The configuration file for a figure
Colaboratory for a hosted python notebook is a great way to get things done in this scheme. My only peeve with colaboratory is that I could not get it to work in directly interfacing with
github; otherwise in terms of building real time collaborative data analysis in python, hard to find a fault (particularly with Google power). Try
r-brain.io if you want. Another option is to use
mybinder.org. Any environment that allows you some slack to use git along with analytical tools should be fine.
While we are on this, some technical nitty gritty of setting up is in order, and once you see through the process, you will realise there is hardly any technicality involved other than just pushing a few files here and there and it does the magic. To keep things simple, we will focus on three different documents that can be updated and dynamically change the configuration of this text depending on how the work is being done and where the analyses are being conducted: the two config files and the layout file.
Authorea is so good that if you can click on a linked
jupyter notebook, you can even verify the analyses in Authorea's resulting jupyter notebook on their server in a Python 2.7 version. The Python 3 version does not always work with all modules, in particular the pandas module does not work. It also cannot accommodate julia or R. A much better idea is to download the file and process either at your end or use one of the web services.
So let's start with the configuration file for the figure (it's called config.yml) and it lives on the figure subfolder of the data folder. This file is editable, best of all. If you use git and your plain text editor, you can also get the job done that way. Anyway, it looks like as follows: