How does maven's declarative convention over configuration stack up against scripted solutions like buildr, groovy/ant or scons when it comes to maintenance costs for large projects? Is there anyone out there who is involved with a large complex project who can weigh in on how sustainable maven has been and how much work it has required to keep bit rot at bay?
I have a project with 20+ sub-projects which I'm planning to convert from a home grown maven like system written in shell scripts and ant to maven. It seemed like a no brainer, the system's process matched maven's model quite well. Then I read these two pieces and while I disagree with them I need to ask... "how sustainable is maven for large scale projects" http://blog.labnotes.org/2007/04/18/introducing-buildr-or-how-we-cured-our-maven-blues/ http://blog.labnotes.org/2007/05/03/buildr-or-when-ruby-is-faster-than-java/ Scripted solution can be much more concise than declarative, but they can also become very complex very quickly. I'm of the belief that I can use or misuse any technology and create a smoothly running system or a maintenance nightmare. I just want reassurance that maven doesn't have some hidden cost, or how best to avoid the pitfalls if it does. FYI, I already know about the Better Builds with Maven and the sonatype books and am working my way through them. Thanks for the help... -- Peter Kahn [EMAIL PROTECTED] [EMAIL PROTECTED], [EMAIL PROTECTED] http://kahnstipation.blogspot.com | http://analogoustendencies.blogspot.com/ Awareness - Intention - Action
