Development of IoT applications brings a new movement to the functionality of wireless sensor networks (WSNs) from only environment sensing and data gathering to collaborative inferring and ubiquitous intelligence. In intelligent WSNs, nodes collaborate to exchange the information needed to achieve the required inference or smartness. Efficiency or correctness of many smart applications relies on the efficient, timely, reliable, and ubiquitous inference of information. In this paper, we introduce the RUbIn framework which provides a generic solution for such ubiquitous inferences. RUbIn brings the reliability and ubiquity for inferences using the redundancy characteristic of the gossiping protocols. With RUbIn, the implementation of such inferences and the control of their speed and cost is abstracted by providing developers with a proposed middleware and some dissemination control services. We develop a prototype implementation of the RUbIn framework and a few inference examples on TinyOS. For evaluation, we utilize both the TOSSIM simulator and a testbed of MicaZ motes in various densities and different number of nodes. Results of the evaluations demonstrate that in all nodes, the inferring time after a change is about a few seconds and the cost of maintenance in stability state is about a few sends per hour.