2014: Kubernetes

2015: TensorFlow

2017: Flutter, Transformer

Those are some big gaps. It looks like Google is in the position Microsoft was in for so long right down to being dependent on a cash cow (their ad network) and lock-in (Workspace). They've rested on those laurels for so long they forgot how to innovate. The difference is Microsoft realized they were on the path to irrelevance and started working on it years ago. Google just realized there's a threat to their dominance, and that threat is moving faster than anything that challenged Microsoft. LLMs are making leaps measured in weeks and months.

Google Tables (2020) could be amazing, but who's going to risk it with their reputation for shuttering new things once the people responsible have been promoted? Think about your list: it's all open source. Nobody's trying new Google stuff unless they know it can be forked.