With ever more devices being connected to the Internet and everyday objects 
becoming "smart" due to embedded processors and communication 
capabilities, the provisioning of intuitive user interfaces to control smart 
things is quickly gaining importance. To address this issue, we present a 
model-based interface description scheme that enables automatic, 
modality-independent user interface generation. User interface description 
languages based on our approach carry enough information to suggest 
appropriate and intuitive interfaces. Still, they are simple enough to enable 
developers to describe the interaction semantics of a smart thing using very 
little, easily producible markup. This is enabled by describing the atomic 
interactive components of a device rather than the device as a whole, and 
capturing the high-level semantics of an interaction. As a concrete language 
based on this approach, we propose a taxonomy of abstract sensing and 
actuation primitives and present a smartphone application that can act as a 
ubiquitous device controller. An evaluation of our approach in a laboratory 
setup, home environments, and a lecture hall automation system as well as the 
results of a user study highlight the accessibility of the proposed 
description scheme for application developers, its suitability for controlling 
smart devices, and its generality with respect to describing heterogeneous 
smart things.