VERSION CONTROL FOR CONTINUOUS DELIVERY

Continuous delivery depends on the ability to deliver reliable software releases through build, test and deployment automation. This is often easier said than done, as it depends on a lot of hard work building a robust infrastructure to support the flow from development to production. One of these crucial tasks is to design a reliable, flexible, easy-to-use and easy-to-comprehend strategy for version control.

I have experienced many times that teams are not able to put new features, or even bug-fixes, into production within a reasonable amount of time (i. e. less than a day). The reason is often that their code base is unstable because they don't have a reasonable branching-, merging- and release strategy.

I will in this talk present an easy and comprehendible version control strategy that allows team members to develop a shared understanding of the branching and releasing processes. Teams not practicing continuous delivery will also find this model useful to avoid "merging hell", unstable code bases and the like. In the examples I give I will use Git, but any version control system supporting branching and merging, like SVN or CVS, could be used to implement the strategy.