Allen Newell had a long and rich scientific career that contributed to multiple subdisciplines in computer science. Still, each individual endeavor was pursued with a characteristic style that his colleagues, students, and friends recognized as essential to Allen. The criteria by which a research project is judged is predicated, foremost, on the belief that a good idea is not enough. The qualities that transform a good idea into good science can be captured in three maxims (attributable to Allen in spirit, if not in words): Good science responds to real phenomena or real problems. Good scientific problems are those that reflect real phenomena in the world or real problems that need to be solved. A candidate project should include a clear statement of the real phenomena or application it addresses. Good science is in the details. A scientific solution to a problem takes the form of a working model that produces observable behavior or empirical data that can be compared and contrasted to the real phenomena. A candidate project should take the form of a detailed analysis or implemented computational model and should include a clear statement of the scope of its performance. Good science makes a difference. The measure of the contribution of a piece of work is in how it solves real problems or shapes the work of others. A candidate project should include a clear statement of future work that is facilitated by the research presented.