views:

101

answers:

3

I am trying to become more familiar with test driven approaches. But I am embarrassed because I have to report the effect of TDD. I should report on the UT result to the superior. (I reported value of bugs, when I enforced UT(unit test).)

But I can not obtain the UT(unit test) result if I try TDD. Please let me know how to report the effect of TDD.

+1  A: 

If you are using TDD, then the outcome of unit testing is not successful unit tests - it's the code that you wrote in order to make those unit tests succeessful. That's how to measure - how much code has been written.

John Saunders
A: 

Maybe you could report on the value of bugs already "squashed" before they become problems.

More seriously, the tests in TDD represent the features implemented. So effort should approximately equal to the feature estimates.

If the real question is how much effort was spent on TDD, as if it's extra time, then I don't know. I haven't practiced TDD religiously on the job. The real metric might be your own comparison of similar scale work done with TDD and without TDD.

marklai
Sorry!!!!!!!!, I miss take, effort -> effect or result.
kumagon
+3  A: 

You can't easily measure the quality or value of code. In fact, code is a liability and only the resulting software is an asset.

In any case, TDD has very little to do with bug fixing or Quality Assurance. It is a software development discipline.

Although I know this may be difficult, you need to re-educate your manager.

  • TDD is a development methodology, not a testing practice.
  • Experience shows that TDD is superior in the long run.
  • Expect development time to be longer, but 'testing' periods to be shorter.
  • There will be fewer bugs, but there will still be bugs.
  • As you gain experience with TDD, the first project iteration or two may simply take longer overall, as with any learning experience.
Mark Seemann