The Jupyter developers have presented Jupyter Books, a new open source project for the print-ready publication of computer-aided technical and, in particular, executable content in book and HTML form. As the data scientist and neuroscientist Chris Holdgraf announced in the Jupyter blog, Jupyter Book is part of The Executable Book Project, which roughly means “project executable book” (short: EBP).
Jupyter’s executable narratives
EBP is an international community with an interest in open source tools that work together to promote the publication of computational narratives (such as “computer-aided narratives”) in the Jupyter ecosystem (and beyond).
At this point in time, Jupyter Book is still in beta. The technical goals are already set, and the The status of the development can be seen in the documentation: Developers should be able to use it to create publication-quality technical content using markdown. They can either use Jupyter Markdown or, alternatively, use an extended function of the new tool with publication features. The format apparently supports rich syntax such as citations and internal references, mathematical formulas and figures. Code and output can be written in Jupyter notebooks and then exported in book form.
Writing books with markdown
Individual issues can be temporarily saved and recycled for later use, numerous interactive elements can be integrated, including the connection to online services such as Binder. There are several output formats to choose from for the actual output, users can choose between single and multi-page websites or have their compiled content output as PDF. Books can then be created from the command line with the command
jupyter-book build mybook. The Jupyter documentation website is already created with Jupyter Book and can serve as a trial copy.
Further Information on the subject of the Executable Book can be found on the EBP website. Details on the announcement of Jupyter Book can be found in Chris Holdgraf’s blog entry on Jupyter. The Project is anchored on GitHub.