Just Do It

I had given a talk @ Toronto a few weeks back about Feature Toggles and Mainline Development. Few people came and told that the talk was great and a few others came up to me for further clarifications.

One of the questions that came up was, this [Mainline development] is a great idea. The team always wanted to do this, but not sure when and how to get started. We discussed a few options about how to get started and what might work.

One person who had the question, was still concerned because he wanted to be safe. He asked me about how I got started and what my experience was. I told him we “just did it.” He seemed to have convinced and told he will try the same approach, i.e., “just do it.” I hope he does and achieves excellent results.

A few years back, I was reading the Continuous Delivery book about Feature Toggles and Mainline Development and realized how much waste is created by keeping multiple branches. When I shared my learnings from the book, the team also felt that we are making the delivery more complicated than required. And we just switched on the spot. It worked smoothly.

The best option is “just do it” for bringing in change than analysis-paralysis for a good time. We did the same for pair programming, TDD, etc., and most of the time it worked.

Small Steps

Do in small steps to mitigate the risk in case the changes require a lot of steps. The advantage of small steps is that they can be reversed, in case it's necessary. So always try to work in small batches.

Someone told me recently about merge conflicts while checking into source control:

I never fix merge conflicts. I always overwrite with their changes and then add my code and commit. I can code it again really fast because every commit will be just a few lines. It is much faster than merging conflicts.

Wow, what a great perspective. I never thought so. If the commit is small, I can always add my code, because it’s fresh in my mind. Why should I spend time analyzing the conflicts and fixing the same?

Misconception about Risk

We want to be safer and try to avoid taking risks. As Mark Zuckerberg mentioned, the biggest risk is not taking any risk. Another way to look at it is as Martin Fowler says, if it hurts, do it more often.

Changes are inevitable, so bring in change. Instead of bringing in a big-bang change, bring it in small steps. Don’t overthink it, most of the time “Just do it” is the right option.

Maybe “Just do it” works better because as per The As-If principle, changing our behavior changes our thinking.