The author does not know the first thing about building digital repositories. Maybe that is a strange thing to say, given that he works in a repository development group now, worked on the original DSpace project years ago, and worked on a few repository research projects in between. Given how long he has been around people and projects aiming to "build repositories," and how little confidence he has that everyone knows what it really means to build a repository, the author would guess there are many people who do not know, either. On one hand, sometimes it is OK not to know--if a person has a good goal in mind, such as collecting faculty research or making rare local materials available, the details of how he achieves his goal are less important than regularly measuring against the yardstick itself. The author would bet that there are plenty of projects that do not even have this much clarity driving them. This article discusses several mistakes that a person can make along the way to achieving clarity in determining why one wants a repository.