In 1999 I worked as a programmer at IBM T. J. Watson Research Center on a project involving aggregating multimedia and structured text data. Potential applications of this system were supposed to be advertising agencies, catalog retailers, stock photo and video providers and similar customers. As part of the project I was invited to co-author a publication for presentation at Engineering Federated Information Systems workshop in 2000.
Since the fall of 1999 there have been many new developments in media asset management technologies so the paper may be quite a bit out of date.