From my PhD research experience, I can see that researchers sometimes use some of these terminologies without any clear restrictions. I see that they refer to distributed intelligent systems that contain a lot of architectural challenges and many conflicting quality features orbiting around context-sensitivity and adaptability. What do you think?