I am a bit of a vocal proponent of the BDD methodology. I've been applying BDD for a couple of years now, and have adopted StoryQ as my framework of choice when developing DotNet applications. Even though I have been unit testing for many years, and had previously shifted to a test-first approach, I've found that I get much more value out of using a BDD framework, because my tests capture the intent of the requirements in relatively clear English within my code, and because my tests can execute multiple assertions without ending the test halfway through - meaning I can see which specific assertions pass/fail at a glance without debugging to prove it.
This has really been the tip of the iceberg for me, as I've also noticed that I am able to debug both test and implementation code in a more targeted manner, with the result that my productivity has grown significantly, and that I can more easily determine where a failure occurs if a problem happens to make it all the way to the integration build due to the output that makes its way into the build logs. Further, the StoryQ api has a lovely fluent syntax that is easy to learn and which can be applied in an extraordinary number of ways, requiring no external dependencies in order to use it.
So with all of these benefits, you would think it an easy to introduce the concept to the rest of the team. Unfortunately, the other team members are reluctant to even look at StoryQ to evaluate it properly (let alone entertain the idea of applying BDD), and have convinced each other to try and remove a number of StoryQ elements from our own core testing framework, even though they originally supported the use of StoryQ, and that it doesn't impact on any other part of our testing system. Doing so would end up increasing my workload significantly overall and really goes against the grain, as I am convinced through practical experience that it is a better way to work in a test-first manner in our particular working environment, and can only lead to greater improvements in the quality of our software, given I've found it easier to stick with test first using BDD.
So the question really comes down to the following:
What arguments can I use to really drive the point home that it would be better to use StoryQ, or at the very least apply the BDD methodology?
Can you point me to any anecdotal evidence that I can use to support my argument to adopt BDD as our standard method of choice?
What counter arguments can you think of that could suggest that my wish to convert the team efforts to BDD might be in error? Yes, I'm happy to be proven wrong provided the argument is a sound one.
NOTE: I am not advocating that we rewrite our tests in their entirety, but rather to simply start working in a different manner for all future testing work.
