Frankie, Keiran, TJ and I had a great time at YOW 2018. Practical and buzz free Judging from the agenda, the new hot things are (unsurprisingly) serverless and Kubernetes. Agile, microservices and AI continue to be popular.
Kent Beck had a great talk on his theory of the 3 stages that a company (or part of a company) can be in: explore, expand and extract. You see what he did there with the e's. Exploring is about experimenting feverishly to find a market. You've got nothing to lose and lots to gain, so the key is lots of low-cost experiments that have a small chance of a very big payoff. Expanding is when you find that 1 experiment that finds traction and you're growing into your newfound market. Here, the main focus has to be systematically predicting and removing barriers to growth. You have a high chance of a big payoff, so it is critical that everyone is focussed on making this happen. Extract is that plateau where things are stable. At this time, most changes are incremental. The goal is to nudge your market share a fraction higher, and your risk a fraction lower. Different teams can be in different stages.
Kent made the point that recognising where you are on the curve is critical. You'll know when you make the shift from explore to expand. But it can be harder to detect the transition into extract. This is important, because very different engineering and business practices are useful at each stage. When you're exploring, the most important thing is learning as quickly as possible. You need to be running lots and lots of low-risk experiments. When you're exploring, tests only make sense if they let you experiment faster. Impact metrics can be a great way to make sure you're being rigourous in your experiments.