Frequently, the structure and description of the data in a database bears little resemblance to the structure and description of data as it appears in the tool that captured it. This makes it difficult for users to write queries because they receive little information from the database schema regarding the precise meaning of the data. We assert that the semantics of data can be more reliably understood by viewing the data in the context of the user interface (UI) of the software tool used to enter the data rather than the bare framework of a database. GUAVA (GUi As View) presents a conceptual model that captures information about user interface components. In this paper, we describe how to model a forms-based UI using a GUAVA-tree (g-tree), which can be used to generate a natural schema against which querying is simple. We then introduce and formalize the notion of a channel of database transformation operators from the natural schema to the underlying physical schema.