MDA 2

As a representative of an organization were information systems are vital, I would be very skeptical of MDA, and I would do a lot of research and testing before deciding to, or not to, implement MDA in my company. I would make MDA prove its worth.

I think MDA is good for a company that does a lot of development, and has many programmers on its payroll. The main benefit would be to eliminate the high costs of human programming, an the costs will be higher, and the benefits greater if there are a large amount of programmers who will not be needed after the implementation of MDA. I’ll pretend the company I represent has many programmers.

I feel that MDA is the next big thing for software development. But I’m not sure if I’m ready for it yet. Some of the criticisms of MDA is that it is so new, and not fully developed, so the standards aren’t quite standardized yet. I would be hesitant to implement it, only to have to change major components in the future to conform. Also, not many organizations are using it, and it is relatively new, so it hasn’t been proven to stand the test of time.

I would look very carefully at some of the companies who have successfully, and not so successfully implemented MDA. I would like to find a company similar to mine who works with it and it works for them. I would also to hear the criticisms of MDA from the companies who abandoned it. I would look very carefully at what other companies have to say.

I will look at a company, Navitare, who develops software for airline reservations. They produce code as a core competency. There are many developers, all over the world. And for every developer, there is a business analyst/consultant who is in charge of meeting with customers and gathering requirements definitions and communicating them to developers.

Navitare has been in business long enough, and has experience enough, that they have software development for reservations down. The product sold to airlines is very similar, and each customer modifies the software to tailor fit their needs. I see pros and cons for implementing MDA for this company.
One reason not to implement is because Navitare already h
as the software developed, and only minor changes are made to the code. The development of the base software is complete. MDA tools wouldn’t be used much because they already have software that matches their model.

One reason to use MDA is in case of major changes to the software in the future. Maybe they would like to have a variety of base codes to give their customers more options. In this case, they would want to create more models. Another reason is if customers are asking for the software in various different platforms, because their internal IT group maintains the software. With MDA tools, the software output could be in any platform the customer wanted.