The various aspects of Open Source

A funny thing that I just realized is that the term "open source" is generally used to denote several types of software. Without actually defining what "open source" means, I am thinking to several categories:
1) Hobby projects. These are projects started just for fun, from someone that writes a cool little app and wants to share it with everyone else.
2) Disruptive projects. Here we have a specific category of projects that are funded and maintained with help from (big) corporations, with the specific intention to disrupt other forms of commercial software (usually their competitors).
3) Pixie-dust projects. Here I put big, commercial projects that were morphed in an open source variation, as in an attempt to sprinkle the open source pixie dust to help them to remain competitive. Examples: Netscape (and probably Solaris?).
4) Sample code. This is a self-explanatory category, which is sometimes not considered "true" open source.
5) Shared source. This is also self-explanatory.

The interesting thing is that these categories are not clearly delimited. In fact, most open-source projects started as (1) and ended up in (2) like in the case of Linux. JBoss is another example.

[update]

By "disruptive" I meant the term explained here: https://en.wikipedia.org/wiki/Disruptive_technology