Abstract
Spatial databases are becoming more important in decision and analysis processes. Currently there are very few that allow for spatial data to be truly distributed among many computers without utilizing a central repository for information. Users can benefit from distribution via faster retrieval of local data and localized editing control. Because the same spatial object can appear in several distributed data sets there are several problems that must be addressed stemming from the uniqueness of spatial data. Among these is the need to develop an architecture to manage distribution and propagate editing updates to the same object when it occurs in several separate sets of data on different computers. Distribution also implies data transmission which needs low overhead, specialized methods for meeting authentication requirements on spatial data. There are few is any existing systems that have investigated these issues in the context of application to spatial data. In this paper a new architectural model supporting solutions to the above mentioned problems is presented.