Jira is Turing-complete

Jira is Turing-complete
3 minutes by Nicolas Seriot

Jira's automation system can simulate a Minsky register machine, a model proven equivalent to a Turing machine. Registers map to linked issue counts, the program counter maps to an Epic's status, and instructions map to automation rules. A working addition machine and a Fibonacci generator both run inside standard Jira Cloud. This confirms that Jira automation is Turing complete.

Ship it right, not twice: How to reduce accessibility rework
sponsored by Level Access

Ship. Fix. Repeat. If addressing accessibility issues feels like a never-ending cycle, the
problem isn't effort—it's process. Learn how to stop issues at the source in the
Accessibility Resolution Playbook. You'll get a practical roadmap to shifting from
remediation to prevention, so you can stay focused on building new features, not fixing
what's broken.

Should you normalize RGB values by 255 or 256?
8 minutes by Pekka Väänänen

Dividing image pixel values by 255 maps black to 0.0 and white to 1.0, which is clean and matches how GPUs work. Dividing by 256 with a 0.5 offset places each value at the midpoint between two integers, offering slightly better theoretical precision. In practice the difference is tiny, and for images made by others, dividing by 255 is the right choice. Division by 256 only makes sense when you control both saving and loading.

Every byte matters
5 minutes by Farid Zakaria

How you organize data in memory can matter as much as your algorithm choice. Using cache sizes and access patterns, Farid compares Array of Structs and Struct of Arrays. He shows that organizing data to match CPU cache behavior can significantly improve speed, especially for large datasets and random memory access. The key message is that reducing data size and improving cache efficiency can have a major impact on performance.

The history of pets vs cattle and how to use the analogy properly
7 minutes by Randy Bias

Randy explains the origin and proper use of the “Pets vs. Cattle” analogy in cloud computing. He describes how the analogy helps distinguish traditional, unique servers, “pets”, from modern, disposable, and automated servers, “cattle”. He argues that the key idea is server replaceability and fault tolerance, not whether applications are stateful. Randy emphasizes preserving the analogy’s original meaning to avoid confusion.

Fast is better than slow
5 minutes by Patrick Dubroy

Speed makes great programmers, not the other way around. Moving fast means quicker feedback, better decisions, and more learning over time. Simple habits help: start tasks right away, use small pockets of time, share work early even if imperfect, avoid pointless debates, and do only what is asked. These small changes add up to major gains without working longer hours.

And the most popular article from the last issue was:

• AI engineering for developers