A Java Geek weekly 125

I was curious to compare what I learned by doing to the structured approach of a book. I ordered the book last year when it was still being written. It was published only early this year, and I was already reading (and reviewing!) DuckDB in action. I put it on the top of my reading pile list; I finally finished it: here’s my review.

What is a problem on Twitter et al will become the same mess on GitHub od Gitlab - just a matter of time.

For this reason Forgejo is working on federation for git platforms. Some months ago we released the first user visible feature the federated star and we will continue our path.

We will talk about our motivation, point out the general roadmap and show implementation details.

Across multiple coding agents and LLMs, we find that context files tend to reduce task success rates compared to providing no repository context, while also increasing inference cost by over 20%. Behaviorally, both LLM-generated and developer-provided context files encourage broader exploration (e.g., more thorough testing and file traversal), and coding agents tend to respect their instructions. Ultimately, we conclude that unnecessary requirements from context files make tasks harder, and human-written context files should describe only minimal requirements.

A central misconception the OpenTelemetry team addresses is the belief that OpenTelemetry is itself an observability platform or monitoring product. In reality, OpenTelemetry is a vendor-neutral instrumentation and data collection standard, not a backend for storing, visualizing, or alerting on telemetry. It provides the APIs, SDKs, data models, and collectors needed to generate and export telemetry, but organizations must still choose a backend, open source or commercial, to make that data usable.