| Choosing which naming method to use depends on your environment and on your priorities. Each naming method carries with it a certain set of advantages and disadvantages, summarized in Table 2.1. Table 2.1. Comparison of Net8 Naming Methods | Naming Method | Advantages | Disadvantages | | Local naming | Simple to understand and implement; no need to configure and support a Names server; can take full advantage of Net8 features. | No central control; changes need to be propagated to each client machine. | | Directory naming | Central control; very little client configuration needed. | Resources are needed to configure and manage directory servers. | | Centralized naming (Oracle Names) | Central control; little to no client configuration necessary. | Resources are needed to configure and manage Names servers. | | Host naming | Simple to set up; no client configuration needed. | Only works under TCP/IP; no access to advanced Net8 features. | | External naming | Leverages existing name resolution services; provides central control. | Ties you to a specific external name resolution service; not widely used; may be difficult to configure and manage. | The naming method is configured at the client level using the NAMES.DIRECTORY_PATH parameter in the sqlnet.ora file. For example, the following setting causes Oracle Names to be used: NAMES.DIRECTORY_PATH= (ONAMES) It's entirely possible to use more than one naming method. For example, a developer may want to attempt name resolution using a local tnsnames.ora file before querying a Names server. Net8 gives you complete control over which naming methods to use, and in which order. The default order is shown by the following setting: NAMES.DIRECTORY_PATH= (TNSNAMES, ONAMES, HOSTNAME) The default is to try tnsnames.ora first, then Oracle Names, and lastly the hostname method. Chapter 3, talks more about the NAMES.DIRECTORY_PATH parameter and about some of the things to think about when you set it.  | 
