January 2023 Summaries

AWS CloudTrail and LaunchDarkly have integrated their services, allowing customers to ingest LaunchDarkly activity data into CloudTrail Lake. This integration enables organizations to control, store, and query events recorded by both AWS CloudTrail and LaunchDarkly in a single location. The consolidated audit data provides a comprehensive view of changes across the entire infrastructure, making it easier for development teams to stay compliant with fewer tools and overhead. The integration is easy to configure and offers deep visibility into LaunchDarkly accounts.

A canary deployment is a controlled rollout strategy that involves pushing out changes to a small subset of users first before releasing to the rest of the user base. This pattern allows organizations to limit or avoid downtime, gather feedback on new code behavior in production, and easily roll back in case of bugs or performance degradation. Canary deployments are inspired by how canaries used to serve as an early warning for toxic gases down coal mines. They help test the new feature in production while minimizing impact to users. Some benefits include getting early feedback from real users on performance, bugs, and user experience before exposing all users to the new version, quicker rollback in case of significant errors, and reduced deployment stress due to lower stakes. However, there are also drawbacks such as two environments to maintain, technical challenges, support for multiple versions, and some errors not making themselves known on a user's first exposure to the new version.

Datadog is a cloud observability and security platform that provides visibility into metrics, traces, and logs across your tech stack. It includes Real User Monitoring (RUM), which allows you to visualize and analyze the real-time performance and user journeys of individual application users. To collect events, Datadog provides an easy-to-configure RUM Browser SDK. LaunchDarkly uses Datadog RUM software to monitor its applications as customers interact with them. By integrating LaunchDarkly's feature flags with Datadog RUM, you can supercharge your configuration and control various aspects of data collection for different users or environments. This integration allows for more granular control over the data collected without requiring extensive custom code changes.

LaunchDarkly has introduced the first feature management solution to be authorized by the Federal Risk and Authorization Management Program (FedRAMP): LaunchDarkly Federal. This authorization demonstrates LaunchDarkly's commitment to supporting U.S. government agencies, system integrators, and commercial partners of the U.S. government. The company has been delivering enterprise-class feature management solutions to public sector communities for years, enabling organizations to accelerate digital transformation and mitigate risks in software delivery. With LaunchDarkly Federal's FedRAMP Moderate authorization, the platform can better support the missions of government agencies with its scale, performance, and ease of use.

Cody De Arkland shares his enthusiasm for OpenAI's ChatGPT and its potential to demystify complex coding concepts, drawing from his personal journey from automating tasks with bash scripts to developing applications with languages like Python, Flask, and React. He highlights how ChatGPT provides users with practical coding examples and explanations, such as implementing LaunchDarkly feature flags in React or Python, while also offering guidance on rolling out new features or migrating databases. Despite some limitations, like incomplete code snippets and outdated data, ChatGPT serves as a valuable educational tool by lowering entry barriers and encouraging users to ask pertinent questions to refine their coding skills. De Arkland concludes that while ChatGPT is not set to replace developers, it effectively showcases possibilities and serves as a starting point for further exploration in coding.

LaunchDarkly has introduced a new approval requests dashboard to enhance tracking and managing approvals for its customers. This feature aims to improve the release process by providing a singular place to access all flag approvals, reducing the need for navigating through individual flags. The dashboard allows users to visualize, filter, and sort approval requests easily. With improved Slack integration, users can stay updated on new requests and boost collaboration between teams using LaunchDarkly. This feature is now available in Professional and Enterprise plans.

LaunchDarkly introduces its new Professional Services team aimed at assisting customers in their feature flag adoption journey and accelerating software delivery solutions. The team offers four paid engagements, including Platform Readiness, Application Readiness, North Star, and LaunchDarkly for Developers. These services focus on different aspects of the LaunchDarkly implementation process and provide best practices to help customers achieve key milestones during their journey.

A production environment is where the latest versions of software, products, or updates are pushed into live operation for end users. It differs from test and staging environments in that it hosts new applications for real-time use by users. While testing was once mainly performed in staging environments, feature flags now enable developers to conduct various tests and checks in production settings. This allows for better releases as user data and feedback can be gathered, and bugs not apparent in artificial staging environments can be uncovered. Some examples of what can be done in a production environment include A/B testing, canary testing, application monitoring and performance testing, and User Acceptance Testing (UAT). Feature flags make production testing safer by allowing developers to enable or disable specific functionalities without the need for code deployment. LaunchDarkly's feature management platform helps teams conduct safe tests in production environments at a high frequency and large scale.

The AWS re:Invent 2022 conference featured numerous world-class sessions and impressive announcements. Dr. Werner Vogels, Amazon's CTO, delivered a notable keynote discussing asynchrony in nature and loosely coupled systems. Other highlights included insights into the history of AWS architecture and real-world examples from companies like Trustpilot. The Amazon Builders' Library was also mentioned for its valuable resources on building and operating software. Various sessions were conducted by experts, covering topics such as feature management, observability, security, and more. Notable announcements included the Step Functions Distributed Map, AWS Application Composer, EventBridge Pipes, and Amazon CodeCatalyst. The event also featured fun activities like serverlesspresso and a light speed tunnel at LaunchDarkly's booth.

Platform engineering is an emerging field focusing on integrating tools and designing workflows to enable self-service capabilities for software developers, particularly in cloud computing services and Infrastructure as Code tools. It differs from DevOps by reducing the need for coordination between development and operations teams, and from site reliability engineering (SRE) by serving developers rather than operations teams. The benefits of platform engineering include faster, more frequent releases, standardized workflows, and a better developer experience. However, it may not be necessary for early-stage startups or small companies. To build a successful platform engineering team, one must treat the platform as a product, connect it to business goals, manage expectations, effectively communicate with internal customers, and possess empathy, strong communication skills, willingness to teach, and experience in relevant technologies.

Feature flags are a software development process that uses switches or control features to toggle functions without needing code deployment. They enable testing and isolation of features before making them publicly available, allowing greater control over the development cycle. LaunchDarkly offers feature management tools for automation, analytics, and coordination of feature flags across diverse architectures. Proper feature management is crucial in microservice-based development to handle complex interconnected codes and varying code infrastructures. LaunchDarkly provides an SDK suite supporting major languages and frameworks, comprehensive feature management capabilities, and easy-to-use feature flags. Building a feature management solution from scratch can be resource-intensive and time-consuming compared to using an enterprise-proven system like LaunchDarkly.