Keynote 2. A Transformative Internet?

In public discourse, big data is often discussed in terms of "meta data." In Internet of Things (IoT), there is a similar concept: meta-ilities, and in particular, (1) Scalability, and (2) Heterogeneity. These 2 meta-ilities are unique due to complexity and diversity, and exemplify why IoT is different than past efforts with distributed systems, communication networks, and systems-of-systems. This is ultimately a sensors and algorithms problem with abundant wireless security concerns. Note that all of the classical "ilities" still apply (i.e., reliability, performance, fault tolerance), but they are sub-ilities to the meta-ilities. The tenets of IoT are: (1) Things communicate, (2) Things may sense, (3) Things should be physical, (4) Communications are mostly wireless due to scale and limitations on wired infrastructure, (5) On board algorithms and software implementations own and control sensor I/O, (6) Things are likely heterogeneous. So this begs questions such as: (1) Are you a thing or human? (2) Who are you if you are human? (3) Where is where (geo-location)? (4) When is when (time is tamperable)? These all play into the final IoT trust story. The key points dissolve down to the fact that truth is not malleable, trust is, and in IoT, that malleability is concerning. It also begs the question as to whether we need an IoT separated from an Internet of Humans (IoH). Based on the above, we believe: (1) Trust in IoT is minimally a function of: wireless, security, privacy, sensors, algorithms, interfaces, and interoperability, (2) Scalability makes this a highly complex issue with even few nodes/hops/branches, (3) Heterogeneity exacerbates the interoperability problem, and (4) Privacy falls victim to sensors and wireless communications. If time remains, I\´ll discuss the supposed relationship between IoT and neuromorphic computing.