Digital libraries and intranets often include a wide mix of document types, ranging from catalog entries consisting entirely of metadata, to documents with both substantial text content and thorough metadata tagging, to documents with hardly any metadata at all. This paper describes a retrieval model for use in such environments which can accommodate searches in a variety of meta-data schemata. It uses both content search and three-valued satisfaction of metadata constraints (yes, no and maybe). In this model queries may include explicit metadata constraints as well as free text terms. The latter may impose additional implicit constraints as well as contributing to content relevance scores. Result lists are presented in tiers, in which the top tier corresponds to full satisfaction of all constraints and subsequent tiers are derived by progressive relaxation of constraints. An eecient and scalable implementation of the model has been developed and has been in production use on the Australian National University intranet since July 1999. Details of the implementation are given and performance results are presented.