Dear all, we just published a paper that describes how Notebooks and myBinder were used to engage students with self-paced problem solving exercises based on sophisticated computer simulation.
The content of the course is about magnetic materials. A Python package (Ubermag) is used to interface to (compiled) legacy simulation tools. The Python interface can be controlled from the Jupyter notebooks, and provides a learner-friendly interface to the functionality of the legacy simulation tool (OOMMF).
The paper describes this in more detail, and includes some feedback from students and teachers involved.
There are also some recommendation for such “Computational Projects” that should be transferrable to other domains.
I found this passage interesting: “The installation of software for teaching purposes can be challenging: The university’s or the students’ personal laptops may be running a variety of operating systems (typically Windows, MacOS or Linux) with different versions. More complex simulation software environments may need multiple libraries of compatible versions to be installed simultaneously.”
Did your team considering trying to run the notebooks on something like JupyterHub with LTIAuthenticator then connecting it to Canvas via LTI (such that the student wouldn’t need to set anything up)? Or was that kind of approach not feasible?
From your paper it seems like some students set up Ubermag themselves and some relied on URLs on MyBinder that launched a pre-configured enviro. But either way the resource wasn’t part of the Canvas-based learning content. That might not have been a perceived issue, just wondering if you considered an LTI connection to integrate content or if it was too problematic for some reason.