Bitrise

Why the Author Still Writes How-To Posts

The author reflects on why they continue to write how-to articles despite changes in search behavior driven by AI. Citing research and industry reports, the post notes that in 2026 less than a third of Google searches end with a click and that AI Overviews on search results can reduce click-through rates on those queries to as low as 17–20%. The author argues generic, reference-style how-tos are increasingly captured and summarized by AI, while posts that describe real production experience, mistakes, and original insights retain value and higher-quality traffic. The author plans to write fewer generic tutorials and more firsthand, experience-driven guides.

AI Accelerating Shift from Open Source to Paid Products

The author observes a growing trend of popular open-source projects moving to commercial or dual-licensing models, citing examples from the .NET ecosystem (AutoMapper, MediatR, Fluent Assertions, MassTransit) and frontend libraries after PrimeTek's announcement that future major versions of PrimeNG, PrimeReact and PrimeVue will not be released as open source. The post argues that AI may be accelerating this shift: AI enables rapid code generation but also produces large volumes of issues, pull requests and feature requests that maintainers cannot easily review. Concerns highlighted include maintainers choosing to keep code private to avoid unconsented model training, difficulty proving GPL influence on AI-generated competing implementations, and faster discovery/exploitation of vulnerabilities by attackers using AI. The article is framed as an analysis and asks whether these dynamics will change developers' willingness to publish open-source projects.

Auditable Regulatory Reference Data via Versioned CSV

A July 28, 2026 DEV Community article by Yann_ describes a practical workflow for converting legal text into small, versioned, auditable reference datasets (CSV). The author argues for designing datasets around decisions (who must act, when, and how), preserving precise primary-source metadata (source_url, version, last_verified), modeling timing explicitly across organisation categories, and defining maintenance triggers and versioning practices. The goal is to make regulatory claims inspectable and machine-reusable without attempting to automate legal judgment.