The other day I was looking at a proposal K had written for an upcoming conference, and - as I have done many times in the past - I saved a new version of the proposal with my comments, adding a version number onto the end of the filename. I've seen this done in the past; when K and Stan, a colleague of ours, have worked on a paper together, they update the document's filename with a new version number: v1, v3, v12, etc. It slowly piles up, and you're left with 15-20 documents in a folder, all with different content, showing a messy evolution of the proposal. I understand the desire to create new versions - as an academic, you never want to throw away an idea, even if it's edited out. But surely there has to be a better way than creating dozens of duplicate documents.

As a web developer who has also worked in the software industry, I am familiar with versioning systems. I use Subversion at my current job, and used Visual Source Safe in a previous position. Software developers have the same concerns as academics - you never want to throw out a piece of useful code, even if it isn't being used in the application you're currently working on. Likewise, you want to be able to go back in time to revert to an older copy of your code in case something goes wrong with it.

As I mentioned above, I currently make use of Subversion. One of the reasons is that it's open source (and therefore more awesomer). But the main reason is that it seamlessly integrates into Windows Explorer, which makes committing changes (checking in) all the simpler. I can also, using Explorer, view the history of a particular file, and revert to a previous version if necessary (TortoiseSVN, the Explorer integration portion of Subversion, allows you to enter comments when committing a file; I generally note down what has changed since the last commit). I can also see, at a quick glance, what files - if any - have been changed since the last commit, as shown below.

So academics and developers have similar concerns - they want to preserve the content they generate, and be able to go back to a previous version if needed. So the question is, why aren't more academics looking into version control systems for their documents? Is it an issue of technophobia with regards to installing the software, creating a repository, and checking in/out documents? Or is it an issue of the software required honestly being too complex for the "average" academic? How do you version your documents, and why? And if you're a Mac user, how does Time Machine answer this problem for you?