Success or Quality? You cannot have both of them, they say…
13 03 2008The project managers of the projects I work for can be classified into two groups: either they say quality comes later, first we have to finish the project, i.e. to implement all the functionality. Or the say, we want to make it right from the beginning, implement in a more quality conscious way, and hope that we will finish the project on time.
I tend to belong to the second group, except in projects where it’s really incredibly important to be as quick as possible and where nobody has the nerves to think about making it right, just working;-)
It is my deep belief that it is one of the most important things to write test cases and to test right from the beginning of the project. And that’s not only about unit tests with JUnit or whatever, it is also e.g. about testing the written use cases or the UML diagrams. Actually that’s something which is overseen very often. But only because you cannot test those artifacts automatically (as far as I know), it doesn’t mean that it shouldn’t be done.
Explain the use cases or diagrams to some other people who have some knowledge about the project or the business and ask them to find out what’s wrong. They surely will find a lot…
I think that the time one spends in testing from the beginning will pay later in the time of bug fixing. In many projects developers spend an incredibly amount of time for bug fixing, and this mainly because it was never a goal to develop test-driven oriented.
So what can you do in such projects? Not much, first the managment has to be convinced to change their way - and that’s probably very difficult. If tests are an important value for the management, then the rest will follow.






Recent Comments