I have taken what you might consider low-DRY approach also so it's interesting to see what mechanisms people use to manage reuse without introducing complexity

I have taken what you might consider low-DRY approach also so it's interesting to see what mechanisms people use to manage reuse without introducing complexity into the system itself. I think taking another look at project agnostic generators is something worth doing too. It has the benefits of automating some of the duplication without the rigidity of deep abstractions and dependencies. I am still exploring that though.