October 2021 Summaries

In the engineering improvement runbook, Don Brown discusses testing and code coverage. He introduces the concept of a testing pyramid with smoke tests at the top, end-to-end tests in the middle, integration tests below that, and unit tests forming the base. The number of tests decreases as you move up the pyramid. Brown emphasizes that setting code coverage targets depends on the complexity of the code being tested. For instance, a parser or protocol implementation should be massively tested, while getters and setters may not need unit testing at all. Unit tests can also serve as documentation for developers, reducing risks associated with changes in production. In conclusion, while 90-95% code coverage might be necessary in some cases, it's not always required. The key is to understand the complexity of the code and adjust testing efforts accordingly.

The text provides a step-by-step guide on setting up Sleuth, an AI tool that automatically tracks DevOps Research and Assessment (DORA) metrics in minutes. It involves visiting the Sleuth website, signing up with a preferred provider like Google or GitHub, naming your project, adding a code repository, and specifying how to determine service health using supported services such as Sentry. The guide also recommends configuring a webhook for more accurate metrics tracking.

The text discusses the use of DORA metrics for tracking software development performance. It introduces Sleuth as a tool that accurately tracks these metrics with minimal effort from users. Sleuth tracks change lead time, change frequency, change failure rate, and mean time to recovery by analyzing code deployment, service health, and other relevant factors. By pointing Sleuth at the code repository and sources of service health, it continuously generates accurate DORA metrics without requiring maintenance or manual intervention.

The text discusses using DORA metrics for improving team performance, specifically focusing on reducing change lead time. It introduces Sleuth, a tool that tracks these metrics, and demonstrates its usage in identifying problem areas. The author explains how to interpret the change lead time graph, zooming into specific periods and analyzing contributing factors such as review time. They also highlight Sleuth's ability to provide insights and automate notifications for keeping track of changes over time.