November 2019 Summaries

LaunchDarkly's rigorous testing strategy for their SDKs involves continuous integration workflows with unit tests and integration tests across various platforms. They utilize "restwrappers" to expose core SDK functionality through a consistent REST API, allowing for consistent testing across multiple languages and platforms. The Roku SDK required more effort due to the lack of existing virtualized environments and HTTP server support on the platform, necessitating an in-house solution. Despite these challenges, LaunchDarkly's commitment to thorough testing ensures reliable product delivery and continuous improvement of their SDKs and infrastructure.

LaunchDarkly introduces a new integration for mParticle in its Data Export feature. This allows teams to analyze user cohorts and combine data sets by piping real-time raw event analytics directly into mParticle. Once in mParticle, the data is unified into single customer profiles and used to create enriched customer audiences that can be activated across the tech stack in real-time. The integration helps teams make smarter decisions with their customer data by monitoring rollout success and debugging problems without worrying about duplicate customer profiles or data silos. To get started, contact [email protected].

LaunchDarkly introduces a new Segment integration for its Data Export feature, allowing teams to analyze user cohorts by combining real-time feature flag evaluations with existing data sets. This integration enables teams to monitor rollout success, construct enriched customer cohorts, and debug issues within their preferred data analysis tools. The Data Export add-on is available for LaunchDarkly Enterprise plan customers, who can access it by contacting [email protected].

In this talk, Katherine Gallagher discusses the importance of onboarding new hires effectively to improve retention rates and productivity. She introduces the concept of Retention-Driven Development, which focuses on empowering new hires to onboard themselves by leveraging their own skills and knowledge. Gallagher emphasizes that effective onboarding should be employee-driven, with new hires taking ownership of their learning process. This approach helps combat issues such as disenfranchisement, overloaded managers, and stale documentation. To implement Retention-Driven Development, Gallagher suggests creating a touchpoint matrix that outlines the key topics and resources for onboarding. New hires can then be tasked with identifying gaps in documentation or processes and working with subject matter experts to address these issues. By empowering new hires to contribute actively to their own onboarding, companies can foster a more collaborative and inclusive work environment, ultimately leading to higher retention rates and improved productivity.

In this talk, Rick Zucker, an experienced Site Reliability Engineer (SRE) Manager, shares spooky scenarios in debugging and lessons learned from these experiences. He discusses two debugging scenarios he personally dealt with - one involving a compiler bug and another related to hardware verification. The key takeaways from his talk are: 1. Be careful about what you take for granted. 2. Consider corner cases that aren't recommended but are permitted. 3. Always be observant about everything. 4. Sometimes, you have to be paranoid and skeptical about things. 5. Trust in your supply chain is crucial, and it's essential to think about where bad actors might take over.

Experimentation is crucial in validating new ideas, methods or activities for an application, feature, or infrastructure. It differs from testing as experiments lead to new knowledge while testing only verifies behavior. Common types of experimentation include A/B testing, multi-armed bandits, and chaos engineering. Experiments help validate changes proposed to a product, learn which features impact users the most, iterate and improve on features, and discover what doesn't work. The frequency of experiments depends on the company, but they should be run for major features, site redesigns, significant functionality changes, or performance tweaks. Involvement in experiments varies based on their type, and it is recommended to run them in production with appropriate controls and kill-switches.

LaunchDarkly has released its SDK for Roku (BrightScript), making it available for use in digital streaming apps. This addition allows developers to manage features, conduct progressive rollouts, and perform experiments across multiple platforms consistently. The Roku SDK reference guide provides information on how to embed the SDK in an app, view open-source code, and run a sample application. Support is also provided via email at [email protected].

The text discusses the benefits and use cases of Terraform, an open-source infrastructure as code software tool created by HashiCorp. It explains how Terraform simplifies managing complex infrastructures across multiple cloud services, reducing manual work and enabling consistent processes for creating and changing configurations. The author also introduces a new LaunchDarkly Terraform provider that allows users to manage feature flags using Terraform. Two examples are provided: one demonstrating how to create environments without storing individual flag details in Terraform, and another showing how to manage and protect flags without changing their state. The text concludes by discussing the potential use of feature flags in HCL files and encouraging feedback on this topic.