Shadowfetch

Journal beat

Linux and open source

This beat runs from the first screen of an installer to the machines in the back of a building. There is an essay about what a system asks you before it will do anything, a look at open-source AI tooling reaching the point where a small team can actually run it, a Home Assistant release that makes a smart home less brittle, a project that turns kernel security lag into something an operator can see rather than assume, and one company walking away from VMware and finding out what that costs at the edge. Shadowfetch Linux itself is documented under /linux; this beat is the surrounding ecosystem, not the product.

These are existing pieces from the Shadowfetch archive, still hosted on the previous site — every link below opens at its original /blog address. Nothing on this page is new writing.

Essays

First-person writing from the studio. The oldest material in the archive, and the reason this grouping exists — newest-first would put all of it below months of news.

  1. The First Question the Machine Asks

    The installer stopped on the keyboard screen again. Not the disk screen. Not the network screen. The keyboard. A plain list, a highlighted default, one…

Reporting

Reported pieces on the tools, defaults and decisions around this beat.

  1. Patchless turns Linux kernel security lag into an operator-visible problem

    A new public monitor highlights the gap between upstream Linux kernel security fixes and the kernels actually running in fleets.

  2. Sheetz’s VMware exit turns a licensing gripe into an edge-computing warning

    Sheetz’s migration of more than 830 stores off VMware shows how Broadcom’s licensing changes are becoming a real edge-infrastructure decision for distributed enterprises.

  3. Home Assistant’s 2026.7 update makes the smart home less brittle

    Home Assistant’s July automation overhaul makes connected homes easier to run, but the real win depends on honest setup, privacy, and lock-in tradeoffs.

  4. Open Source AI Tooling Keeps Maturing for Small Teams

    Open source continues to lower the barrier for small teams shipping AI features. Local model runners, lightweight vector stores, and permissivelicense…

Beat assignments on this page are an editorial judgement made by Shadowfetch; they are not part of the original article metadata.