The other day, I mentioned that I was preparing for a presentation.
The aim was to – as a minimum – get people to follow the existing process. The ultimate aim was to try and get buy-in for following a full out Test Driven Development methodology (TDD).
My previous post about this generated some awesome advice – some mentioned that I should pin point a case study of where a process wasn’t followed. A former manager suggested I avoid death by power-point, and instead cover a live example of how TDD can speed up the development process. Others suggested that I follow a four point rule.
It’s just sods law that I woke up in the morning with a cracking headache and feeling sick. Awesome. I dragged myself into work, regardless – I’d put far too much thought and effort into this to call in sick on the day.
I think I managed to incorporate elements of all the advice I received. I opened up with a brief section on ‘when things go very, very wrong’, to establish that I’d already had a trial by fire for this kind of thing. I then demonstrated how time can be saved and pit falls avoided using TDD. I closed off with my suggestions that we adopt a TDD-like methodology, or at the very least follow the process we already have across the board, with no excuses. All issues were condensed down to a maximum of 4 points each, to avoid information overload. I then opened up the room for discussion.
Unfortunately, the management were unable to attend the meeting, which meant that I was really working from the ground up. It turns out that the most powerful tool in my arsenal was the fact that the product recall notice (the price of not doing this right) is still on the internet. The twitter posts are still there as well. It’s hard to argue against that.
Overall – I don’t have full buy-in of TDD yet. However, there is an agreement that the process as it currently exists will be followed more stringently. The value of unit testing was accepted as common sense, but the same old arguments reared their heads – resources and time. It was agreed that we would need more drive from a management level to get resources dedicated to setting up a proper nightly build system, and that is still the main issue with what I’m trying to get in place.
After the presentation, my manager asked me to send him and the other managers a summary of the results of the presentation, which has given me the opportunity to not only give feedback, but to also push for giving the presentation to the management.
So, the battle isn’t over yet. I haven’t been shouted down, meaning I have plenty of room to keep the argument for this going. I feel I have managed to move things forward a small step, which as the new girl in the office, I don’t think is bad going. There is still a long way to go, and I’ll no doubt be blogging about it as I continue pushing for it.
I’d like to once more give a huge thanks to all the people who gave up their time to give me advice on how to proceed with this presentation – it’s very much appreciated, and I don’t think I would have gotten a team to agree to follow the process more closely without following your advice! You’re all awesome!