The purpose of this lecture is similar to that of lecture 13 or lecture 9, in which we paused and looked at uses for the technologies we've developed in real-world settings originating in the Web Services arena or other similar environments. 

With this in mind, the lecture explores a series of problems that ultimately reduce to data replication.  The slide set isn't especially long, but this is because I go fairly slowly when covering this material -- I happen to know many of the people who worked on problems of the kinds we explore here, and can even comment on the many false starts that the area has known (and there are a LOT of them)!  So the point is as much that with group communication we can reduce these many solutions to a common substrate as not.

But in fact there are some other comments to make.  In exploring these technology uses, one does need to keep asking basic questions: what form of consistency is needed?  How strong an ordering or safety property do we need?  How fast will the application run if we do it this way?  How much concurrency can this solution achieve?  The lecturer must take the time to pose such questions, invite the class to debate the issue (perhaps on a chalkboard!) and ultimately to learn by digging below the surface.