What software do you use to write?  I wrote three complete novels using Apple’s Pages.  It’s very simple and clean, although it does have limitations.  For example, I had separate files for each chapter because putting it all into one file meant too much scrolling and performance that was far too slow.  To make a PDF of an entire novel, that required me to make a single large file from the smaller ones.  It was a pain, but workable enough that I stuck with it for three books.

At the beginning of February 2018, I downloaded Scrivener from Literature and Latte.  I had heard of it a long time ago, but had never checked into it.  It was recommended, however, by people at a seminar I attended in January 2018 by Hackerfarm and Zoot Publishing.  It took a few hours to copy Under Shōko’s Bed into Scrivener scene by scene, but it has been wonderful to work with the entire novel at once.  The navigation and search capabilities are great.  I can add metadata to each scene (dates, characters, symbolism, and in Neyuki, even the depth of the snow).  Scrivener’s “binder,” a sort of outline, makes it easy to move from one place to another in the text.  It really is miles ahead of using a word processing program.  I highly recommend Scrivener to anyone who plans on writing anything big.

But—getting your work out of Scrivener can be a terrible pain.  On the positive side, you can create almost any kind of file with Scrivener, including EPUB and MOBI (Kindle) files, so a single Scrivener project can suffice for all.  If you make changes to the text, you can generate new output files in all of your various file types absolutely painlessly.  The problem is in generating those output files the first time.  That is anything but simple, and as much trouble as it is to set up one output format, each other output format requires just as much trouble.  Coming from the world of Pages, a WYSIWYG (what you see is what you get) word processor, I was used to being able to see my formatting changes as I made them.  With Scrivener, I have to define how each type of text section will be formatted in the output, and it is not WYSIWYG at all.  I make changes to the formatting, then compile the document and look at it.  At which point, more often than not, I scratch my head and wonder why things have come out with some element of my desired format missing or wrong.  Then I adjust a setting and compile and see the results again—and again and again and again.  With some of the changes, I finally just gave up and went with what Scrivener had produced.

The power in Scrivener’s approach to formatting is that you can make a change for one type of text section and that change is automatically applied to every section of that type.  You want a Roman numeral “I” as your chapter title instead of “Chapter One”?  Then change the code in the appropriate section formatting and all of your chapters are fixed.  It’s great, if you know the code.

Where all of this has left me is capable but not confident.  I have gotten far enough with the formatting to be able to generate manuscripts, PDFs, EPUBs, and MOBIs that I have used to give friends, alpha and beta readers, copies of novels to read, but I don’t quite trust any of my formatted output for actual self-publishing yet.  Maybe it’s a matter of trying and succeeding a few more times.  Maybe.  But I still fear that I will put my MOBI onto Amazon and have complaints start streaming in by the thousands.

Of course, that would imply that I was selling thousands.

Hey, an author can dream, right?  In the end, though, maybe I should worry less and just publish.