<h2>Must know</h2>
<ul>
<li><strong>Microsoft deprecates SCOM packs, pushes Azure monitoring</strong>: Microsoft is deprecating System Center Operations Manager (SCOM) Management Packs for SQL Server Reporting Services, Power BI Report Server, and SQL Server Analysis Services by January 2027. This move forces organizations to migrate to Azure-based monitoring solutions, requiring on-premises servers to register with Microsoft’s cloud. Analysts, as reported by various sources, warn this increases exposure to Microsoft’s license compliance audits, raising concerns about data privacy and compliance risks for complex or legacy infrastructures. <a href="https://app.daily.dev/posts/Lyoqg9qSt" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Intel faces server CPU supply constraints amid AI demand</strong>: Intel has warned of its inability to meet the surging demand for server CPUs, particularly those used in AI data centers, leading to a 13% drop in its share price. Despite manufacturing facilities operating at full capacity, the company faces supply constraints and production lag times. Intel’s CEO, Lip-Bu Tan, expressed disappointment, highlighting that demand for server CPUs alongside AI chips is the unexpected spike, further complicated by a global memory chip shortage, according to Intel executives. <a href="https://app.daily.dev/posts/4urCLb8ya" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>MySQL community debates future amid Oracle governance concerns</strong>: MySQL community developers are addressing governance concerns over Oracle’s stewardship, fueled by recent layoffs affecting its core team and a perceived decline in code commits. Discussions include pressuring Oracle for transparency or forking the project, with options like a hard fork (MariaDB style) or a tracking fork (Percona Server style) on the table. Engineers from Percona and PlanetScale are involved, consulting with major cloud vendors to address issues like the absence of vector search functionality, as reported in community forums. <a href="https://app.daily.dev/posts/l2bePOl0A" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Rust 1.93 updates musl library and improves networking</strong>: Rust 1.93 upgrades the bundled musl library to version 1.2.5, bringing significant DNS resolver improvements for static Linux binaries, especially for large DNS records and recursive name servers. <a href="https://app.daily.dev/posts/wF6DW7Hp4" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Linux 6.19 NEXT_BUDDY scheduler disabled due to regressions</strong>: The Linux 6.19 kernel has disabled the re-implemented NEXT_BUDDY scheduler by default after performance regressions were observed in workloads like MySQL and Nginx, with a patch queued to mitigate these issues. <a href="https://app.daily.dev/posts/iDlOb4czh" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Vulkan roadmap 2026 released with descriptor heap overhaul</strong>: The Khronos Group published the Vulkan Roadmap 2026, introducing Vulkan 1.4.340 with the VK_EXT_descriptor_heap extension for direct memory access, aiming for performance and scalability improvements, with SDK support expected in Q1 2026. <a href="https://app.daily.dev/posts/Ow0tbWANK" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AWS launches EC2 M4 Max Mac instances</strong>: AWS has released M4 Max Mac instances for EC2, utilizing Apple M4 Max chips to offer up to 25% better application build performance for Apple platform apps, available in US East and US West regions. <a href="https://app.daily.dev/posts/DEbhYteDE" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GNU Guix 1.5.0 released with RISC-V and Hurd support</strong>: GNU Guix 1.5.0 is out after a three-year cycle, featuring KDE Plasma 6.5, GNOME 46, rootless daemon security, 12,500+ new packages, and official support for RISC-V 64-bit and experimental x86_64 GNU Hurd kernels. <a href="https://app.daily.dev/posts/3NyywHGNR" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Zlib-rs 0.6 enhances AVX-512 support</strong>: Zlib-rs version 0.6, released by the Trifecta Tech Foundation, enhances AVX-512 support for improved performance on AMD Zen 4+ and Intel Xeon processors, integrating new gz* functions and hiding internal implementation details. <a href="https://app.daily.dev/posts/dIKTdK0cf" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>OpenTelemetry eBPF SIG 2026 roadmap targets 1.0 stability</strong>: The OpenTelemetry eBPF Instrumentation SIG’s 2026 roadmap aims for a stable 1.0 release, prioritizing broader protocol support (MQTT, AMQP, NATS), enhanced NoSQL database telemetry (MongoDB), and .NET support, alongside hybrid instrumentation and eBPF profiler integration. <a href="https://app.daily.dev/posts/mC6R9vti3" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>Anthropic CEO predicts AI will replace software engineers in 6-12 months</strong>: Dario Amodei, CEO of Anthropic, boldly declared at Davos that AI models could handle most software engineering tasks within 6-12 months, relegating humans to mere editors. He admits his own engineers already use AI to write code, but also concedes his timeline is, shall we say, ‘uncertain.’ So, while we’re all supposed to panic about our jobs, he’s hedging his bets. Classic AI hype, long on promise, short on verifiable specifics. <a href="https://app.daily.dev/posts/0fyNmKnQE" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI coding agents disrupt developer workflows and education business models</strong>: Jeffrey Wei claims AI coding agents have ‘drastically changed’ his development process, leading to a 40% staff reduction at his ed-tech platform because, apparently, nobody needs to learn to code anymore. He now enjoys programming more by acting as an ‘architect and reviewer’ of AI-generated code. It’s almost as if the AI is doing all the actual work, and he’s just there to make sure it doesn’t accidentally summon Cthulhu. What a time to be alive, where ‘productivity’ means less thinking and more reviewing AI’s questionable output. <a href="https://app.daily.dev/posts/KfJ8s7nDm" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Context transfer is the ‘main barrier’ for AI coding tools</strong>: Apparently, the ‘main barrier’ for AI coding tools in 2026 isn’t security or code quality, but ‘context transfer’—getting engineers’ tacit knowledge into AI systems. Because, you know, AI can write code, but it can’t quite grasp why that one obscure workaround from 2008 is still critical. So, instead of fixing the AI, we’re building ‘infrastructure’ to capture human ‘tacit knowledge.’ Sounds like a fancy way to say we’re still doing the AI’s homework for it. <a href="https://app.daily.dev/posts/DczaKkmGw" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
