Ideas for RSE Guidelines Contents

[lsetiawan]

Overview

The RSE Guidelines have been very useful for our internal team as well as collaborators. It's very important to capture our best practices as well as have direct guides on all the different tooling that we utilize within SSEC that is also useful for others.

---

AI for software engineering

---

The idea of this section is to have a collective guidelines on the usage of AI for RSE. An example of this is utilizing copilot within VSCode for day to day work, any pitfalls of AI to look out for, and other resources that will help development process.

• # for Copilot
• Working with ChatGPT
• Utilize local LLMs?

---

Development Environments

---

All about development environments. Currently there's a Computing environment section. We'll probably need to revamp this to be a development/computing environments

• Conda/Mamba
• Pixi
• Docker
• Notebooks and interactive environments

---

Code style and conventions

---

This section could introduce to the various coding conventions. For python mainly the pep8 and numpydoc or googledoc.

https://peps.python.org/pep-0008/
https://numpydoc.readthedocs.io/en/latest/format.html
https://google.github.io/styleguide/pyguide.html

We currently have a commit convention in the fundamental.. but I think that's more of a process convention.

---

Code collaboration tools

---

This could be a sections on how to do pair-programming and what tools we can use

• VSCode liveshare
• Github Codespaces

---

Visualization

---

This will be some section about visualization and storytelling techniques, we can cover different visualization tools that we've used

• Streamlit
• Holoviz

---

Debugging

---

Currently we have a Debugging with VSCode section. This needs to be revamped and expanded to include more concepts for debugging.. what it means to debug, different strategies, etc.

https://code.visualstudio.com/docs/editor/debugging

---

Code Profiling

---

This will be a section where one can learn about the code profiling, what is it and how to do it. Some guidance of what to look out for would be great and how one can improve code performance by profiling and modifying a piece of code.

• Memory profiling
• CPU profiling
• GPU profiling?

---

IDEs

---

This will be a fairly short section introducing IDEs and have a list of preferred, some recommended, and IDEs to avoid.

---

Github Project Templates

---

This section is more of a pointer to the various templates that we currently have such as:

• project-template
• python-project-template

Any additional templates can be pointed from this section.. also if there's any more advance cookiecutter such as the Scientific python cookiecutter, this can be added here as well.

---

Advanced

---

This will be an advanced section for tools and infrastructure. There will be subsections below.

Code packaging

This is all about code packaging for various programming languages.

• Code
  • Python
  • C++
  • Rust
  • Etc...
  • Multilingual package
• Docs
  • Sphinx
  • Jupyterbook
  • Mkdocs
  • Quatro
  • Rustdoc
• Test
  • Python: https://docs.pytest.org/en/7.4.x/
• Infrastructure
  • Python
    • Readthedocs
    • PyPi Publishing
  • Github Infrastructures
    • Actions
    • Dependabot
  • Pre-commit
  • Codecov
  • Conda-forge

Code Deployments

All about code deployment tools and how one can deploy things to production

• Cloud
• Kubernetes
• Docker Advanced
  • Docker swarm
  • Docker registries
    • https://github.com/features/packages (Docker registry)
• Mobile App
  • APK
  • PWA

---

Fostering community

---

This section is about fostering an open source community

Code backbone

This is a subsection for fostering community and covers what needs to go as part of the code repository to make it easy for members within a community to contribute to the code and how to prepare for citations.

• Community health files
• Zenodo
• Citation.cff

Community workshops

A small guide on how to hold a community workshop or any optional programs such as hackweeks.

[lsetiawan]

I think having a part on geospatial information system would be very useful

Geospatial Information System

This part covers topics about working with geospatial data in an open-source way

• Intro to coordinate systems: https://ihatecoordinatesystems.com/
• Vector: https://geohackweek.github.io/vector/
• Raster: https://geohackweek.github.io/raster/