A good reality check for the viability of publicly announced research projects such as Jini is to look for other companies pursuing similar projects. While it has not been publicized much by Microsoft, if you dig several layers deep into the Microsoft Research web pages there is talk of a project dubbed Millennium, which promises "a new self-organizing , self-tuning distributed system" where "any code fragment might run anywhere, any data object might live anywhere ." In fact, since the public announcement of Jini, Microsoft has made some attempts to compare Millennium with Jini. Marketing talk aside, there are some very fundamental differences. Communication in Jini is based on the Java RMI (remote method invocation) API. Interestingly enough, the Java RMI classes are part of the core Java specification that Microsoft has chosen not to implement in some versions of its Java Virtual Machine (and is one of the factors in Sun's license infringement lawsuit against Microsoft). At present, there is not enough publicly available material on Millennium to complete a thorough comparison between the two. It's hard to believe, however, that Microsoft would give in at this point and base Millennium on the Java RMI specification. However, given that by the year 2000 interoperability between CORBA and DCOM will have been suitably addressed for most developers (via various vendors ' uni and multidirectional bridges), don't be surprised to see Millennium versus Jini to become one of the new defacto standards battlegrounds.