People like Dieter, Holger, Daniel and I have invested lots of our spare time in the early years. So funding for that came from other paid work, and Osmocom projects were a hobby. It was a lot of fun, but it’s not a sustainable basis for the scope of the projects at this time.
Today, our development is driven by
This means in general that very little will happen, unless somebody commits to putting resources at it!
→ typical 2G setup will become OsmoBSC + OsmoMSC + OsmoHLR
Signal Transfer Point, part of libosmo-sccp
structured approach to state machines
structured approach towards primitives at SAP between layers
Generalized finite state machine (FSM) abstraction
Much more maintainable and deterministic than the many implicit or non-existing state machines in older code
More osmo_fsm Would greatly improve code quality, testability and maintainability
→ GOAL: Daily functional testing on all supported platforms/configs
TODO:
⇒ Roadmap is to put more effort into CTRL completeness
Big Wishlist item:
But who will fund this / put resources at it?
→ we need to generate more alarms in abnormal situations
→ we need aggregation/interpretation/analysis of that data
End of File