Observatory Weekly: Software Signal Engineering, use AI with your brain, and

In The Observatory, the MoTaverse community shows up to share links they are reading all the time.

The value is not only in the links shared in the moment, but also in the accumulation of links that become searchable knowledge over time.

There are currently 2890 searchable links.

Your Testing Tools Leave Them NO CHOICE

ACCELQ builds tests in plain English, no code, cuts maintenance 70%. Book a demo to see it across your entire stack!

🎉 Featured link of the week:

Software Signal Engineering — A discipline for testing and reliability in complex dynamic systems.

Interesting idea of "signal engineering" as a non-deterministic approach to testing — Lisa Crispin

10 links from the Observatory this week:

The Archaeologist and the Witness — Simon Stewart

A long time ago, I used to say that there were two missing roles on any software project: the Software Archaeologist and the Project Witness.

The Archaeologist would be able to read commit logs, meeting notes, story cards, or any other form of written context to try and piece together what had happened and why. It’d be possible to do this at any stage of a project, even long after it had finished.

The Witness, however, would be the holder of all the oral history of a project, the teller of tales, and the person who could describe how decisions were made, and why they were made that way.

Are you using AI? Do not forget your whip — Skye Cleary

On creating distance from AI, with some Nietzschean advice

Accessibility Game Theory — Sheri Byrne-Habler

Why do capable organizations keep buying and building inaccessible products, and how do you change the payoffs?

Game theory studies situations where the best move for one party depends on what the other parties do. This includes competitive procurement, contract negotiation, and litigation. A substantial amount of accessibility strategy can be linked to game theory.

Testing authentication "unhappy paths" using a mock — Wonderproxy

Handling user authentication and making sure the many different scenarios work as planned can be a significant challenge for many development teams. Especially considering that the authentication layer is commonly handled through a separate service or even a third-party provider.

Sainsbury's branch halts AI use as shopper ejected — BBC

Sainsbury's has suspended its use of live facial recognition technology at one of its London branches after a customer was wrongly challenged as a shoplifter and asked to leave.

The Bechdel Test for Building a Real Developer Community — Leon Adato

For those unfamiliar with it, the “Bechdel test” is named for the American cartoonist Alison Bechdel. In one strip from her series “Dykes to Watch Out For“, a character described their 3 basic requirements to go see a movie. The

How to Use AI With Your Brain — Kristin Jackvony

In spite of its name, artificial intelligence is not actually intelligent. The AI agents we use are based on a Large Language Model (LLM), which is a database of trillions of words. When we ask an AI agent a question, it is answering based on the information in that LLM. It makes predictions based on training and previous interactions about what you are asking and what you might want next. Do not make the mistake of believing that the AI knows how to “think”. It knows how to fetch data, organize information, and make predictions.

Beads: A Git-Friendly Issue Tracker for AI Coding Agents

AI coding agents have transformed software development workflows, offering the ability to write code, fix bugs, and accelerate project timelines. However, these agents face a fundamental constraint: finite memory, commonly known as the context window. As conversations grow and files accumulate, the context window fills up. When you start a new session, the agent loses all previous understanding of the project's goals, progress, and architecture. This creates a persistent challenge in maintaining continuity across development sessions.

How we tracked down a 16-year-old SQLite bug — Tailscale

At the end of last year, our uptime was pretty shaky. You can see this trend on our status page, and that instability continued into the new year. Many of these outages were caused by a single bug, deep in SQLite. It took months of intense forensics to track it down.

Cyberattacks expose U.S. water vulnerabilities — Axios

Suspected Iran-linked cyberattacks are exposing long-standing vulnerabilities in fragmented U.S. drinking water systems facing aging infrastructure and intensifying climate threats.

Explore all Observatory links…