Another factor is sure the lack of frameworks. For example OrbiterSound and UCGO are not yet updated to 2016 and both are pretty wacky to use there.
An open-source framework for making add-on development easier is still out of sight. The next big question would for example be: What should such a middleware do?
TBH, I would not invest time in open-source frameworks replacing and/or extending on OS, UCGO or UMmu. You will immediately run into licensing and/or versioning problems due to it being a middle-ware (what license to avoid another Vinka, what policy on deploying together with addons?), you have a fragmented "market" with the legacy framework status being in limbo, and you risk your work becoming obsolete if the "true" framework suddenly returns with a new and improved version.
You can mitigate the later point a bit by means of working out the first version behind the scenes, so the original author is not "teased" too much. Place it when ready, then lure legacy users with some shiny new feature. In the end, the peer pressure ("does it have UMmu?") will do its work. Still risky, but might work.
Of course it would be more faithful to win people by arguments like "it is open, so you won't risk these update-problems any more". But in my experience this doesn't work, because folks don't give a dime on the philosophies behind it, they only see the immediate advantage of shiny features. Not that I can blame them for that, mind you.