I can give you the case of the lightbulb. It's not a very complicated thing. Metal heats when a current is applied. And there's an evacuated tube around it - if it wasn't evacuated, then the air would heat and it would explode. Evacuated tubes were already well in use, in the form of scientific instruments - Crookes and Geissler tubes. There's a few material concerns, in that many such tubes were individually blown for experimental purposes, and evacuation is a tricky process. Also, they needed a metal which wouldn't simply denature - to even get up into visibile light for a blackbody requires incredible effective temperatures. Scientific instruments, to my knowledge, were using an order of magnitude fewer volts.

And then you have the complex history of the wiring of America. The lightbulb is less sophisticated than the Crookes tube, but its ease of use and ubiquity, alongside simply being so darn bright, distinguish it as an invention.

I'm not an expert, truly. But in general, invention has a lot to do with who has 1) the base technologies needed to build in, 2) the space and time to research, 3) luck, 4) the econopolitical circumstances that affect production and distribution. There are cases of rather ingenius inventions that nobody else would have thought of, but often the principles behind a thing are remarkably simple, and all it takes is the right place and time to hammer out the details. The lightbulb, for example, is terrifically simple. But Edison had to blow up n of them to get the materials right, and even then, the electrification of the world, which was outside his power,was the biggest contributor to his success. That said,two hundred years before his time and nobody even knew enough about electricity or evacuation to make it work.

(EDIT: As an aside, a trained monkey could probably have invented the lightbulb. This isn't to deride Mr.s Edison, Swan, Lodygin, etc., they still did the extensive dirty work. But you put a few million filaments in glass, evacuate them, and run current through the ones which survive, until they burn out. You make a nice chart of these, and if you happen to make decent tests on the carbonized bamboo ones, you'll invent the lightbulb.

This goes for many inventions. Computers are really about getting more transistors together. If you know something about semiconductor chemistry, about relativistic and quantum EM effects, and you have the right tools, then you can make billions of transistors, reliably, very small. Binary logic follows from transistor logic. Everything since those first logic gates has been the result of pounding away at transistors, either making them better or using them to compute. You can follow this same sort of logic in many cases. So many things aren't about genius moments, they're about applying the basics and testing until something sticks.)

You mention violins, but I'd say such things are likely to straddle between invention and art. The violin is somewhat technologically distinct from it's predecessora, but you could also say it's a style of stringed instrument, whose popularity is dependent on its place and time, instead of really being something new.