Network

Defining Trust

What is trust in general terms? Before categorizing people and resources, trust must be defined. Trust is the likelihood that people will act the way you expect them to act. Trust is often based on past experiences. You could also say that trust can exist only between two individuals who know each other. You can never trust a total stranger, but you can start to trust one over a certain period of time. An exception to this rule exists in the context of networking. You might be willing to trust a stranger if you know that someone you trust trusts him. This is, after all, the basis for Secure Sockets Layer (SSL) and certificate exchange, as discussed in Chapter 13, "Public Key Infrastructure."

Now that trust is defined, a list of resources can be developed that ranges from most trusted to least trusted, as shown in Figure 1-1.

Figure 1-1. Security Zones

Most Trusted

The most trusted network resources in an organization are internal servers, domain controllers, and storage devices attached to the network. Only a limited number of well-known people should have access to these devices.

Less Trusted

This category includes the internal users and the remote, authenticated users. On a certain level, an organization has to trust its users, internal or remote, because otherwise these users cannot perform their jobs. Despite the trust granted to them, some people in an organization use the passwords they have to do things they are not supposed to do. Although most employees can be trusted, it is because of the minority that abuses its privileges that this group is categorized as less trusted, not most trusted.

Least Trusted

The least trusted (sometimes referred to as untrusted) resources and users are Internet servers and remote, unauthenticated users. You can never trust an Internet server because you are not sure what is behind it. That is the reason for using digital certificates, which are explained in more detail in Chapter 13.