A simple product though won't likely be used commercially by professionals, and then you are stuck at the low end serving students/dabblers, which is not a place were you will make much money. Remember for software tools, it is usually only those who are making money using your tool who are willing to pay for it.
We are developing an open source 3D system, and what could set it apart from existing tools such as Blender is an interesting question. In our case we are taking a different tack by developing in Common Lisp, which will give us a powerful REPL and whole-system extensibility.
Though the web was a tempting platform, we are keeping it as a desktop app for now. Based on the comments I read here that may have been a wise decision.
We shall see if there is a niche for a system such as ours.
Same issue in progress here! I've been working for a few years on an editor(forked from Three.js editor) for my browser game bad.city and now I'm separating the editor, rewrote it in react and the current goal is making an easy-to-start-with game level editor with many assumptions. We'll see how that goes
