I had the same thought once, and I found something called ThreeNode.JS that fit the bill. I took this and added a JavaScript based physics engine on top of one of the existing 3D programming examples. It was easy for me, but it wouldn’t be easy for everyone. I think what you will find as you get really good at programming is that building that crud app is not really “science” but it’s more like information processing, but as you get more advanced you want to do more difficult things and so you go out and you READ, you read a LOT, learn as quickly as you can, and when you’re ready you apply it. If you had a massive library of drag and drop components instead of typing code you’d spend all of your time trying to figure out a faster way to access all that information than clicking and scrolling, like SEARCHING, and at this point you’re back to square one.

If you are truly spending too much time typing then try a higher level language like Ruby/Rails.

I think the holy grail is visual programming by speaking to the computer, using a spoken language with structure and a visual context, so that you can see what you’re building as you speak, then make typing and clicking secondary, maybe even more advanced things like hand gestures or eye tracking.