Changes right. So...)

So, I understand--somewhat--why you've taken the path you have, in importing their content to OO/LO, cleaning it up, etc. I'd do it in Word, but that's all a matter of preference. Don't ignore what I said about building the OPF, though. If you want to go direct-to-mobi, that's probably your best path. (Although, again, to me, they're much of a sameness. Yeah, verily, there are a few differences in the formats, but....{shrug}, not enough to really think about. BTW: you do know that your authors could send ePUB format to their Kindle email addresses, and get a formatted MOBI in their personal docs section of their Kindle, right?)

Hitch

[Sella174]

Originally Posted by Hitch

Which is why I said, in my post: just build an OPF and an NCX, for your superclean HTML. Then drop the OPF onto Kindlegen or KindlePreviewer, and it will build a dual-MOBI file. Voila, no ePUB and you have what you need.

Kind of figured that as well, reading the Kindle Publishing Guidelines.

Originally Posted by Hitch

I will say this--I've used LO and OO and Word and Pages and all kinds of word processors, over the last decades. If OO or LO works for you, great, but the cruft output by any of them is pretty much the same. What needs cleaning is always pretty much the same, if, like most of us, you have utterly untrained (and untrainable, typically) authors. I can clean Word the fastest, of all the word-processing solutions. Just offered FWIW.

Oh, how I miss that feature of WordPerfect to view the source!

I haven't looked at MSWord since I was no longer required to provide user support for it, but LibreOffice has really worked on their regular expression engine the past few updates. Not perfect, but pretty good. I also like the feature to search for applied styles, fonts, etc.

Originally Posted by Hitch

So: just build an OPF and an NCX. From what you've said, that's obviously well within the scope of your abilities. Place it in the correct relative position, with your HTML files directory, and drop that OPF on Kindlegen or Kindle Previewer. You'll have a dual mobi without any problems--and without an ePUB.

Certainly will give this a bash, instead of a whole ePub.

Originally Posted by Hitch

BTW: you do know that your authors could send ePUB format to their Kindle email addresses, and get a formatted MOBI in their personal docs section of their Kindle, right?)

Whilst I do, for (most of) them this is akin to designing a returnable Mars rover.

[radius]

Originally Posted by Sella174

Certainly will give this a bash, instead of a whole ePub.

I just wanted to tell you again that the difference between this and making an epub to convert to mobi is negligible. If you are scripting it, then maybe five minutes of time up front and then you have one more format to use forever after. Maybe nobody is clamoring for it right now, but it comes at almost no cost to you.