Data Analysis with R, Second Edition - Second Edition

Version control

A very compelling benefit to our neat hierarchical organization scheme is that it lends itself to easy integration with version control systems. Version control systems, at a basic level, allow one to track changes/revisions to a set of files, and easily roll back to previous states of the set of files.

A simple (and inadequate) approach is to compress your analysis project at regular intervals, and post-fix the filename of each compressed copy with a timestamp. This way, if you make a mistake, and would like to revert to a previous version, all you have to do is delete your current project and un-compress the project from the time you want to roll back to.

A far more sane solution is to use a remote file synchronization service ...