As part of a collaboration with the White House Office of Media Affair
s, members of the Artificial Intelligence Laboratory at the Massachuse
tts Institute of Technology designed a system, called COMLINK, that di
stributes a daily stream of documents released by the Office of Media
Affairs. Approximately 4,000 direct subscribers receive information fr
om this service, but more than 100,000 people receive the information
through redistribution channels. The information is distributed throug
h e-mail and the World Wide Web. In such a large-scale distribution sc
heme, there is a constant problem of subscriptions becoming invalid be
cause the user's e-mail account has terminated. These invalid subscrip
tions cause a backwash of hundreds of bounced-mail messages each day t
hat must be processed by the operators of the COMLINK system. To manag
e this annoying but necessary task, an expert system named BMES was de
veloped to diagnose the failures of information delivery.