Sounds to me like making an out-of-the-box OACS stack installer is a task best suited to a set of RPMs (something that somebody is hopefully thinking about doing once stable releases start getting cut?)... a webmin-like interface could be packaged up as yet another RPM. There are really two different issues here: easy setup of a vanilla server stack and simplified configuration of a running server stack. Both of these tools would be valuable independently, and both projects would benefit from working together (or at least the community would benefit as a result of such a collaboration), but they don't seem like the same project to me.