We go to a lot of trouble to make our book descriptions as good as we can — not just their content, but the way they look. Many of us have learned basic HTML tagging in order to provide formatting for book descriptions as part of our work publishing a book.

As soon as you go beyond the world of Amazon-only, you begin to lose control of what distributors and retailers do to your pretty book descriptions. And it takes special effort to make the book description inside your book look good, too.

Doesn't always work, though. Not all methods are suitable for all situations, and there are limits to what you can control.

As in all such things, the devil's in the details.

(click on any image below to enlarge it)

HTML Markup to use for websites

What I think of as the original text usually appears on the book's product page on your own author and/or publisher websites.

You have complete control over what this looks like.

You can see a simple use of <strong> and </strong> to mark the bold section (same as using <b> and </b>). This was created in WordPress, and the paragraph behavior is instantiated by the underlying WordPress theme (in other words, I didn't need to place paragraph marks <p> and </p> around each paragraph.)

HTML Markup to use for ebook front-of-book blurbs

Ebooks are special packages of HTML files, in a compressed (zip) container. I make my ebooks using Sigil, and each chapter is an underlying file in the container. The opening page shows a book description.

You have complete control over what this looks like, within the limitations of HTML supported by ebooks.

The HTML code behind it includes styles I created in a CSS style sheet, like “block” which makes the first paragraph not indent, or “chapter” and “block sgc-1” which add bold to the text. Note that every paragraph is surrounded by paragraph marks <p> and </p>.

HTML Markup to use for retailer product book pages and others

So far, so good. Where things become tricky is when you can't be sure how (and when) HTML code is permitted and where it can be used in less obvious situations.

Closed systems

There are retailers with modern technology operating within closed systems. Take Amazon, for example. Not only do they provide simple formatting like bold text, but they support paragraph breaks.

See how this Amazon Author page shows you what it produces. Instead of paragraph tags <p> and </p>, it supplies line break tags <br/>.

Everywhere in the Amazon set of stores, the formatting is preserved.

Then there are the almost closed systems, like Kobo. Not only does Kobo support stores in many countries, it is also the source for other retailers.

It provides similar tools for book description entry as Amazon.

You can't tell how it tagged the paragraphs or line breaks, but it looks fine on the Kobo closed-system sites.But Kobo distributes to other retailers. Even Indigo, a parent company, has lost the paragraph breaks (or, more accurately, has defined them with a different style re: spacing) and the bold text markup.

Older technology sites provide no markup. Sometimes they at least preserve the paragraph marks in some form, like this Smashwords input.

Which looks fine on the Smashwords site.

But by the time Smashwords forwards the information downstream, to Blio for example, even that modest formatting is lost.

Ingram LSI, for example, formats text nicely, like Amazon.

Though it marks up the paragraphs with <p> and </p> tags, instead of line breaks <br/>.

That works fine for downstream retailer Barnes & Noble…

But it doesn't survive the handoff to downstream retailer Foyle's, demonstrating that the problem is sometimes in the sender, and sometimes in the receiver.

 

HTML Markup to use for inside the book metadata and other do-not-interpret situations

Sometimes you can't use straight HTML markup. Here's the problem (grossly simplified)…

Let's say that you have a book description with HTML markup tags, and that goes into a database system somewhere. Some other program looks at that data. If that program is itself HTML-based, it might interpret those HTML tags instead of just passing them along as part of the data. So, instead, you have to use what are called escape characters.

Here's how the book description is formatted within the ebook in the Content.opf file in the ebook container. The “<” is the XML escape character for “lesser than”, or “<“. So, “<div><p class” is actually “<div><p class”.

Programs which read this data, such as Calibre, can interpret it as a fully formatted description (along with the other metadata listed here).

The description inside a Bowker ISBN record, for example, only uses the HTML <p> tag (and not the </p> to close a paragraph). That's a common situation in some other contexts. You can also use escape characters here.

It comes out looking like this.

 

We're lucky it isn't worse, in my opinion.

Irritating, though.