Last week, author Charlie Stross posted his review of the process of writing using Scrivener, a specialized story-based word processor I’ve mentioned a few times. Stross has a good overview of the program’s strengths and weaknesses from the point of view of a professionally-published writer.

The program’s biggest weakness, he finds, is that it essentially becomes useless at the point a novel is finished and submitted to the publisher—because the Word document output isn’t quite ideal for submission, and then the publisher will send revisions in the form of Word documents, and expect them to be processed accordingly. Since Scrivener can’t do anything with the revisions, Stross has to use LibreOffice from then on.

This isn’t a formal review: it’s just a comment to the effect that Scrivener works pretty much from the moment of conception to the hour before final submission of a finished manuscript. It doesn’t completely replace the word processor in my workflow, but it relegates it to a markup and proofing tool rather than being a central element of the process of creating a book. And that’s about as major a change as the author’s job has undergone since WYSIWYG word processing came along in the late 80s (actually the late 70s if you were a researcher at Xerox PARC, but the rest of us had to wait). My suspicion is that if this sort of tool spreads, the long-term result may be better structured novels with fewer dangling plot threads and internal inconsistencies. But time will tell.

Stross also notes that Scrivener produces “some of the cleanest EPUB files that [he’s] ever seen,” which makes it a little puzzling that the help file comes as a PDF instead. (Someone in the comments points out that’s probably because almost everyone has a PDF reader on their desktop but few people have EPUB readers there.)

I’ve lately been using Scrivener for writing stories in a collaborative Internet fiction setting a friend and I have come up with between us. I’ve found it works great for writing the story, and for producing output files in several different formats if you’re not going to be sending them any further than out to the Internet for other people to read. The scene-based nature of it works great if you want to edit or rewrite just one specific scene—you don’t find yourself pressing control-end and then jumping all the way to the end of the document.

I’ve only run across a couple of major problems for me.

First, in my experience it’s really difficult to get things written elsewhere imported into the program such that you don’t lose any italics you’ve got. Either it pastes in the wrong font, or if you force it to match the style you lose the italics. I’m not too sure what to do about that. I may need to do more experimentation to see if I can figure out a way to make it work better. This is problematic if I’m doing any sort of collaborative work, like writing out a scene with someone using Etherpad.

The other problem is that it can be tricky to get your work to output properly into formats that aren’t standard markups. For example, I needed to post my stories to the Shifti transformation stories wiki. But in order to do that I had to pass them through Word or RTF output, open them in LibreOffice, copy and paste them into the WikEd wysywig-conversion text editor to wikify them, then copy and paste them into an Emacs clone to remove all the extra line spacing and do some final wiki markup cleanup (such as formatting the chapter titles as wikicode section titles), then copy and paste them back into WikEd for full wiki conversion. (In fact, I had to copy and paste it back and forth several times to do a Wiki preview and see if it looked right.) I did get the stories posted all right, but there’s got to be an easier way!


  1. You write: ” it’s really difficult to get things written elsewhere imported into the program such that you don’t lose any italics you’ve got.”

    You might try writing in Markdown, which Scrivener understands. Simply tag every italicized word or phase like this: *This is in italics.* That’ll even work inside pure text editors without formatting. You can also use the # sign to tag chapter and scene changes.

    My only major complaint, which I have shared with the developer, is that Scrivener shares the inadequacies of OS X’s text functions. You can’t apply a named style to paragraphs, something that’d make it very easy to export even a complex book to a layout program such as InDesign.

    Apple’s execs apparently still think WordStar circa 1982 is the definitive word processing app. Their own Word-like features don’t even match those of Word 4.0 in the late 1980s. Pitiful.

    OS X’s spell-check suggestions are awful too. About 1/3 the time, it has no suggestions. I paste that same misspelled word into a Google search and Google gives me the correct spelling about 95% of the time. Again pitiful. Apple spends tens of millions making audio and video compression a tiny bit more efficient. It could spend a few hundred thousand getting its text functionality out of the 1980s

    Change marking is a major issue with anyone writing for a publisher. Writing, editing and publishing desperately need a solution to change markup that’s more universal than requiring everyone to introduce Word into their workflow. Personally, I find marking up a PDF and hand-entering the changes much better. It lets you skip Word and work directly with an InDesign typeset copy. That also catches any error the Word to InDesign/Quark process might create–and that can be quite a few errors.

    –Michael W. Perry, author of Untangling Tolkien

    • I tried that. But then I found there was no way to import it properly in the Windows version. It’s one of those features in the OS X version that they keep meaning to bring over and never get around to.

      (Didn’t help matters that when I posted about this on their forums, one of the programmers or support people swore up and down that, no, being able to write italics in text editors and import them as italics into your project was not what MMD is for.)

  2. I love Scrivener, but I find that the final polish needs to be done in Word (or another word processor).

    I am self published and still find that a .doc works best for critiquing and editing. Now that I publish directly to Kobo Writing Life, I copy the whole manuscript back into one scrivener page to compile as epub – another reason to love Scrivener.

  3. Stross is correct, and yet this is no knock on the program. Scrivener was always intended to serve as a means to finish research and organize a long-form of writing. When the organization is complete, the writer is supposed to export to rtf or doc and then polish and copy-edit and style in MS-Word or other word processor.

    It’s only because users keep asking for more and more features and bloat, that Scrivener is as close to a word processor as it has become.

    Remember yWriter? That was the inspiration; since there was no version for Mac, the author/programmer created Scrivener.

  4. Count me another Scrivener fan. I saw so many positive comments on Kindle Boards I had to try it and downloaded the free trial. However, I can’t imagine using it to produce a finished product and will export to Word Perfect (yes, I’m stubborn and not using Word in this life if I can help it) for the final work. I almost hate hearing Scrivener exports to ebook formats so well because I’m not going to put my work back in Scrivener for that purpose.

The TeleRead community values your civil and thoughtful comments. We use a cache, so expect a delay. Problems? E-mail