Subversion as a Distributed file system
March 29, 2008
Posted by on
The majority of my private data are storied in subversion repositories. This is not because I write a lot of code, which I don’t. Neither do I have any special need to control different versions of the pictures in my photo gallery. No, in my case it’s simply an attempt to gain the effects of a distributed file system.
While I’ve previously been using SSHFS with pretty good results, it’s no doubt nice having a local cache (working directory) available. You never know when you’ll end up offline or on a slow connection.
Of course, there are lots of different software available if you merely want to sync files back and forth. The benefit of using a version control system is that you really don’t have to worry about unknowingly overwriting files. Besides, I do like the clean server-client model which Subversion provides.
Still, this isn’t exactly what Subversion was made for and, no doubt, there are caveats to be aware of. Myself I’ve had problems handling filenames containing “funny” characters (such as the swedish letters å, ä and ö). Just to be on the safe side I now try to stay in the lower ascii. There’s also the pristine copy, which might give you an unpleasant surprise if you for some reason would ask Subversion to handle something like your multimedia library.
By now I’ve been keeping my documents, etc in Subversion for a couple of months and as a whole I’m more than satisfied with the solution. Besides, it’s a lot easier to setup than your own AFS environment.