Traditional paper remains a key medium in many domains of our daily lives. We use handwritten grocery lists, we sign contracts on paper and read printed articles because of the mobile, robust and highly flexible nature of paper. Instead of being replaced by digital systems, paper artifacts coexist with these systems, inevitably leading to the problem of integration: the ”Digital-Physical Gap”. In order to bridge this gap, a new generation of user interfaces successfully employed digital pen technology, giving shape to a new style of interaction: Pen-and-Paper Interaction (PPI).
While most prior PPI systems focused on processing in a classical desktop setting, recent systems focus on processing settings of mobile and ubiquitous computing. To provide support for PPI in applications and systems within these settings, developers need adequate infrastructures and toolkits. Letras provides such tool support. It is a novel toolkit for pen-based interaction, with a focus on the mobile and ubiquitous use of PPI and sharing of processing resources among applications: interaction in mobile and changing environments, input on arbitrary surfaces and simultaneous use of multiple pens becomes easily possible.
Letras is based on the ubiquitous computing middleware MundoCore and leverages a distributed pipeline architecture. Successive processing stages with clearly defined separated interfaces channel digital ink data to applications and transform it appropriately. Letras employs a gray-box framework approach: application developers can construct applications using a set of predefined components which only have to be “plugged together” or developers can choose to extend and customize components if required by the targeted setting. Highly flexible deployment schemes allow for easy adaption to the specific requirements of the environment at hand. Hence, Letras provides an extensible platform for rapid development of ubiquitous PPI based applications, empowering application developers to base on a common infrastructure and processing model.
The project homepage of Letras (including downloads) can be found here.