How deal with user interaction in 3D geovisualizations

This paper is devoted to the user performance within interactive 3D geovisualizations. Most of the usability studies in cartography used static 3D geovisualizations (perspective views) as stimuli or if interaction was possible, it was not even monitored or analysed in detail. Only few studies as Wilkening and Fabrikant (2013), Treves et al. (2015), McKenzie and Klippel (2016) and Jurik et al. (2017) are exceptions. Their approaches and methods used for 3D User Interface evaluation, especially by the screen logging method inspired our work. We designed and used unique web testing tool 3DmoveR (3D Movement and Interaction Recorder) for the collection of data about user interaction and virtual movement. Methods that allow us to study or comparison of user interaction and virtual movement can be divided into two main groups: numerical metrics and visualizations methods. Numerical metrics are calculated from the virtual trajectory (length, average speed), virtual camera positions (average height, rotation characteristics), or totalled from the duration of individual movement types. Visualization methods include virtual trajectories visualization, screen video recording, area of interest (AOI), sequence chart displaying sequences of virtual movement types, height chart of the virtual camera or timeline chart (distance between the virtual camera and selected point). The method is between these two categories is String Edit Distance method calculated by software tool ScanGraph (Dolezalova and Popelka, 2016). All these methods can be classified according to their deployment capabilities: - preservation of spatial component, - preservation of time component, - data aggregation, - possibility for comparison, - dimensionality of movement mode. Examples of the utilization of above mentioned methods are demonstrated on selected examples and discussed in the conclusion.