The garbage collection handbook

The garbage collection handbook
a few hours by Richard Jones et al.

This handbook provides a comprehensive overview of garbage collection, a key technology for automatic memory management in programming. It covers both traditional and cutting-edge techniques, including parallel and real-time collection methods.

Hands-on AWS Skill Paths built for developers
sponsored by Jetbrains

JetBrains and AWS created hands-on Skill Paths built around real developer workflows. Code in your IDE, practice in AWS sandboxes with no surprise costs, and get a certificate at the end. The DevOps Skill Path is free to complete on the AWS Free Tier. AI and LLM Skill Paths include full AWS lab access for training models and deploying GenAI apps. Start with DevOps and join the AI and LLM waitlist.

How to make logging better
15 minutes by Boris Tane

Boris explains why traditional logging fails in modern, distributed systems. Logs lack context, are hard to search, and don’t reflect how requests actually flow. He introduces “wide events”: one rich, structured event per request with full context. Combined with smart sampling, this approach makes debugging faster, clearer, and more reliable.

Software engineers should be a little bit cynical
9 minutes by Sean Goedecke

Sean defends being called cynical for advice like "do what makes your manager happy." He argues that idealistic views of software engineering are actually more cynical because they assume companies are fundamentally corrupt. In contrast, engaging with company politics to ship meaningful changes to millions of users is genuinely idealistic. A small dose of cynicism helps engineers understand how large tech companies really work.

When compilers surprise you
3 minutes by Matt Godbolt

Matt explores how modern compilers optimize simple code in surprising ways. Using a loop that sums numbers, GCC cleverly processes two numbers at once, while Clang eliminates the loop entirely with a closed-form formula. These optimizations turn an O(n) operation into O(1), showing the power and sophistication of compiler design, and highlighting how even experienced programmers can be amazed by their clever transformations.

Economics of orbital vs terrestrial data centers
11 minutes by Andrew McCalip

The article analyzes whether space-based computing can compete economically with ground datacenters. Andrew built a cost model comparing orbital compute to terrestrial alternatives. The physics don't immediately rule it out, but economics are challenging with costs roughly 400% higher than ground-based systems.

And the most popular article from the last issue was:

• How to review code