Foreign language translation and transcription in the U.S. federal government has changed dramatically over the last fifteen years. The introduction of shared on-line foreign language dictionaries via large, fast databases on large area networks has increased the speed, accuracy, and consistency of the work while also drawing the federal language analysts closer together in technical communities of common interest that share knowledge at the speed of light. The introduction of UNICODE UTF8 served to speed and simplify this process greatly as a technical common denominator that handles all the world's languages equally well in database searches that power federal multilingual processing to new efficiency levels. Foreign language dictionary databases have revolutionized federal translation and transcription; they promise even more improvement over the next few years as these database tools are pushed out and shared further than ever before. The development, use, and expansion of the Chinese dictionary database is only one example of the beginning of this process; but it illustrates some of the best practices in meeting the problems and challenges besetting the users, developers, and administrators of these large systems as the U.S. federal government builds a multilingual infrastructure based on UNICODE.