Repository for the presentation about "development and publishing of reproducible statistical workflows" (working title).
- simple rather than exthaustive, to also include and attract "non-computerscientists"
- git
- introduction
- what is git/github
- Purpose: demonstrate that everyone should use git/github when working at the computer
- overview on functionality
- basic commands clone/pull/add/commit/push
- release?
- private/public/github at our servers
- intended use/limitations (e.g. filesize)
- markup?
- demonstration
- kept simple
- demonstration of basic commands (short)
- more demonstraton combining github and jupyter after in the next section
- introduction
- jupyter notebooks
- introduction
- what is jupyter notebooks
- Purpose: demonstrate that it is easy to use (also remote) and looks nice
- how to install kernels: python3/R
- overview on functionality
- layout of page
- code cells/markdown cells
- some usefull magic commadns
- outlook: jupyterLab? "release JupyterLab 1.0 later in 2018"
- demonstration
- use port forwarding to run notebook on the institutes servers
- example including git and jupyter notebook
- realistic, drawn from everyday work, generic but still simple
- include git
- guidelines for using git with jupyter notebooks
- introduction
- git
- commands
- link to this repository?
- ...
- notebooks
- setup kernels
- howto start locally
- port forwarding
- check forwarded ports
- notebooks in git
- are rendered
- guidelines: export code, save images
- cool drawings / sketchnote style?
- check current ssh tunnels (also from other users, without beeing root)
- ps aux |grep ssh seems to work
- netstat from the net-tools package (difficult to install here, but I managed to patch it)