Following the verification operation of the acquired marker, the sentinel establishes the state of the authentication operation. Once this state is established, the authentication operation is over. The departure stage allows for shutting down this particular protocol and gracefully moving to the next one. As we’ve noted, such protocols can be effected recursively. Indeed, the three approaches to authentication suggested by the NIST document referenced earlier can be applied in repetition; performing any one of them constitutes single factor authentication, while using two or all three constitutes two factor or three factor authentication. Ostensibly, the level of trust established through the authentication operation increases as the number of factors used is increased.

Networks

Computer networks derive from the mathematical reality of pair-wise interactions; the number of resultant connection pathways goes as the square of the number of end-points. For large numbers of computers, it becomes problematic to provide a dedicated physical connection between each pair. The same problem arises in all manner of interaction environments; the streets within cities, roads between cities, sewer systems among homes and offices, voice telephony, radio telephone and so on. The solution for all of these environments is the same; use shared connection pathways with a single connection for each end point. The shared connection pathways allow for real, logical or pseudo-shared access mechanisms.

When a road is built between two cities, the interactions between the cities take the form of traffic between the two. The traffic may be comprised of a person walking between the two cities or a truck load of goods being hauled from one city to the other. If commerce with a third city is desired then a single road can be built from one of the two initial cities to the third. Thus, three cities can be supported using only two roads. One of the cities now has two roads into and out of it; it has a fashion of redundancy. Regarding interaction capabilities, it is in a superior situation to the two cities with only a single road connecting each to the outside world.

A general network is a totally open network, amenable to all applications. While we really can’t list any networks that are purely general, there are several domains which exhibit near-general facilities.

There is for certain one global network at least, the now traditional worldwide network of computers: the Internet. We’re going to look a bit at its fundamentals and the relationships among the computers involved, including trusted platform cores. Then we’ll use that model as a reference for the very important industrial networks that have otherwise developed.

The original computers of the 1950’s were standalone machines, not really thought of as networks of devices by themselves. They had, nevertheless, a central processor unit connected to devices meant to manage the input and output of information to and from the central processing unit. Originally, each link to an input/output device, be it a keyboard, a tape reader, or a printer, was independently conceived. It didn’t take long, however, before an organizing component, called a bus, came along as a way to connect the central processing unit to output devices in a homogeneous manner.

A precursor to the understanding of a standalone computer as a central processing unit connected in a star network with multiple input/output devices was Control Data, under the guidance of Seymour Cray, who would later become the creator of the famous super-computers of the 1980’s bearing his name. In the Control Data model, a standard language of communication (a protocol in

Book available at Midori Press (regular)
Book available at Midori Press (signed)
Book available at Amazon (regular)