Integration card at the end of user history (TDD)

Our team develops with TDD and when new features are introduced, sometimes at the end of the story, when all the cards turn green, an "integration map" appears, which means combining the implemented components so that they play with each other. I feel bad about this map because it means nobody tried the code in real life in tests only, and the integration map means try and make it work.

Is it a good agile practice to put an integration card at the end of each story that leads to a new feature? Or should it be part of every issue card when you can integrate it along with your existing code?

+3


source to share


1 answer


Here are just my 2 cents:

TDD itself has nothing to do with your way of working as described in your question. Agile / Scrum / Lean / Kanban in my opinion.

Writing test units before you actually write your code is a good practice you guys are doing if I understand you correctly.

I agree with you that it's a little weird to postpone the actual code integration until the end of the sprint. If hell breaks down, you won't get anything at the end of the sprint. In other words, your potential shippable product will not exist until the last risky card is done.



If you want to break through this way of working, I would consider googling for continuous integration . There you try to integrate the code as often as possible in order to find integration errors as soon as possible.

Shortly speaking:

Hopefully this helps you take the step towards a true agile / lean team! Trust your instincts, you will get it :).

+3


source







All Articles