[edwinnathaniel]

This is part of the reason why I really appreciate m2clipse (Maven Eclipse IDE) ability to download the source code (when available in the repository, which, in Java land, almost always) when I tried to navigate to the Class definition, method implementation, interface or whatnot.

It's very empowering to read the source code because one can learn a lot.

A few more things to add from your points:

To work-around a software, sometime I have to understand how it works internally and given the source code, I don't have to spend a lot of time to reverse-engineer it.

Having said that, sometime the customers do care if it's my fault or IBM faults if we stated clearly in the deliverable that "hey, we're using IBM WebSphere for this...". They would give IBM some time to fix it, or ask us if we can work-around it, or they would bend certain requirements (i.e.: work-around the requirements, not the code).

This is why sometimes, probably a dying trend given the quality of open source software, an outsource project chose software based on the company behind it.