Hello :)
I used to work with svn some years ago (I learned it for programming at first). I mostly then used GUIs with it (TortoiseSVN on windows and kdesvn on linux).
Then I moved to Mercurial (hg) after comparing between different softwares. One of the main reason I chose it over git was the availability of a good GUI on windows
(TortoiseHG works very well and is integrated both with windows explorer and nautilus under linux/gnome). I also sometime use the command line directly, but mostly for administration tasks or scripting.
I frequently use an addon to hg called mercurial-queues (mq) that allow to work on a patch queue. So most of the time, I use patches to try some new things (in scores also), or can switch between different queues depending on which project I work on (each change on patches can be versionned in a secondary repository).
When things get a little more stable, I commit changes to the main repository, and I can push to a server if I'm working on a shared project (I guess hg and git are quite similar in term of possible workflows).
To the question :
> Simon Bailey:
> > so, my follow up question: why use git as a single user?
The main advantage I would see on using hg (but I think it is the same with git) over svn, as a single user, is that :
in svn, you basically (if I'm not mistaken) need both the repository itself (either local or distant) and a local checkout/working copy. If by any chance by mistake you delete the repo folder, then all your history is lost (er... yes, I did it once).
With hg, each clone of a repository IS a complete repository itself. So your "working copy" is a repository itself (the main folder just contains a hidden .hg folder with all the history data). I feel this is an advantage over svn when working locally, as your data and the history are at the same place.
Yann