not a lot of time to reply- but briefly: Blender has a very high ceiling when compared to a lot of other mainstream apps. There is a lot of market consolidation happening (with regard to tools) and as that happens the companies that make those tools are less incentivized to innovate or improve. So while they have a huge head start in terms of reliability and adoption, that lead is unlikely to last without some major push to keep up with Blender’s rate of growth. I think with 2.8, Blender crossed an inflection point in the hearts and minds of professional artists so we’re going to start seeing a snowball effect where more people try it out and stick, some of them contribute to the dev fund or help improve future releases, or tell their colleagues (which perpetuates the cycle).
So while Blender might not be there quite yet, it’s visible on the horizon, meanwhile Autodesk seems to be pulling into the breakdown lane.
For me personally (having a strong technical background), it’s the robust Python API. All of the problems that I mentioned above I was able to solve for the artists at my studio because of a Python API that gives a technical artist or programmer plenty of rope to hang themselves with. Yes, it sucks that any of that work was necessary- but the fact that it could be solved in the first place without having to break open the source code is nothing short of amazing. Combine that with the high growth potential of Blender and you can imagine a lot of new technical types coming to Blender and bringing new functionality with them. Addons like HardOps and MeshMachine are already way ahead of the bleeding edge for modeling- Maya scripts and plugins aren’t able to keep up.
I could go on, but that’s a pretty good summary for the time that I have