The development of the 3270 coincided with, and in part caused, the explosive growth of online transaction processing that began in the late 1960s. Consequently, the 3270 was a major influence in the design of transaction processing systems such as CICS®.
The earliest terminal devices for online processing were adaptations of the teletype, the original and most basic computer terminal. Output was typed, and structure in the input typed by the operator was determined entirely by program convention, without any assists from the hardware. Cathode-ray tube terminals brought a revolutionary improvement in output speed, allowing a complexity of application not previously possible, but formatting on early CRTs was not much more sophisticated than on their hard-copy predecessors.
The 3270 transformed the user interface by introducing the concept of fields on a display screen. Each field on the screen has a starting position and individual attributes, such as display intensity, color, and whether or not you can key data into it. Fields introduce structure into the communication between program and terminal operator in the same way that fields in a file record provide structure for interaction between programs and data.
Organizing a screen display into fields has many advantages:
The advent of personal computers (PCs) and intelligent workstations brought a second revolution in terminal display function. These terminals differ from 3270s in two important respects:
In contrast, a 3270 screen is divided into an array of character positions, typically 24 down and 80 across. Each position consists of an array of raster points, but you cannot address them individually. You can only select a character, from a set of about 190 choices, for each position. Some terminals allow you to select from several character sets and to load new sets, allowing a rudimentary form of graphics, but essentially you are working with a terminal that displays text, numbers and symbols. You get some control of how the characters are displayed, but the choices are very limited in comparison with a PC display.
These characteristics make possible a much higher-function end-user interface than that of the 3270. You can draw pictures, select from a variety of fonts, scale images in size, and so on. If you are writing a new application, and all of your users access it from such terminals, you may want to take advantage of this function to create the most efficient end-user interface possible for your application.
CICS cannot provide this type of function directly, but it does provide a number of ways for a task to communicate with a workstation, so that you can use a software package tailored for your particular workstation in combination with CICS. One popular approach is to use one of these packages, executing on the PC, to build your screens and handle the interactions with your user--that is, to implement the "front end" of your application. This code can then communicate with the part of your application that does the actual processing--the "back end" or "business logic" part--executing under CICS on the host. Communication between the two parts of the application can be done in several ways, depending on what your workstation supports:
When you do this, you can execute specific commands on the host (file operations, for example), or whole programs, or whole tasks. Executing commands remotely is called function shipping, executing a program remotely is called a distributed program link, and executing the whole task remotely is called transaction routing. See the CICS Intercommunication Guide for a full discussion of the possibilities, and the CICS Distributed Transaction Programming Guide for implementation details.
If some of your users have 3270s or other nonprogrammable terminals, on the other hand, or if you are modifying an existing 3270 application, you need to use either terminal control or BMS commands.
Although there is a different programming interface for a PC display, you can use PCs as "3270" terminals. Almost all PCs have programs available that emulate a 3270. These programs convert output in 3270 data stream format into the set of PC instructions that produces the same display on the screen, and similarly convert keyboard input into the form that would have come from a 3270 with the same screen contents.
Under an emulator, the PC display has essentially the same level of function as a real 3270. This limits your access to the more powerful PC hardware, although an emulator program often gives you a means to switch easily from its control to other programs that use the display in full function mode. Moreover, the hardware on a particular PC does not always permit exact duplication of 3270 function (the keyboard may be different, for example). Consequently, your PC may not always behave precisely as described in this chapter or in the IBM® 3270 Information Display System Data Stream Programmer’s Reference manual, although the differences are usually minor.
[[ Contents Previous Page | Next Page Index ]]