After decades of struggling with the concepts and technology that would facilitate information sharing across the many organizational, jurisdictional and disciplinary boundaries, information sharing has come a long way. When the Director of National Intelligence declared that it was the job of intelligence agencies to make their data discoverable rather than to suppress it[1], and when the Congress dictated the creation of a program manager to implement an information sharing environment involving all levels of government[2], we realized as a nation that the time had come for information sharing to take its rightful place in government and in society. Commercial information sharing had already taken hold, as banks and other institutions had realized that common standards and protocols were essential to do business and the globalization of society and government led us to the same conclusions.
Technology supported this positive development, as concepts such as cloud computing, the eXtensible Markup Language (XML), and service oriented architectures made it possible to create more capabilities for information sharing across boundaries. However, a lot of the progress toward a more sophisticated view of information sharing has come from the analysis of business operations, in which we have discovered that there is no reason to continue the extensive duplication of effort and lack of productivity that characterizes so many of our stovepipe systems and methodologies, regardless of the field.
One result of this new realization is that we have seen the emergence of standards such as the National Information Exchange Model (NIEM)[3] and the methodologies begat by cloud computing and shared services that have led to this progress. No one would argue successfully that we have achieved all of the objectives for information sharing, but it is clear that there has been progress. The awakening of federal, state and local agencies to the deployment of open data portals is but one example of the consequences of this convergence of technology with policy directions toward information sharing.
The problem and the pain associated with this evolutionary progress is felt in the enormous deluge of information that is now presented to us on a constant stream that has often been referred to as the task of drinking out of a fire hose. As individuals, businesses, and government agencies attempt to discern insights from the incredible swarm of data feeds, we realize that it is not really information we seek but knowledge. At every personal and professional level, we wish now that there would be some magic answer to extract the wisdom from the information that is shared with us. In other words, we plead for the capability to implement knowledge sharing at the next (higher) level beyond information sharing. Yet we are ever concerned that the knowledge is based on data transformed into evidence so that the knowledge we want shared is sound and not, as has been popularized, alternative facts.
Although we yearn for a level of abstraction from information to knowledge, the intensity of our search for wisdom requires a discipline to the process of sharing knowledge at this level. Since we have yet to postulate the discipline of extracting knowledge from information, we have also not matured a set of principles let alone standards on which to base the sharing of knowledge. There have yet to be developed standards for the representation of knowledge and for its transmission protocols in a way that automates the exchange of knowledge particularly across boundaries of distinctively separate communities of interest. As an example, we appear to have concluded that a strong collaboration between schools, law enforcement and social workers is essential to preventing active shooter incidents at schools, dealing proactively with the behavior of troubled children, but to assemble and exchange the knowledge of any individual child’s situation and potential for violence still remains a significant challenge.
Sharing knowledge across boundaries of communities of interest, such as between health and human services, is complicated by the language variations of each individual community. It is, for example, very difficult to reconcile all the meanings of the word “case” across health, human services, public safety and justice agencies, just to name a simple example. Attempts to resolve such differences lead to harmonization efforts where constituents from the various collaborating communities must gather to reconcile the difference in meaning and find some common ground. This was the relatively successful model used by the NIEM community to define the core elements that were to be applied across boundaries.
It is likely that machine learning and natural language processing technologies will help solve the problem of language ambiguities across domains but until these technologies are farther along it will fall to human interaction to find common positions on the meaning of terms as well as the forms for knowledge sharing. We are also constrained by the need for human involvement to develop common notions of knowledge sharing formats and methods so as to enable automated exchanges. Some of this work is very basic—determining the elements of what constitutes knowledge which we already know is dependent on context. In quite a few circles today, there is in common use the notion of “situational awareness” which has at its core the concept of knowledge sharing. The question remains, for any given context, of how we should represent the knowledge that must be shared across boundaries in order to provide situational awareness.
It should be obvious that if we are to determine how knowledge, as opposed to just information, can and should be shared, then there are many opportunities to create new ideas, concepts, methodologies and standards for the derivation, representation and dissemination of knowledge. The opportunities for research, design, innovation in technology applications, etc., are endless.
Paul Wormeli, December 2017
[1] https://fas.org/irp/dni/iss.pdf