We envision using the DQI algorithms in tandem with a query previewer [3] that allows the user to browse a huge database and select a manageably small subset to scan. Once the user selects such a subset, the query previewer passes its bounding rectangle to the DQI, which then takes control. The bounding rectangle for the active subset determines the extreme values for each attribute. Therefore, the query preview approach can also be considered as an application of the incremental querying paradigm (in a more broad sense). If, at some later time, the user wants to look outside the active subset, then the simplest solution is for the DQI to return control to the query previewer. This will be considered in a future paper on the interaction between the DQI and the query previewer. The data structures and algorithms are given in the following section with an example on range sliders.