The Windows Sockets specification was created by a group of companies, including Microsoft, in an effort to standardize the TCP/IP suite of protocols under Windows. Prior to Windows Sockets, each vendor developed their own proprietary libraries, and although they all had similar functionality, the differences were significant enough to cause problems for the software developers that used them. The biggest limitation was that, upon choosing to develop against a specific vendor's library, the developer was "locked" into that particular implementation. A program written against one vendor's product would not work with another's. Windows Sockets was offered as a solution, leaving developers and their end-users free to choose any vendor's implementation with the assurance that the product will continue to work.

There are two general approaches that you can take when creating a program that uses Windows Sockets. One is to code directly against the API. The other is to use a component which provides a higher-level interface to the library by setting properties and responding to events. This can provide a more "natural" programming interface, and it allows you to avoid much of the error-prone drudgery commonly associated with sockets programming. By including the control in a project, setting some properties and responding to events, you can quickly and easily write an Internet-enabled application. And because of the nature of custom controls in general, the learning curve is low and experimentation is easy. SocketWrench provides a comprehensive interface to the Windows Sockets library and will be used to build a simple client-server application in the next section of this document. Before we get started with the control, however, we'll cover the basic terminology and concepts behind sockets programming in general.

Copyright © 2008 Catalyst Development Corporation. All rights reserved.