I would highly recommend learning git to a deep level as it confers a huge force-amplifying power for both day-to-day development as well as the long-term maint

I would highly recommend learning git to a deep level as it confers a huge force-amplifying power for both day-to-day development as well as the long-term maintainability of a well-constructed commit history. The CLI is definitely obtuse, but the underlying data model is quite simple and elegant. Once you grok the primitives you can learn the idiosyncrasies of the CLI at your own pace, secure with knowing that you always have reflog and cherry-pick as a safety net. My recommendation is to use a client like Sourcetree to visualize branches/commits as you learn new operations. Having come up in the CVS, and then Subversion eras, I acutely remember in 5 years of using Subversion that I never truly understood the underlying data model (a branch is a tag is a directory!?) as well as I understood git after 3 months. It's a very very good data model. Mercurial is arguably equivalent with a better CLI. I won't argue the comparison here, but given the mindshare git has it seems like you'd be losing a lot by constraining yourself to a different version control system. I promise git is extremely usable once the primitives click in your brain, and likely to outlive most of your other dev tools like editors, languages, other toolchains, etc.