2.5 Admins 293: Reduced Flicker

In episode 293 of 2.5 Admins, Joe, Jim, and Alan dive into Microsoft's latest Windows Insider blog post promising a 'less shit' Windows 11 experience, focusing on reduced AI integration, fewer disruptive updates, and improved File Explorer performance. The hosts express skepticism, particularly about the vague promise of 'reduced flicker' in File Explorer, which they dismiss as unnecessary gaming-level optimization for a basic file browser. They contrast Windows' clunky update system—slow, unreliable, and often requiring full reboots—with the seamless, background-driven updates on Linux and macOS, highlighting how Linux systems can apply all updates in one efficient batch. The discussion then shifts to a real-world case of a cyberattack on Intoxalock, a company providing court-ordered ignition interlock devices, which left drivers stranded when servers went offline. The hosts critique the over-reliance on internet connectivity for essential functions, calling it a classic 'internet of shit' failure. They also address a listener's question about building a new home lab with TrueNAS and virtualization, advising against running TrueNAS or routers as VMs due to complexity, performance issues, and bootstrapping risks, instead recommending bare-metal ZFS setups for reliability and control. Key takeaways include: 1) Avoid nesting ZFS in VMs for performance and redundancy reasons; 2) Run TrueNAS and routers directly on bare metal for stability and simplicity; 3) Linux’s update model is superior due to background application and batch processing; 4) Critical systems should have offline fail-safes to avoid dependency on cloud services; 5) Microsoft’s promises of reduced AI and smoother updates are welcome but lack credibility without better QA; 6) Always consider boot order and network dependency when virtualizing infrastructure; 7) For learning and control, use general-purpose OS with ZFS over purpose-built distros; 8) The 'reduced flicker' claim in File Explorer is a red flag for misplaced priorities in software development.