Mobileread
Workflow for simultaneous EPUB and PDF production?
#1  Monaghan 08-15-2019, 07:53 AM
Thanks for taking the time to read this post.

I have a client for whom I am currently designing documents in EPUB format. These are relatively simple, in terms of design and CSS, and we are aiming to keep the workflow short in order to accommodate updates to the documents down the line.

They have asked whether it would be possible to have PDF versions of the documents produced at the same time, from the same source document.

My initial thought for the workflow was to import the original files into InDesign, from which I would output PDFs using InDesign styles using the same names as the existing EPUB CSS styles. Then I would export an EPUB from InDesign and apply the current stylesheet.

Is there a simpler method I haven't thought of? Or a piece of software that would do the job? Unfortunately, Vellum does not have the design flexibility for what we need to do. Software on PC, Mac or Linux is fine.

Ideally this workflow would be simple enough that the client could update the book once it's set up. This is asking a lot, I know!
Reply 

#2  BetterRed 08-15-2019, 08:17 AM
Try opening one of the EPUBs in the calibre viewer and hitting the print button - you can set page and margin sizes, and page numbering - not fancy, but surprisingly effective.

BR
Reply 

#3  Monaghan 08-15-2019, 08:36 AM
That is surprisingly effective, thank you. Do you happen to know if there is a way to add custom page sizes and not include the external TOC at the end?
Reply 

#4  BetterRed 08-15-2019, 09:38 AM
Quote Monaghan
That is surprisingly effective, thank you. Do you happen to know if there is a way to add custom page sizes and not include the external TOC at the end?
'Fraid not - ask in the main calibre forum, someone's bound to know.

But a quick glance suggests you might be able to use calibre's Conversion - PDF Output feature looks like it has what you want

show attachment »

The Viewer Print would be a fixed version of that

BR
Reply 

#5  Doitsu 08-15-2019, 12:45 PM
Quote Monaghan
Is there a simpler method I haven't thought of? Or a piece of software that would do the job?
Have a look at Prince PDF and the Calibre Prince PDF plugin.
Reply 

#6  Tex2002ans 08-15-2019, 06:04 PM
What you'll want to do is pick one "source format" or "master document" which you derive all others. Which format that is, that's up to you.

I prefer an EPUB-first (or HTML-first) workflow.

Others prefer InDesign-first, and others prefer DOCX-first.

You want the conversion to happen as late as possible, or else the amount of labor needed will multiply dramatically.

Quote Monaghan
My initial thought for the workflow was to import the original files into InDesign, from which I would output PDFs using InDesign styles using the same names as the existing EPUB CSS styles. Then I would export an EPUB from InDesign and apply the current stylesheet.
It could work, but the InDesign EPUB cleanup would still require manual intervention.

(InDesign usually outputs a lot of iBooks code that doesn't work well on other readers. Or a lot of Print-specific cruft gets carried over.)

Quote Monaghan
[...] we are aiming to keep the workflow short in order to accommodate updates to the documents down the line.
This throws a huge wrench in your workflow, because it limits which source format you choose to what the AUTHOR knows.

What's the format of the source document? DOCX?

If you choose InDesign as the source, the author will have no idea how to do anything within that program.

If you choose DOCX, you have to make sure the author knows how to properly use and apply Styles.

If you choose HTML/EPUB, the author will have to consistently apply proper/clean HTML+CSS.

Quote Monaghan
Ideally this workflow would be simple enough that the client could update the book once it's set up. This is asking a lot, I know!
... each conversion is going to require some amount of manual intervention.

You'll want to take care of as much as possible at the source document level, and split into different formats as late as possible.

Example, if you go from InDesign->EPUB:

0. You "finalize" the Print design.

1. Export to EPUB. Complete all your manual adjustments. EPUB "finalized".

(This is what I call the "bifurcation". You're going from 1 type of document to 2.)

2. At some later date, the author finds 25 corrections.

You correct 25 typos in the source file (InDesign).

3. Do you manually duplicate the same 25 fixes in EPUB?

(Along with all that entails: making sure the corrections were the same, don't introduce new typos, retesting, etc.)

Or do you re-export from InDesign, and start your EPUB-specific tweaks all over again? (Back to Step 1.)

The more extensive the corrections are, the more likely it's "easier" to just start the EPUB from scratch again.

And you can see, it's not "just 25 typos" we're correcting now, it's 25 * # of formats.

* * *

DOCX->EPUB, DOCX->InDesign, EPUB->InDesign, [...] each one is going to bring along its own unique challenges.

Trying to allow for author-tweaks beyond this bifurcation point... Each new format will multiply the amount of work needed to update.

Quote Monaghan
They have asked whether it would be possible to have PDF versions of the documents produced at the same time, from the same source document.
Depends on the extent of the changes you're expecting the author to make.

Print comes along with plenty of its own challenges, and the smallest of changes (like adding a word, deleting a footnote) can cascade into very large reflows of text.

If you rely on more complicated things like referencing Page Numbers, Figures/Charts (floating), Lists, [...], this also throws huge wrenches in each format.

Quote Monaghan
That is surprisingly effective, thank you. Do you happen to know if there is a way to add custom page sizes and not include the external TOC at the end?
What's the intended use of the PDF?

Is it intended to be an actual Print book? If so, Calibre's PDF output is... abysmal.
Reply 

#7  Hitch 08-18-2019, 02:33 AM
Quote Tex2002ans
What you'll want to do is pick one "source format" or "master document" which you derive all others. Which format that is, that's up to you.

I prefer an EPUB-first (or HTML-first) workflow.

Others prefer InDesign-first, and others prefer DOCX-first.

You want the conversion to happen as late as possible, or else the amount of labor needed will multiply dramatically.
AMEN, brother!


Quote
... each conversion is going to require some amount of manual intervention.

...snippage....

Trying to allow for author-tweaks beyond this bifurcation point... Each new format will multiply the amount of work needed to update.

....more snippage....

Depends on the extent of the changes you're expecting the author to make.

Print comes along with plenty of its own challenges, and the smallest of changes (like adding a word, deleting a footnote) can cascade into very large reflows of text.

If you rely on more complicated things like referencing Page Numbers, Figures/Charts (floating), Lists, [...], this also throws huge wrenches in each format.

What's the intended use of the PDF?

Is it intended to be an actual Print book? If so, Calibre's PDF output is... abysmal.
I want to say that Tex is my resident process expert--what he doesn't know about it ain't worth knowing. Now, we argue about a few things--for books that will have both print and ebook components, I prefer INDD-->ePUB, but much of this is personal preference.

As he says, if you're thinking of using this for print, run like hell from the Calibre->PDF option. It's fantastic for hadning a client a pdf that they can read for whatever reason, or something that they can send to the Last Book Reviewer on Earth That Won't Accept ePUBs/MOBIs, but it's absolutely not a print layout option.

Lastly, I'm sorry, but I see absolutely NO possibility of you giving the client an ePUB and a PDF that they can update. How would they do that? Editing a PDF is always--always--a lousy outcome, other than letter-for-letter replacements (e.g., "their" for "there" or the like). Sure, they can learn--possibly--to open an ePUB in Sigil or Calibre's editor and make minor text tweaks, but what happens when they can't deal with code view? What happens when they decide to "just" edit a table or whatever, and they corrupt the markup/CSS?

And as I said, editing a PDF? That's no bueno for anyone. Believe me, I've spent a lot of time trying to figure out, lately, if PDFs are REMOTELY editable in any realistic way and the answer is--they ain't.

This is typical client pipe-dream nonsense--they think that everything is a Word file. If they REALLY need editable files, then you give them a Word file--that's what they have. They upload it at the KDP, they upload it at B&N, and they export it to PDF. That's the only realistic, viable path.

Otherwise, educate them and move on down the road. Tex and I have done many thousands of eBooks and hundreds of print and there's simply no "good" way to do this with a client that wants to be able to update the files themselves.

Hitch
Reply 

#8  Tex2002ans 08-18-2019, 03:23 AM
Quote Hitch
I want to say that Tex is my resident process expert--what he doesn't know about it ain't worth knowing.
Also depends on how technical your team is, their current skillsets (do they all know Word or InDesign? Do they even know HTML/CSS?), and how much you can automate beforehand.

For example, I also convert my EPUBs into HTML articles to insert into websites (or specific CMSes).

Side Note: So, I automated a lot of Regex to convert clean EPUB/HTML -> HTML-for-CMS. I can now "one button push" convert my EPUB into something I can copy/paste into the CMS I work with.

Depending on your output format, things like Footnotes or Poetry might have to be handled quite differently, and you can't rely on Ebook-specific CSS. Take the CMS example, you have zero control over the CSS used on the website.

Also, the further you go down the rabbit hole of ONE format's specifics (example: EPUB3 Footnotes), the more you're going to have to work to convert the code into a more "format-neutral form".

There isn't no way in hell the CMS is going to handle EPUB3's:

Code
<a href="#fn1" epub:type="noteref" >1</a>
<aside id="fn1" epub:type="footnote">
It'll probably have to be converted to basic links.

Maths??? FORGET ABOUT IT!

Quote Hitch
Now, we argue about a few things--for books that will have both print and ebook components, I prefer INDD-->ePUB, but much of this is personal preference.
You're wrong.

Quote Hitch
Otherwise, educate them and move on down the road. Tex and I have done many thousands of eBooks and hundreds of print and there's simply no "good" way to do this with a client that wants to be able to update the files themselves.


If you're working B2B, consistently working with the same publisher across hundreds/thousands of documents (Journal articles, or consistently Styled books), then perhaps you can talk... but it would require a lot of customized workflows/backends/tools.

And you're always going to come across these insane edge cases (2 columns with 1-column abstract up top? Top/Bottom floats? I want my figures to float into the margins? My book requires text to shrink like Alice in Wonderland?).

Throwing an author updating things into the mix? Yeesh... that's starting back from square one, unless you have some way to completely control/sanitize their input.

And again, every step in conversion isn't automated, there's still quite a bit of manual labor and checking that has to happen any time you convert between formats.

And Print as output? My gods, the Widows/Orphans alone...
Reply 

#9  Hitch 08-18-2019, 03:36 AM
Quote Tex2002ans
Also depends on how technical your team is, their current skillsets (do they all know Word or InDesign? Do they even know HTML/CSS?), and how much you can automate beforehand.
Yabbut, if the client was like that, they wouldn't be contracting the work. Right?


Quote
Maths??? FORGET ABOUT IT!
Oh, sure, RUB IT IN.
Spoiler Warning below






(Tex is consulting with me on a book that's pretty much all scientific notation/math, all the time.)

Quote
You're wrong.
Nope, you're just addicted to Open Source and you have that mindset that anything NOT OS = "eeevil." But I love ya anyway.

Quote

If you're working B2B, consistently working with the same publisher across hundreds/thousands of documents (Journal articles, or consistently Styled books), then perhaps you can talk... but it would require a lot of customized workflows/backends/tools.
It's always something. And if you think about it, all this customization, etc.--what's the upside to it? That the client can do the work themselves? (And not pay you, the originator and creator of this out-of-the-box system?)

Quote
And you're always going to come across these insane edge cases (2 columns with 1-column abstract up top? Top/Bottom floats? My book requires text to shrink like Alice in Wonderland?).
Yep.

Quote
Throwing an author updating things into the mix? Yeesh... that's starting back from square one, unless you have some way to completely control/sanitize their input.

And again, every step in conversion isn't automated, there's still quite a bit of manual labor and checking that has to happen any time you convert between formats.

And Print as output? My gods, the Widows/Orphans alone...
Indeed. (What are you doing up at what, nearly 3? You're getting as bad as I am with the "no sleep" thing....)

Hitch
Reply 

#10  Tex2002ans 08-18-2019, 04:12 AM
Quote Hitch
Yabbut, if the client was like that, they wouldn't be contracting the work. Right?
Completely depends.

Quote Hitch
Oh, sure, RUB IT IN.
Spoiler Warning below






(Tex is consulting with me on a book that's pretty much all scientific notation/math, all the time.)
Maths is a hard problem, and not many tools can even handle MathML:

https://indesign.uservoice.com/forums/601021-adobe-indesign-feature-requests/suggestions/32805088-mathml-support



Quote Hitch
Nope, you're just addicted to Open Source and you have that mindset that anything NOT OS = "eeevil." But I love ya anyway.
Proprietary tools can go suck it. :P

There are only very few that are the top of their game, and I will begrudgingly allow within my workflows, but they would have to jump a very high bar before I consider them (Toxaris's EPUBTools for DOCX cleanup is one).

Side Note: Did you see Microsoft got rid of full-license purchase of Microsoft Office 2019??? Now you have to pay monthly/yearly fees!

Quote Hitch
It's always something. And if you think about it, all this customization, etc.--what's the upside to it? That the client can do the work themselves? (And not pay you, the originator and creator of this out-of-the-box system?)
Those darned spreadsheets putting our slide-rule-operators out of business. :P

What's the purpose of any automation and efficiency gains?

In IT, you automate everything so you don't run into the issues in the first place, the entire place runs like a well-oiled machine. Things that took hours/minutes now take a button press. Uptime trends towards 99.999+%.

And there's always going to be those newer edge-cases:

I teach everyone I work with about Styles? I just saved myself (and them) tons of headaches, and now they need me for higher-level Problem X instead of (monotonous, horribly boring, "Enter-at-the-end-of-every-line" cleanup).

Quote Hitch
Indeed. (What are you doing up at what, nearly 3? You're getting as bad as I am with the "no sleep" thing....)
Thought you knew I'm a night owl.

That's something else I have to automate away, sleep!
Reply 

  Next »  Last »  (1/2)
Today's Posts | Search this Thread | Login | Register