Following up on last week’s tip, here’s a near-data-disaster from Rollen D’Souza:
I decided when I started grad school that I would always keep track of my research and general course notes in repositories. This wasn’t entirely just for backup purposes. It turns out that when you want to work on three different machines — work desktop, home desktop, surface laptop — making sure they are all synchronized with your latest work is non-trivial without some automated or manual tracking software. I use Mercurial (distributed vcs) because then there is an entirely cloneable copy of all my work on every machine I work on. (Why Mercurial versus git? Another story.) I’ve gotten into the habit of pulling, committing and pushing whatever I have whenever I work on a given machine.