The author first studied the board in an effort to understand the implications of the transformation from a board composed of housing authority officials to one staffed by citizen volunteers. A 1987 study examined the changes that had occurred since the first investigation. This reflection on the research confirms the author's uneasiness about court-docket studies. Although plausible theories may fit the data, examples from the Hawaii study show they may have little to do with what actually occurred. To minimize the problems of interpreting docket data, this article advocates a "local-knowledge" approach. This involves understanding the court system as it was understood by those using it rather than imposing "objective" explanations on visible patterns. It also involves understanding the context in which cases arose and were litigated. This requires an appreciation of factors affecting the understandings of those who brought or or failed to bring business to the courts. An alternative approach to obtaining local knowledge is to examine docket data for a period that is sufficiently limited -- 20 or 30 years perhaps -- to allow an indepth examination of contextual forces throughout the entire period. This should produce more detailed information about the interaction between judicial business and the contexts in which courts operate. 15 footnotes.