This is a bit of a rambling post, but stick with me, I think it's an interesting question.

I use a poor-mans versioning system for organising code at the moment. When I get to stage where I want to clean up the code and get ready for the next challenge, I copy the directory, increment the directory number and move on.

Now already I can hear a whole bunch of people crying themselves to sleep as they realise people still do this and don't use tools that were made exactly to solve this problem (cvs, svn, et al). I want to do this, but I've just not gotten around to it yet, and likely, until I stumble into a project of sufficient size to warrent any sort of investment, I probably won't.

But that's not really where I'm going with this piece. I find myself replicating that methodology with the rest of my documents, especially things like my dissertation and my resume.

Just one of the many folders I've got kicking around looks like this (and this is a tidy one ... ):

The date is there, obviously, and the text strings are the initials of my colleagues and supervisors, with the suffix value representing an individual's increment (after they've run through and made a bunch of changes).

Now I'm finding I'm trying to apply similar logic to things I'm doing on my own, but all I'm really doing is ending up with a gazillion different versions of things that, to be honest, I really don't need to keep.

Wouldn't it be great if I just had 1 document and all the revisions hidden until I actually want to get at them? Hmm, smells like versioning ...

So, has anyone ever used or ever investigated the use of a versioning filesystem like ext3cow or cvfs-fuse?

If yes, how did you have to change (if at all) your working habits?

Alternatively, links to help organise my filesystem/life are more than welcome. It's likely the only thing actually wrong is my working method and existing habits

Looking forward to your thoughts!

- Piete.

OMG, so you're saying this .odt thing doesn't include versioning features itself? ;-p


Ext3COW is experimental as far as I know (still haven't gotten round to testing it properly) and you're forgetting FUSE-CopyFS (IIRC that one works).


For configuration files I use a wrapper script and good old RCS, for code that requires versioning I use CVS and for text I use a text editor that has versioning, like OOo.org. For me it's not about what tool or method to use but the discipline to use them.