Tuple trees provide a convenient and efficient structure for storing conventional relational rows in a prejoined fashion. Schema trees provide an easy, intuitive way for programmers (and even non-technical people) to define a database with fairly complex relationships among the various attributes. Tuple trees have many nice properties: they are easy to build, easy to manipulate, efficient to store and search, and easy to view. Since a tuple tree's leaves can be stored contiguously, locking a set of related rows (one tuple tree) is straightforward.
Since schema and tuple trees are easy to understand, we have found that many people incapable of defining simple databases with standard relational tools can now build fairly complex databases in a matter of minutes.