I believe that if we expect people to, for example, collaborate on software systems, we need the best human factors people we can find to be on, and have an equally respected voice on, the design, development, implementation, test and evaluation teams. In addition to human factors people, we also need to open the teams to lots of end users – novice as well as experts – also with an equally respected voice on the teams. Also, we need to test and evaluate our systems on people representative of all levels of computer expertise, again from novices to expert users. We need systems that are as intuitive as a Google search, rather than requiring end users to take classes on how to use, for example, a collaborative system complete with a 150-page manual on how to use the system. I’ve evaluated many a system for usability, only to find that people desparately want to use the system, but cannot get beyond the first few pages of instructions, much less remember those instructions. In testing and evaluating a collaborative system, we need to use several methods that incorporate both subjective and objective metrics to find out if the system is more effective, efficient, and satisfying than what people use now to collaborate (e.g., the phone, videoconferencing, e-mailing, face-to-face meetings, etc), If the new system is no more effective, efficient, and satisfying that what people use now, we should not invest in or expect people to use the new system.