Topics to cover in Personal Research

[dcsoto]

Fundamental tools:

• Choose your language: R, Python, bash, others? Should be based on your objectives (e.g. is it mainly statistical analysis of data? will it be eventually packaged as a command line tool? do you require pre-existing code out there? (mention interfaces if more than one language is needed)
• Where to start writing code: text editor (Sublime, Atom, Visual Studio Code), IDE and notebooks (RStudio, Jupyter Notebook, Jupyter Lab), CLI
• Literate programming: Markdown, R Markdown, Jupyter Notebooks.
• Coding good practices (variable naming, linter, pep8, commenting, functions, modularize, paradigms): mainly for your future self
• Version control: Git and Git GUIs (GitHub Desktop, GitKraken), GitHub/GitLab/BitBucket
• How to organize your project structure: reproducible research template initiatives (basic ideas of keeping your data separate from code, versioning your results, etc)
• Software versioning and environment managers (Conda, pip, virtualenv, pyenv, etc.)

[BenjaSanchez]

A non-exhaustive list of useful links:

• https://stackoverflow.com/questions/41573587/what-is-the-difference-between-venv-pyvenv-pyenv-virtualenv-virtualenvwrappe
• https://github.com/Reproducible-Science-Curriculum/rr-init
• https://github.com/drivendata/cookiecutter-data-science

[dcsoto]

Other useful tools:

• Data version control https://github.com/iterative/dvc. A version control system for data and models that works in combination with GitHub.
• Ways to share (big amounts of) data and outputs with the community: Open Science Framework https://osf.io/, FigShare https://figshare.com/, Git LSF https://git-lfs.github.com/

[dcsoto]

We shouldn't forget to include somewhere:

• Coding best practices: include assertions
• Best practices writing commit messages
• Pair programming
• Rubber duck debugging
• Google CoLab

[dcsoto]

Some items to discuss in our next meeting:

• Maybe change steps to: (1) project structure, (2) programming languages, (3) working environment, (4) good practices
• two table alternatives
• Include appendix with recommended packages? (e.g., https://github.com/const-ae/ggsigni, patchwork)
• I did not dived into the differences of compiled/interpreted programming languages (low-level/high-level, memory efficiency, speed) . I can look for a reference or include a more detailed explanation in an appendix.

[BenjaSanchez]

@dcsoto we did not mention pair programming + rubber duck debugging in the latest version of the manuscript, do we want to add those?