Today I’m looking at a new method of Source Code (or Control) Management (SCM). As many people out there, I’ve been developing code for a while, and I’ve used CVS for about as long. As a result, I’ve got used to it’s foibles, and it’s quirks and the various irritations. Which is a little unfair, CVS does a simple job well and above all quickly and fairly un-obtrusively. With the maturity of integration into the various popular environments and tools such as TortoiseCVS life was pretty easy. Although I remember having to spend a fair chunk of time determining the branch and tag approach. My main complaints were limited to the lack of changesets, inability to delete directories and list modules remotely. There were various tools, notably ViewCVS (now ViewVC).
Nowadays, I use SVN, but I’m not impressed 😦 It does have a much better model for changes and changesets, and I can now query the repository remotely. However the integration is not as seemless (it’s really slow with TortoiseSVN) and overall it’s slow, that and I’m not impressed with the tag and trunk artifacts being part of the filesystem.
So for today, I’m going to have a go with Mercurial and see if I can get back to the ease of use I use to enjoy.