The Paradox of Information Access

A significant contribution of the embedded systems research community to a broad spectrum of modern-day applications has been the attainment of dependability of various technological artifacts in the face of increasing unknowns. The term “dependability” here is used broadly to mean assurances on freedom from unwanted behavior. For instance, research on temporal guarantees offered solutions for satisfaction of time constraints of increasing complexity in the presence of different sources of uncertainty. Research on formal methods and model checking ensured that software behavior was trusted in increasingly unpredictable environments. While landmark advances were made in the areas of timing dependability and software dependability, today, the world is plagued with a new dependability problem that remains unsolved. Namely, we lack dependability of information.

Can wisdom accumulated over decades of embedded systems research, addressing various dependability challenges in engineered systems, be leveraged to redesign the world’s information delivery architecture? Significant consequences are at stake. While incorrect timing and unreliable software may lead to fatal catastrophes in safety-critical systems, it is argued that unreliable information may undermine the foundations of entire democracies. Democracy is based on the assumption that constituents are well-informed. If that assumption is violated due to an information dependability problem, the foundations of democracy itself may be jeopardized.

One traditional solution to ensure data dependability in engineered systems has been to leverage redundancy. With more data available from multiple sources (e.g., from redundant sensors in an IoT application), outliers may be eliminated and correct data identified. Unfortunately, the same observation does not work well for systems where data comes from human sources. On the surface, accounts suggest that when a crowd is asked to estimate an unknown value (say the price of a product or the weight of a bull), the median answer is often very close to the truth. This phenomenon, however, requires that individual biases be independent. In turn, independence (in a statistical sense) favors lack of communication (because communication may exert influence and thus result in error correlation).

It is this last observation that sheds light on why improved access to information in fact exacerbates information dependability challenges. It is well known that dependability suffers when errors can have large correlations. Evidence suggests that ease of access creates a situation where alternative individually coherent and largely internally corroborated versions of information emerge that make it harder to reject the false ones. The phenomenon is explained by two consequences of increased ease of access: (i) a tendency for bias synchronization as a result of local interaction, and (ii) a global overload leading to partitioning and global divergence. This phenomenon is termed the paradox of information access.

The idea that increased access creates fragmentation into isolated internally well “synchronized” but externally divergent clusters is not new. For example, historic accounts suggest that the creation of the Interstate Highway System in the US increased societal fragmentation and geographic polarization in metropolitan areas. This is because the resulting increased ease of commute facilitated resettlement into more isolated and locally homogeneous geographic neighborhoods. As cities sprawled, different neighborhoods acquired significantly divergent characters across geographic locales, accentuating various socioeconomic and ethnic divides. This is an instance of local synchronization and global divergence among the locally synchronized clusters.

Similar observations apply to information access. The creation of the Internet, the Web, and then social media created the information superhighway that greatly facilitated global information access. Individuals found it easier to connect with others of similar bias. This resulted in local synchronization, producing what is often referred to as echo-chambers. Moreover, since an individual’s cognitive capacity is fixed, as the volume of online information increases, each individual consumes a vanishingly smaller fraction of all posted public content. With preference for information that supports their local bias, information supporting other beliefs increasingly gets filtered out. The very channel that makes information sharing ubiquitous therefore partitions the whole into sets of locally synchronized echo-chambers that become globally isolated and divergent, thereby creating societal fragmentation and ideological polarization.

From lessons in system design, embedded systems researchers know that common modes of failure (across components that lack diversity) reduce overall system dependability. Similarly, the homogeneous nature of individual echo-chambers that emerge from the paradox of information access makes them more susceptible to misinformation spread. In turn, misinformation contributes (often by design) to increased polarization and bias, producing a positive feedback loop. The cycle magnifies misinformation and bias exacerbating the information dependability problem. Since affordances of connectivity and ease of access are here to stay, how does one solve the information dependability problem, when it is created by increased connectivity and ease of access itself?

The author believes that the embedded systems research community has a lot to offer towards resolving the above problem. With decades of experience in addressing tough dependability problems in challenging contexts, this community has the collective know-how to identify innovative modeling and solution avenues, leveraging a wealth of ideas from a long and productive history. In collaboration with social psychologists, information scientists and embedded system engineers can redesign the information fabric of the 21^st century, producing an information service architecture that frees humanity from the effects of the paradox of information access, restores dependability and neutrality of information coverage to which societies are exposed, and frees individuals to make more informed choices, thereby saving democracy from today’s escalating information challenge.

Author bio: Tarek Abdelzaher received his Ph.D. in Computer Science from the University of Michigan in 1999. He is currently a Professor and Willett Faculty Scholar at the Department of Computer Science, the University of Illinois at Urbana Champaign. He has authored/coauthored more than 300 refereed publications in real-time computing, distributed systems, sensor networks, and control. He served as an Editor-in-Chief of the Journal of Real-Time Systems, and has served as Associate Editor of the IEEE Transactions on Mobile Computing, IEEE Transactions on Parallel and Distributed Systems, IEEE Embedded Systems Letters, the ACM Transaction on Sensor Networks, ACM Transactions on Internet Technology, ACM Transactions on Internet of Things, and the Ad Hoc Networks Journal. He chaired (as Program or General Chair) several conferences in his area including RTAS, RTSS, IPSN, Sensys, Infocom, MASS, SECON, DCoSS, ICDCS, and ICAC. Abdelzaher’s research interests lie broadly in understanding and influencing performance and temporal properties of networked embedded, social and software systems in the face of increasing complexity, distribution, and degree of interaction with an external physical environment. Tarek Abdelzaher is a recipient of the IEEE Outstanding Technical Achievement and Leadership Award in Real-time Systems (2012), the Xerox Award for Faculty Research (2011), as well as over ten best paper awards. He is a senior member of IEEE and a fellow of ACM.

Disclaimer:  Any views or opinions represented in this blog are personal, belong solely to the blog post authors and do not represent those of ACM SIGBED or its parent organization, ACM.