Metadata are descriptions of data sets that include information such as data quality, keywords, or the spatial or temporal extent of the data set. They provide sufficient information to enable data search, retrieval, transfer, and evaluation. Standardized metadata that can be automatically used by computers without human intervention or interpretation are key components of effective data management, inter-operability between distributed data systems, and, most importantly, integrative, interdisciplinary science. Such standardized metadata will reduce the semantic heterogeneity and syntactic incompatibility that currently hinder data management and inter-operability. Unfortunately, many ocean scientists provide either incomplete metadata to allow for automatic data search, retrieval, transfer, or evaluation, or they provide none at all. For example, scientists often provide descriptive information such as the where, when, and what of the data set, but little about the logical and physical structure of the data set. Also, scientists use different terms as column headings in their data sets, for example, Temp versus Temperature versus Water_Temperature, thus limiting inter-operability.