Describing the linked requirement as "no double entry" might be clearer. The demographics table within AC would be used as the data source for the PM module. The primary key for a patient would be the same in both systems. Whenever a new patient or existing patient information is changed in AC or the PM side, it is reflected in both sides as the underlying table is the same.
Probably the best example of a shared resource edited and read by multiple software packages might be LDAP and the
x.500 directory services. Multiple applications across an enterprise can use LDAP for authentication or for user lookup, all from the same data source. As long as the directory keeps to a standard, the various apps can plug in and play together. Have to be the same if any AC demographic tables were opened up in an SDK arrangement. The standard tables & fields exposed would need to be versioned & managed by AC team to make it work (which is being done already anyway with the 3.x, 4.x and 5.x releases).
