Several web-based platforms have emerged to ease the development of interactive or near real-time IoT applications by providing a way to connect things and services together and process the data they emit using a data flow paradigm. While these platforms have been found to be useful on their own, many IoT scenarios require the coordination of computing resources across the network: on servers, gateways and devices themselves. To address this, we explore how to extend existing IoT data flow platforms to create a system suitable for execution on a range of run time environments, toward supporting distributed IoT programs that can be partitioned between servers, gateways and devices. Eventually we aim to automate the distribution of data flows using appropriate distribution mechanism, and optimization heuristics based on participating resource capabilities and constraints imposed by the developer.