What is Intentional Programming?

BANGALORE, INDIA: Programmers spend days understanding the requirements and needs for building a software solution for an organization. They interact with business domain experts like financial analysts, HR personals, or marketing managers to understand what they need and try figuring out the specific business functionalities required to build the software solution.

Despite all these efforts, the software solution doesn't come out as desired by the domain experts, mainly due to the fact that the programmers have to convert the domain expert's knowledge into their own programmatical knowledge to build the software.

Now suppose the domain experts have the liberty to design the software solution as per their needs and the software solution reflecting the exact intentions of the domain expert who designed the solution, then the whole concept of Intentional Programming (IP) is achieved. With IP, the domain expert can use his expertise in form of graphical representations to define the logic of the tasks for the application.

What is Intentional Programming?
Intentional Programming was a research project led by Charles Simonyi at Microsoft, until early 2001. Later Simonyi found his own startup company 'Intentional Software Corporation' in year 2002, which was focused on making software based on his concept of IP.

IP at Work
As we are already familiar with 'What You See Is What You Get' (WYSIWYG)-based software like text editors, GUI designers among others where the WYSIWYG software lets the computer users to manipulate and design the appearance on screen without having to master the underlying code for the software. On same principles, the IP facilitates the computer user to express needs in his own familiar language using the domain workbench, then allowing the comprehensible views of the emerging design before assembling the executable code. The Unified Modeling Language (UML), developed in mid 1990s at Rational Software, too relied on philosophy of graphical representations to represent a program's function, structure, and behavior. Unlike UML diagrams that cannot be converted to finished products, the IP-generated representations can be transformed to finished products.

Intentional Programming is similar in concept to WYSIWYG processing programs. It's a collection of concepts that enable software's source code to reflect the precise information termed as 'intention', which a programmer had in mind while building any software solution.

The IP process starts when the programmer defines a proposed application's intended functions through a WYSIWYG format interface. After that an automated system utilizes this information for generating the final software product. The WYSIWYG interface that the programmer uses in the IP concept is called as domain workbench.

The domain experts are presented with an interface where they put along the various components to represent the detailed functioning processes of the software application, as per their intents. And this 'domain code' created by domain experts using the domain workbench is like a flowchart that represents how the actual user screen will look like and what is the formal logic associated with it. This way the domain experts are able to conceptualize their intentions into real-time applications that the business users will be accessing.

The programmers now come into picture who will, depending on the 'domain code', create the program generators that will be used for manufacturing the actual target code that the computer will be able to compile and execute. If the final software product does not perform as per the requirements, the programmers can again work with the domain experts on the domain workbench for redesigning of the software. This way the domain experts and the programmers can fiddle with their visualizations of the application until the desired domain code is arrived upon for generating the final target code for the application.

Advantages of IP
Intentional Programming has few great advantages over conventional programming. One is that the people who design the program are those who have the complete understanding of the task and its functioning that is needed to be automated. Thus, there is no mismatch of functioning of the product when the programmers come out with the final product. Second, the design and the layout of the app can be manipulated simply and easily. This eliminates the scenario where in a simple change for text-based code, one had to go through thousands of lines of codes. But as in IP, every task is represented in graphical manner, the change in part gets reflected to whole of the program. Also the programmers have to focus on application generator part only to build the executables for the application, resulting in reduced bugs and errors in the app. Thus, with IP, the programmers will be able to focus more on detailing of meeting the computer users' intentions rather than meeting and coding the computer hardware requirements.

With its advantages, IP has supporters in the software industry as it addresses real and critical issues being faced by the programmers. But at Intentional Softwares, Simonyi is in no hurry, as of now, to come out with a product, but has showcased some demos of the concept and that does show that the future has some innovative programming change to witness alongwith narrowing the rift between the programmer's understanding and domain expert's requirements.