Currently, the viewer supports grouping in the dataset selection menu by means of dividers. This is helpful, but still of limited use for a large number of datasets. I'm currently deploying a viewer for 276 geographically small datasets grouped by country. In this case, a hierarchical menu, with the individual datasets in country-specific submenus, would be ideal.