This paper examines an emerging technology known as active databases.  The
focus is on issues which need to be considered in their development and the way
they have been addressed in currently implemented systems.   We examine what
defines active behavior, why it is useful, the semantics of its specification 
and choices that have to be made when realising it in an actual system.  We see
that a multitude of options exist when designing an active database and
future research will prove to be both necessary and fruitful.