Arguably, computers and communications have changed the world more than any other science or technology. Yet there are a lot of failures, some prominent, many minor, and a widening gap between aspirations and reality. Borrowing Richard Feynman's criticism of cargo cult science, this article discusses the problem of cargo cult in computer science. To paraphrase Feynman: everywhere I see computer scientists that look like they are working on the right thing, but the thing actually doesn't work. I hope this article could make you reflect the problem of having poor science in computer science research, and what we can positively do about it.