Formatting Tricks

If all goes well, this will be part one of however-many-it-takes posts on tricks for easy ebook formatting. These are not the be-all or end-all, but they’re the various methods I’ve developed to format books. I’ll be doing explanations of each part as I have time, and once I’ve explained all the tricks, I’ll pull them together into a cohesive step-by-step guide. These posts all assume a basic knowledge of HTML and CSS.

I know I said I was going to talk about Sigil next, but I’m taking a brief step sideways into something I take care of before I put my books together. This particular strategy has been a work in progress for several months, starting when I first heard Carolyn Jewel bring up the idea of using redirects to keep the links in the back of your ebooks from “going bad.”

Say for example the rights to some of your books are due to revert. If you link to the retailer listing of the publisher version and then republish it on your own later, every book you’ve sold so far will have a dead link in it. Naturally, you can publish a new version with the updated links, but since 1984-gate, Amazon has been a lot less likely to push new updates to people’s accounts. (And I have had similar trouble getting updated versions of books on Nook.)

Or maybe you know that the next three books in your series are going to be Awesome Book, Great Book and Amazing Book. You don’t have retailer links to these books yet, but you want to put them in the back of this book anyway, so people will know what’s coming. You can point those links to a temporary page, but you’ll still have to update those links at a later date.

And speaking of retailer links, there are some great arguments in favor of vendor-specific links. In the back of our Beyond books, I have customized links depending on the vendor. I do a different version for Amazon, B&N, Apple, Kobo, Google and a general version that links to our website. And in the beginning, that was a fair bit of work. I had tricks to make it easier, but they weren’t tricks I could easily explain to someone without a degree in computer science, and easier still wasn’t easy.

I solved all three of these problems with redirects.

BASIC IDEA BEHIND REDIRECTS

There are several ways to do redirects. If you already know what a redirect is and how to create them, you probably don’t need the rest of this post. You’re already thinking of possibilities. (Though skip down to my naming scheme trick to see how I create my versions for different vendors.) For the purposes of this post, I’m going to focus on one option: a redirection plugin in WordPress.

(Credit to Vivian Arend for finding Redirection and bringing it to my attention. That is the one I currently use. However, there are many, many redirect plugins and since things can change as plugins are abandoned, please do your own due diligence and make sure you find one that works for you.)

A redirect link is like a forwarding address. If you click on http://krocha.link/beyond-shame-amazon, you’ll end up on the Amazon page for Beyond Shame. That’s because krocha.link is a site I use just to manage my own redirects, kind of like my own personal bit.ly. (It cost me $15 to register the domain, wordpress & the redirection plugin were free. However, if your website is on wordpress, you can most likely install the plugin to your own site without the additional hassle.)

The reason this is useful is because I can use http://krocha.link/beyond-shame-amazon as the link in the back of my books whenever I want to link to the Amazon page for Beyond Shame, but the “forwarding address” (the website this link points to) is controlled via the redirection plugin. That means I can change where it goes at any time without having to update the book.

The source URL is yourdomain.com/source-url
The Target URL is where you want that link to point to. In this case, the amazon page.

I use this trick with my upcoming books as well. We know that Beyond Addiction is going to be next up, and since we’ve included the first chapter as a preview in Beyond Jealousy, I wanted to be able to link to it.

For now, this link points to the book page on our website. When the book becomes available, I’ll update the redirect Target URL with the amazon link. I make one of these for each vendor. They all link to the website for now, but will be updated with vendor-specific URLs as they become available.

Vivian Arend takes it a step further. Links to her upcoming books redirect straight to her newsletter, where readers can sign up to be notified when the book becomes available. Since you can change the target URL without touching the book itself, it’s easy to experiment on what you like best without constantly having to upload a new version.

REDIRECT NAMING SCHEME

We know from the previous lesson that I’m a huge fan of search & replace. When I sat down to come up with the naming scheme for my source URLs, I wanted two things:

I wanted it to be easy to remember.
krocha.link/book-title–vendor is something I don’t have to look up over & over. beyond-shame–amazon or beyond-pain–bn are easy to come up with on the fly. If I need to link to the apple version of Beyond Jealousy I know I can just use beyond-jealousy–apple. Easy peasy.
I wanted it to be something I could search & replace when changing vendors.
I make my base ebook file with all -amazon links. When the file has been finalized & tested, all I have to do to make my new versions is Save As (very important–you want to make a new file, not save over the existing one), and then run a search & replace. I find all of the -amazon and replace them with -bn. Then I Save As, find all the -bn and replace them with -apple. This process takes me very little time.

The first time you go through your book, it’s probably not a bad idea to do it one at a time to make sure there aren’t any stray places where you might have -apple or -general in your book’s text. It’s not likely, but stranger things have happened, and I always caution people to take care when deploying search & replace. (Learn from that naive time I changed a character’s name from Dan to Ben without remembering the book also had a Danielle. Or should I say Benielle.)

OTHER WAYS TO USE REDIRECTS

Once you start using them, they can be a little addictive. The Redirection plugin tracks the # of hits, so you can see how many people have clicked on any given link. At base minimum, it’s a good way to tell if people are clicking on your links at all. (I have tons of clicks from amazon books. Almost none from google.)

If you’re willing to take the time to be clever, it can also tell you where they’re clicking. Are people more likely to subscribe to your newsletter at the front or at the back? You can make multiple links to the same place. newsletter-front and newsletter-back can both go to your newsletter, but you’ll know if one gets significantly more clicks than the other.

And there are plenty of other questions to ask. Do readers respond better to a link to the next book directly after The End, or do they seem to like a nice orderly list? I want my ebooks to be exactly what my readers want. I’m constantly trying to think of ways to make them easier to navigate, and how to give readers the information they need without overloading them with stuff they don’t care about. All of this data is valuable.

For me, redirects are the key to making my front & backmatter simple. I have redirects for everything–our Facebook page, or twitter names (hey, I changed mine once… @moirarogersbree still gets follows from the backs of old books) even our websites. By keeping as much control as possible over these links, I’m making it easier on myself down the road. Because the one thing I’ve learned–things are always changing.

No really, next time. Sigil. I swear.

*Note — as some have pointed out in comments, the Mac version of Word is often a different beast. *

How to Convert A Manuscript to HTML with Search & Replace

Software you need to do this:

Microsoft Word
Windows Notepad (or a non-Windows version of a basic text editor)

Note: this is not a lesson on creating an entire ebook. This is just how to turn your manuscript into HTML. The next lesson will be how to turn that HTML into an epub.

I’m not really exaggerating when I say I format an entire manuscript via Search & Replace. When we’ve finished edits on a book, I open the file, copy/paste, run a Word macro I’ve created & end up with the converted HTML of the book itself. It takes me about 30 seconds. This does not involve any frontmatter or backmatter (which are very important and their own separate posts) but at the end of the day, you’re not really going anywhere until you have a book.

There are many good reasons to rely on cleanly converted HTML code instead of the code generated by a program like calibre or Word itself. The two biggest:

Possible Display Errors: the more complicated your code, the more likely it will break somewhere down the line. People read on a huge array of ereaders, phones, tablets, computers and other devices. Every format, platform, operating system and gadget reads and displays code in a slightly different way. The more work you make them do, the more likely they’re going to mess up.
Delivery Fees: complicated code is bulky code. If you’re self-publishing on Amazon via KDP, the 70% royalty option is really 70% – delivery fees. You can find the list of delivery fees on Amazon, but as of the time of this posting they are:
- Amazon.com: US $0.15/MB
- India on Amazon.com: INR ₹7/M
- Amazon CA: CAD $0.15/MB
- Brazil: BRL R$0.30/MB
- Amazon.co.uk: UK £0.10/MB
- Amazon.de: €0,12/MB
- Amazon.fr: €0,12/MB
- Amazon.es: €0,12/MB
- Amazon.it: €0,12/MB
- Amazon.co.jp: JPY1/MB for files less than 10MB, no charge for files equal to or over 10MB
- Amazon.com.mx: MXN $1/MB
- Amazon.com.au: AUD $0.15/MB

It may seem like 15 cents isn’t a lot, but if you sell 10,000 books, that 15 cents is $1,500. Books with a lot of fancy images (like a different header image for every chapter) can quickly scale up to 30 or 50 cents per book.

So, you’ve decided you want to turn your manuscript into HTML by yourself. There are several different considerations to tackle, and I’m going to handle them one by one.

CHAPTER HEADERS

I format books in Sigil. Sigil will automatically generate a table of contents and .NCX file. (The navigation file that tells ereaders, computers & gadgets where everything in your ebook is.) To decide where your headers are, Sigil looks for header tags:

<h1>Chapter One</h1>

So how does Search & Replace help with this? When I’m going through my manuscript one last time to prep it for formatting, I make sure all of my headers look like this:

[Chapter One]

It’s an easy thing to do while you’re writing, or even after you’re finished. Once they’re in place, though, you can use Search & Replace to put anything you want around your chapter headers.

As later lessons will show, I use this for more than just basic header tags. It allows me to tweak my header class, insert page breaks before my new chapters or images after. The first step of my macro puts a sigil split marker before the header (telling sigil it can break this chapter into its own section) as well as putting the header tags around my title.

If you have subheaders (for example, a chapter header and a smaller one indicating POV) you can use different symbols. In some instances, I’ve used [ ] wherever I want to replace something with <h1> and { } wherever I want to replace it with <h2>.

Note: obviously, if you replace [ with <h1> or { with <h2>, you need to do the opposite as well. ] with </h1> and } with </h2>.

SCENE BREAKS

We like to use images for scene breaks. It’s something I like because it’s pretty, it doesn’t take up a ton of space, and it doesn’t often backfire. Some methods of adding extra blank space can be stripped out by less savvy reading apps and devices. There are certainly many ways around this, but if you’re not super savvy at CSS and HTML, a scene break image isn’t a terrible way to go.

While writing our books, we indicate scene breaks with a simple:

When formatting, I search for # and replace it with the code for our dividing image. In Sigil, you can add an image to your epub file and get the relative path to it easily. If you’re simply creating an HTML file to test your formatting, make sure your image is in the same folder where you save your .html file and refer to it simply as “image.gif” (or .jpg or…)

ITALICS

Now we’re getting into something fun.

The key to this trick is knowing that you can put ^& into a REPLACE box as a placeholder for anything was found during the search portion of Search & Replace. The second key is knowing that you can put your cursor into the FIND box and hit CTRL+I and it will search for italics.

So the steps to search & replace italics are:

Open the Search & Replace box.
Put the cursor in the Find What box and hit CTRL+I. This should make Font: Italic show up beneath the box.
Put <em>^&</em> (or <i>^&</i> or your preferred method of emphasizing text) in the Replace With box.
Replace!

IMPORTANT NOTE: this is the sort of situation where a computer not being smarter than a human can come back to bite you if you’re not careful. We’ve all written something in italics, hit return, started typing the next paragraph and realized italics was still on. You can turn it off and keep going, but if you do that–and I can’t stress this enough–Word thinks the italics ends at the start of the next paragraph.

This results in situations like this:

She shook her head. <em>I guess I’ll just have to do this the hard way.

</em>The software silently agreed.

This situation is not terrible, unless you drop your code into something like sigil and let it “fix” your code on its own. It will probably look at that stray </em> tag all on its own and decide you didn’t mean for there to be italics there at all, or-worst case scenario–it could decide you wanted everything that follows to be in italics because it just ate the stray </em> and can’t remember that it was ever there. Those italics could cover entire chapters until it finds another </em> and cuts it out.

(Computers are only as smart as we tell them to be.)

The other problem is multiple lines of italics, which end up looking like this:

<em>This isn’t the greatest example in the world.

This is just a tribute.</em>

The former problem can be detected by searching for ^p</em> and replacing it with </em>^p which takes away the paragraph break before the tag and puts it afterwards. (^p is the next lesson.) The latter…well, you’ll just have to be smart. (And, if you use sigil, manually fix the errors!)

I know that the WARNING, DANGER section of this part is longer than the trick, but by paying attention to the ways these tricks can fail, you’re a lot more likely to survive using them.

PARAGRAPH TAGS

This is, arguably, one of the most basic steps, but I put it after italics for a reason–if you have funny italics, they’re a lot easier to fix before you start messing with your paragraphs.

The two most important codes you need to know for this step are ^p and ^l. Those are the codes for a paragraph and a line break, respectively. Hopefully most of you know which one you use when you’re writing. If you don’t, the quickest way to find out is to turn on the formatting marks in word. That will make your document look like this:

Most manuscripts I’ve seen fall into one of four categories:

First line indented manually, with paragraph breaks.
First line indented automatically, with paragraph breaks.
No indents, two line breaks between paragraphs.
No indents, two paragraph breaks between paragraphs.

Undoubtedly, there are a million more variations. Whatever you have separating your paragraphs, the general idea with this trick is that you want to replace that with </p>^l^l<p>. And no, I did not put those out of order. Consider the following:

This is paragraph one.^p

This is paragraph two. ^p

This is paragraph three.^p

Replacing ^p with </p>^l^l<p> gives us

This is paragraph one. </p>^l
^l
<p>This is paragraph two.</p>^l
^l
<p>This is paragraph three.</p>^l
^l
<p>

This effectively puts </p> at the end of every paragraph, inserts two line breaks, and puts <p> at the start of your next paragraph. This will work throughout your entire manuscript. BUT! It will not put the first <p> before your first paragraph, and it will leave an extra one at the end. So you’ll have to fix the start & end…but everything in the middle should work just fine.

Sometimes it takes a little finessing to figure out what the exact magic substitution is for YOUR manuscript, but you can change just about anything. Some important search/replace codes to know for this section:

Paragraph: ^p
Line Break: ^l
Tab: ^t
Manual Page Break: ^m

You can find a huge selection of codes here.

SPECIAL CHARACTERS

This is apparently a controversial topic. I googled to see if anyone had a definitive guide to best practices for ebook formatting and my browser blew up with drama.

Suffice to say–not all devices can necessarily recognize the fancy smart-quotes and beautiful em-dashes that Word kindly auto-formats for us. And the less standard your character, the more likely you are to have a problem. (Says the woman who somehow uses façade once per book.)

There are ways to get around this. You can remove them completely and replace them with the low tech versions–straight quotes and double dashes–but that won’t help you with your tildes and your umlauts. The best way to make sure everything displays properly is to replace your special characters with the HTML entity. Some very common ones:

Left/Right Double Quotes: “ (ld = left double, quo = quote) &rdquo; (rd = right double, etc..)
Left/Right Single Quotes: ‘ (ls = left single) ’ (…etc)
Em-dash & Ellipsis: — and …

When searching & replacing, it can be easiest to find the special character somewhere (like the link above) and copy it into the FIND box. Then type the HTML entity into the REPLACE WITH box.

Honestly, search/replacing every possible code every time would be prohibitive. And this step is where macros truly become vital. If you sit down once with a list of every special character you might possibly use, you can use Word’s very simple ability to record a macro. What does that mean, practically speaking? Once you’ve got the macro recording, it will keep track of every single thing you run a Search/Replace for. (Even if that thing doesn’t exist in the manuscript.) Then, the next time you sit down to replace special characters, you can run the macro and it will repeat everything you just did.

I have macros for all of these steps, and one macro I can run that runs all of the macros. It took me a little time to finesse all of these steps to work exactly the way I wanted them, but in the end? I can turn out clean HTML in a matter of seconds.

CLEANING UP

There are always some loose ends hanging around. Maybe I have places where I had extra paragraph breaks, so I have <p></p> scattered throughout my manuscript. Well, that’s easy to fix. I just want those to go away, so I can run a Search & Replace to find <p></p> and replace it with nothing. (Yes! That’s an option!)

Well, okay, but if I replace <p></p> with nothing, I might have a ton of line breaks all in a row. All that blank space won’t hurt my HTML, but it’s not very pretty. If I want to decrease my extra line breaks, I can run a search for ^l^l^l and replace it with ^l^l. If there are still extra spaces, I can do it again until I can’t find any groups of three line breaks.

(Do not replace ^l^l with ^l unless you want NO space between your paragraphs.)

Very last of all, I copy everything I’ve just generated and paste it into a very basic text editor. Notepad is a good one. Really, you don’t want anything more complicated. Saving it in plain text makes sure we’ve ditched any odd Word formatting that might be clinging to it–and trust me. Nothing is more likely to go bad on you than Word formatting.

At this stage, I usually put <html> at the top and </html> at the end, save it as testbook.html and open it up in a browser to see how things are looking. You’ll be able to tell straight off if your chapter headers are actual headers, if your italics look okay–and if not, where they broke–and if all of your special characters are as special as they should be. (And if your scene divider image is in the right directory, you should see it, too!)

There are many things I do once the book is in sigil to make this book more pretty. But this very first step is about making the code pretty, and if you’ve gotten this far and made it out alive, your code is probably looking pretty good.

THE BIGGEST SECRET IS

That you can automate just about anything if you think about a way to let the computer know where it is. Word can search for more than italics. It can search for different font faces, different font sizes… if you have different character POVs and want them to display slightly differently, you could use different fonts in your manuscript and find a way to make that work.

The only limit to Search & Replace is how creative you are when it comes to bossing computers around.

QUESTIONS?

Feel free to ask them. I will do my best to answer, or point you to someone who can. 🙂

NEXT UP…

Putting this manuscript into SIGIL! (I love it.)

Formatting Tricks (2/?): Linking With Redirects

Formatting Tricks (1/?): Search & Replace in Word

How to Convert A Manuscript to HTML with Search & Replace