> I haven't heard any arguments for why including SLF4J wouldn't work or poses > a real problem that are compelling to me.
This whole thread has never been about something that just "works" for me. It is about software engineering principles: low coupling, achieving resilience to change over time, abstraction from uncontrolled dependencies, and making lives as simple as humanly possible for end-users. JSecurity has never been about producing something that just works. It is about producing something that is ridiculously easy to use, understand, and deploy, yet still be greatly flexible in almost any environment, regardless of the labor that imposes on the development team. If we don't maintain that philosophy, we could just squash the project and let people build their own or use the insanely crappy and difficult to understand frameworks that already exist out there. The last two days have been painful to say the least, but I don't care - I'd like to think that we maintain a scrutiny of quality better than just "good enough" or "it works".
