Grant Rettke over at Wisdom and Wonder points to an interesting discussion on the Org mailing list about writing a book in Org mode. The thread starts with Vikas Rawal’s announcement of his book, Ending Malnutrition: from commitment to action and a thank you to the members of the list for their help in working out some of the technical details. The rest of the thread discusses some of the problems Rawal solved in the course of publishing the book, including working with non-Emacs-using coauthors. Vikas also linked to the org source for the book.
This isn’t the first time Org mode has been used to write a book, of course, but the issue of its practicality has been discussed lately by Karl Voit and others so it’s convenient to have another example of a finished product. Voit complains that using LaTeX allows the author to tweak the output down to the micro level and thus produce a better final product. Vikas, John Kitchin, and other authors who’ve written in Org point out that you can embed LaTeX micro tweaks into the Org source file and bring to bear the power of Org Babel and reproducible research in preparing the manuscript.
If I were writing a strictly mathematical text with no empirical data, I might choose to use LaTeX/AUCTeX directly but if there were charts, graphs, or data calculations, I’d almost certainly choose to use Org for the manuscript, embedding any specialized LaTeX formatting directly in the source, using Babel for the calculations, diagrams, and graphs, and then export the result to LaTeX for use in producing the camera ready PDF.
The tools to do all this are available right now and they’re getting better all the time. For example, if you don’t like the Org exporter, you can use pandoc to covert your Org to LaTeX. As John Kitchin continues to demonstrate, there is less and less reason to prepare scientific documents in anything but Org.