Time for a revolution
In my opinion, Domino requires a structural revolution, turning folders and views from being internal components of databases to being situated above databases. For simple systems, the database as container is sufficient. However, for both enterprise systems and the enmeshing of small systems into an integrated intranet, this approach is limited. Users don't care about databases, they see views and folders. Lotus has already achieved this with search, supplementing the existing internal database search function with the server-wide Site Search database.

The structure of the databases themselves must be a purely technical matter, to be optimized by an experienced designer or administrator but transparent to the ordinary business users, like the tablespaces of Oracle. Folders within Notes, at present, are simply a collection of document IDs. Extending this to be full IDs (i.e. replica ID + document ID) or even URLs is technically evolutionary but radical for usability.

Directory direction
Another area where database integration is lacking is the Domino Directory. Other vendors (such as Microsoft with Active Directory and Novell with NDS) are developing large, hierarchic directories that contain everything. However, Domino persists in having the handling of its objects split between a Public Address Book (servers, users, domains, users' mail files and mail-in databases), Catalog (all databases except where manually excluded) and Library (manually included---published--databases). Some move in the right direction has been made with R5, where database catalogs become key and not an option for a methodical administrator. This needs to go further, so that there is a consistent manner of addressing any object on a Notes network, and also to shift from server-centric networks to a single, smooth network of objects, which is one of the main functions of a global directory.