RSH is an interactive, streaming web show all about scientific computing and research software.

Programming can be fun. Research can be fun. But when combined, many skills fall through the gaps. We think that computing is best learned like an apprenticeship, but unfortunately most researchers don’t have access to high-quality computing mentorship.

We’ll develop the program based on suggestions and also to maximize fun for us, but broadly speaking we wish to:

  • Have entertaining discussions about research software and computing. Show why we enjoy it so much and how you can, too.

  • Teach tips and tricks (and learn new ones ourselves, too) - at least introduce new ideas and point you where to look for more info.

  • Take code submitted by users and go through it and discuss how we’d improve it - to show our whole thought process. We will do this constructively and also critically review our own codes which we have written some time ago.

  • Show you the errors we make and how we get through them. And do our own learning at the same time, too. We are not experts in everything and we believe in the pedagogy of errors and typos.

  • We’ll cover everything from research software and data to high-performance computing and Linux.

This is an open source production - anyone can help us develop RSH, too. Join us!

About us

Radovan Bast is a research software engineer with background in theoretical chemistry. He’s worked in France, Stockholm, and now Tromsø at the border between science, software, and computational support and enjoys supporting multi-disciplinary research. He now works as part of the Sigma2 metacenter at the University of Tromsø, Norway, and leads the CodeRefinery project.

Richard Darst is a computational data scientist with background in network science and theoretical chemistry, and has worked in the United States and Finland. He now works for Aalto Scientific Computing at Aalto University, Finland. They met through the CodeRefinery project, which teaches best practices in research software, and also collaborate on the Nordic Research Software Engineer community project and NordicHPC.

Together, we have decades of experience programming for science, mentoring, and teaching computing. We have seen a crisis of computing, where computing demands are increasing while there isn’t a corresponding increase in the practical skills that go along with it. Mentoring does not reach enough people, and teaching does not take the role of hands-on mentoring or convey spirit of computing (and also still doesn’t reach enough people). There is also lack of time and academic credit for investing time in improving programming and computing skills.

Thus, we need another strategy. We are trying this web show to combine informal teaching with the feeling and spirit of computing.

Topics

Each week, we want to do some simple stuff, some more advanced stuff, some stuff more towards software/programming, some stuff more towards scientific computing/Linux. Most important, we want to show you the spirit and joy of computing.

See the full list of our latest ideas in the rsh-notes repository and suggest more there. Broad categories and examples below, and our program will evolve as our audience likes:

  • Research software, since these days so many people need to produce it, regardless of background: structuring software, profiling, demonstrating useful packages for research, releasing software, pair programming in our normal work.

  • Scientific computing support skills, tools you need to do scientific computing: git, automated testing, documentation, working with pip/conda, Github, Jupyter.

  • Evaluate watcher-submitted code: We’ll take code our watchers submit, evaluate on stream and give suggestions for improvements. Suggest by making an issue in the rse-notes repository

  • Go through bits of regular courses, such as CodeRefinery.

  • Unix and Linux tools, since this is the base which most scientific computing is built on: bash, ssh, tmux/screen, shell scripting, etc.

  • Data management, since often the data is harder than the code: optimizing data access, data formats, repositories, sharing data.

  • Open science, reproducibility, and credit, since good software leads to good science and vice versa: preparing software for release and wide use, releasing software, making software citeable.

  • Discussion time, we answer any questions you may have. Try to stump us! We will all learn from each other.