forked from swcarpentry/git-novice
-
Notifications
You must be signed in to change notification settings - Fork 1
Rationale for Changes to Git Curriculum
John Pellman edited this page Feb 27, 2020
·
2 revisions
- The command line is intimidating for folks that are new to git. The original curriculum increases cognitive load for beginners by forcing them to keep both unrehearsed knowledge of bash in working memory and new knowledge about git. The command line is a distraction from the fundamental lessons we're trying to teach about git.
- A lot of the command line sections are pedantic (i.e., talking about carriage returns vs newlines) and detract from the main thrust of what we're trying to accomplish.
- Giving researchers choices (e.g., text editor and hosting service) dramatically increases the number of things that could go wrong during a lesson and also diminishes the lesson's focus and scope. We also want to avoid overchoice and holy wars. If people don't like Atom, we'll point them to the traditional Software Carpentry curriculum.
- The Open Science section is an evangelism piece and out of scope. These lessons are meant to teach git, not discuss issues of open science. In all likelihood, this section is either preaching to the choir or unconvincing to folks who don't believe in open practices. We're making a deliberate decision to make these lessons more utilitarian, even if our personal convictions may (or may not) be in favor of open science (whatever that means). Good version control practices are useful even within the context of proprietary IP, and are orthogonal to openness.
- We can provide more detail about licensing practices that are specific to Columbia in the licensing section.
- RStudio broadens the scope too much. I'm unaware of anyone using RStudio at ZI as yet. Perhaps this could come back if there is demand for it.
- The Wolfman and Dracula example is nonsensical and distracting. We want to bring in examples that Zuckerman researchers might relate better to.
Overall, we feel that the original curriculum tries to accomplish too many goals and as a result is lacking in focus. Since we have a very specific group of clientele at Zuckerman, we can afford to tailor it to better suit their needs and Institute workflows.