Configuration management is complex. As a product evolves, CM gets even more complex, as complexity breeds problems. So how do we continually march our product configurations toward higher and higher quality? You need good people, process, tools, and automation. People define and put the processes in place. Tools are used to automate the processes and to bury the complexity of CM below the user level. Development is a process that takes ideas and turns them into products. There are many techniques and tools to do design and implementation. The right mix is required to effectively develop a competitive product. On the management side of things, the same is true. Application lifecycle management requires the right set of processes and tools to allow the design efforts to be properly leveraged and to yield the best quality.

Building is considered to be one of the fundamentals of configuration management, even though some might argue that building isn't really CM. The reason it is fundamental is that the build/verification cycle provides proof of reproducibility if done correctly. It forces CM to be done correctly so that only the objects from the CM repository are used to reproduce the build. When formal build processes are correct, they need no information that resides outside of a CM repository. When properly done, the build record is created prior to the build (i.e., a build notice) rather than as a result of the build, with that record being used to drive the build process. An integral change management capability is an essential component of such a build process.

There are a lot of CM/ALM solutions out there - so where are we headed? In my opinion, the future of CM/ALM will be defined by the level of tool suite integration, more than by any other factor in the 3rd and 4th Generations of tools. And basic "integration" will not cut it. Putting tools together into a single package with some glue and triggers to help the tools interact is helpful, but will fall short of market demand. "seamless integration" will be a requirement. No advance is more important to the next generation of CM tools. What about cost? What about ease-of-use? How about traceability? The answer is simple: first seamless integration, and the rest will follow. If you haven't seen a seamlessly integrated tool suite, you may not fully grasp this statement. But seamlessly integrated tools are the ones that will be dominant in the future. Let's look at why.

Open Source is a great concept, thmough any resisted at first. Can you remember a time when it was valid to use the argument that, "Nobody will commit to supporting it, so you're taking your chances"? Today, I see it differently. There are loads of support for open source products, even better than for commercial products in many cases. So does that make it the winner?