3.0 quite literally nearly bankrupted LibLime, I kid you not, and it was precisely because we spent so much energy making sure that 3.0 was truly stable, and accommodated as much as could be preserved from past versions. It's shocking to me that you're accusing us of the thing we worked so hard to prevent.
I'll add I can well believe this - being RM costs a lot for the organisation doing it. The comments about things not being committed back - from my POV anything not done by us would have been because we ran out of resources to do it rather than any desire to hold some code as "private". How to maintain major bits of functionality that only a few orgs use is a dilema - we ran out $$ to do it for both versions of acquisitions, although now there may be more who want it, and some of the core stuff has stabilised, so it is less of a moving target. Cheers Rachel