Daniel Wheeler is a computer scientist at the National Institute of Standards and Technology. He is primarily interested in the development and deployment of software for applied scientific applications. He has expertise working with Python and numerical tool kits such as Numpy, Scipy and Trilinos in high performance computing environments.
What I found hardest about using Git the first time was…
I was a long time CVS and SVN user. The biggest challenges were moving away from the mental model of central repositories, managing branches in a single working copy and managing remote repositories. After becoming familiar with Git, I switched a project with multiple collaborators from SVN to Git and my intuition was that the Git model would result in complete confusion. My intuition was completely wrong. I now, of course, can’t even remember how SVN and CVS work. More recently, I’ve been working on a project that uses Mercurial. This has proved to be enormously painful because I am entirely confused by its branching model and when to use the bookmarks etc. I really like Git’s lightweight branching compared with Mercurial (I acknowledge that this is due to my lack of familiarity with Mercurial and that both systems seem entirely similar to most).