Everything Old is New Again
In the late eighties, the PC revolution came to the office. Initially the PC was just another terminal: although it was capable of displaying some graphics to the user, interaction was still basically limited to CRUD screens. But sophisticated client applications such as DTP and CAD packages began to transform the PC into a full-fledged workstation. Advances in graphical user interfaces further enhanced the interactive capabilities of PCs.
But as there was still a need for multiple users to access centralized data, client-server applications emerged that sought to bring together the best of both worlds: rich client interaction coupled with centralized data storage and processing. Client-server applications were not without their pitfalls, however. Upgrades to the application were no longer centralized, but instead had to be coordinated among all clients in addition to the server, and configuration of the clients could often be troublesome.
In the late nineties, the world-wide web came to business (or business came to the world-wide web), and neither would ever be the same. Although even the earliest incarnations of HTML and HTTP (the foundations of the Web) contained facilities for users to submit information to web servers, the web prior to this time was primarily a publication and information-dissemination medium. Business changed all that, and in addition to simply publishing advertising materials on the web began creating interactive catalogs. "E-commerce" became a business model.
But when one steps back a bit, one realizes that these early web applications bore more than a passing resemblance to the text-only CRUD displays. The web browser afforded users the ability to enter and submit forms which were processed by the server. Today’s Java applets and Flash games are analogous to the early client-only applications. Zuma(tm) is the new solitaire.
Where the web is headed now is for a rebirth of the client-server application.