Over the years Equinox has become a stable project. The last major code effort was in the Luna release when the Equinox framework was refactored to use the OSGi generic capability and requirements model along with the OSGi Resolver service. Over the past few years the list of active committers has dropped to just a handful. We did add two additional committers this past year, but even the voting turnout from existing committers was pretty low.
The Eclipse top-level project itself has also suffered from a drop in active committers. In an effort to consolidate the committer base and get more active committers across the eclipse project the Eclipse PMC has continued to merge their own committer groups so that they can share the burden of maintaining the rather large code base among the set of active committers included in the top-level Eclipse project.
Many years ago the Equinox project was spun out of the Eclipse top-level project to its own project under the top-level RT project. The Eclipse project continued to heavily depend on Equinox for its runtime and the two projects are still very closely tied together with respect to planning and release engineering for the overall releases produced each year. It has become apparent to me that separating Equinox out from under the top-level Eclipse project is not serving in the continued best interests of Equinox nor the Eclipse project as a whole. As Equinox committers, we deal more closely with the Eclipse PMC in planning our release than we do the RT PMC. I think it is more representative of the work done in Equinox to be placed back under the Eclipse project PMC. I propose we do a move review to move the Equinox project back under the Eclipse top-level project.
I do not anticipate this will change much in the way of how we develop and release Equinox today since much of that is already influenced by the Eclipse Project release engineering team. I also do not think this will inhibit our ability to change how Equinox is developed, if we so choose. For example:
- having more frequent releases to maven central
- moving to bndtools for development
- etc.
We should be able to drive changes into Equinox with the same flexibility as we do under the RT top-level project. But instead of being under the leadership of the disparate RT PMC (which is largely be design) we will be back under the leadership of the Eclipse PMC which I believe is more cohesive to the day to day activities in Equinox than the overall RT PMC. Please let me know if you have any concerns or questions about moving Equinox back to the top-level Eclipse project. If nobody objects then I will start to formalize a plan to do the move.
Tom
