Wednesday, April 7, 2010

Information intelligence

This section describes the last core construct of Secure Natural Communications – the intelligent network architecture that can fulfill and provision the user’s intentions. After the decode engine provided the interpretation of the user’s intentions (and assuming that the core infrastructure has enough capacity to perform these intentions), what remains is for the infrastructure to securely identify certain resources and assign them to the tasks at hand. Many of these issues are standard for designing any network and were addressed in some form for the Internet and Web. However the scale of the set of resources involved results in different requirements. Hence these topics need to be revisited.

Sample information intelligence research issues.

Naming and addressing. This is the first requirement of any infrastructure design. How many elements will need to be addressed (think about all of the sensors, receptors, displays, etc.)? The number is easily in the billions. To support group communication, how will we deal with the exponential number of possible groups? Structure of the address space? Support for mobility?

Location services. Whether via directories or search, the infrastructure must have a method whereby participants in secure natural communications can find each other.

Application programming interfaces. Secure natural communications must provide a standard means for people to get access to its capabilities. These are standard mechanisms to request capacity, share documents, connect to people, apply privacy and security controls, and manage resources, reflected as a set of Application Programming Interfaces.

Security and privacy. As the infrastructure provides more services it also learns more information about people. With more personal information on-line there is greater opportunity for misuse. Social media provides exposures. These issues are additive to today’s unsolved core security problems with the Internet. As we imagine the future infrastructure, we must build in security mechanisms. These include greater use of encryption, privacy hierarchies, and placing greater controls on anonymity – to help discover mischief.

Group management. Group communication is popular and infrastructure support is required. First group communications was popularized with shared workspace solutions – notably Microsoft’s Sharepoint. Blogs, twitter, Facebook are examples of social media which require support for groups to communicate in an ad hoc fashion, share information, and/or multicast. There are discussion groups on the Internet and teams of people from many companies collaborating on a single, brief business transaction. Core infrastructure management must provide support for addressing and locating these groups; mechanisms to track their discussion threads; facilities for people to join and leave and learn the history of the group, and lightweight but adequate controls for security and privacy.

Resource management. The role of an operating system is to manage resources in a single computer. The role of the traditional Network Operations System was to manage all resources in a telephone network. Today, the same infrastructure manages both data (IT) resources as well as telecommunications resources. Moreover, the fact that workloads are more portable - due to cloud computing, virtualization, and virtual appliances - implies that the traditional boundaries between management in a computer (Operating System) and management in a network (Network Operations System) is blurred. Hence, secure natural communications, by fusing together all resources and all applications must develop a universal approach to manage all resources. These resources must be supported across a diversity of operating environments, security needs, virtualization, and leveraging both public and private resources.


  1. Is the Secure Natural Communication Initiatve dead?

  2. Chiku,

    No the initiative is not dead.

    I wanted to get the stream of consciousness out of my head and into a blog. That part is complete.

    I'm talking about it in various conferences. Most recently was today, in SwSTE 2010 in Israel.

    Still contemplating next steps. Probably need a research grant to address some of the technical problems we have been discussing.