In the 1990's, the only people using the software tended to be people capable of debugging the problems. Now due to maturity, code complexity, and the simple fa

In the 1990's, the only people using the software tended to be people capable of debugging the problems. Now due to maturity, code complexity, and the simple fact that there are a lot of "users" means that its a lot bigger deal if a random patch breaks something. Its the result of the success of the project more than anything. Now the maintainers have to start acting like real developers and do the difficult/unsexy parts of maintaining something rather than just hacking at the fun bits. It really points out that much of what people considered the strengths of the opensource model were illusions that evaporated when the ratio of users/developers got very large. That doesn't mean its not a valid model, only that its a mistake to think a project can both be successful and popular while simultaneously expecting everything to be done by volunteers.