Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo
If we speak merely about synching, no other features involved, and considering how all operating systems have more than capable tak schedulers (Windows: Task Manager; OS X: LaunchD, cron; Linux: you name it), is there any benefit in functions, stability or whatever in having a resident-in-memory java process that is permanently sucking 40 MB of (real, non-virtual) RAM, instead of just using whatever the OS provides to run sync-now.sh every now and then?
I definitely second that motion!
Would be easier on the memory and probably also easier to configure if the system does the scheduling.