<h2>Must know</h2>
<ul>
<li><strong>OpenAI goes all-in on ads, privacy be damned</strong>: OpenAI just launched ChatGPT Go, an $8/month subscription tier with ads, rolling out globally after an India debut. This move is a direct response to Google Gemini eating into their market share, and they’re projecting billions in ad revenue by 2029. While OpenAI assures users their conversations are private from advertisers, the entire strategy raises serious privacy and ethical questions, especially when funding AGI development with tactics traditionally associated with user data monetization, as noted by “The A in AGI stands for Ads” and “wtf is going on with ChatGPT?”. <a href="https://app.daily.dev/feed-by-ids?id=ArnlILqap&amp;id=ep7p1hnQy" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>China’s satellite internet ambitions dwarf Starlink, raising global concerns</strong>: China has filed plans with the ITU to launch nearly 200,000 satellites across two constellations, CTC-1 and CTC-2, making Starlink’s 49,000 look like a hobby project. This aggressive move, potentially linked to defense, aims to dominate orbital radio spectrum and has the U.S. FCC fast-tracking its own satellite approvals due to national security concerns. Meanwhile, Iran successfully jammed Starlink during a national blackout, degrading performance by 80% and highlighting the vulnerability of commercial satellite internet to state-level electronic warfare. <a href="https://app.daily.dev/feed-by-ids?id=9Dqo8zul1&amp;id=VVK9EFJ8w" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI is making the internet fake and software buggy</strong>: The “Dead Internet Theory” suggests most online content is now AI-generated, leading to a crisis of authenticity, with AI-generated code and comments appearing in open-source projects. Compounding this, AI code generation tools consistently reproduce a specific memory management bug, demonstrating that LLMs often mirror flawed patterns from their training data. This raises concerns about the reliability of AI-assisted content and the need for human oversight, as discussed on HackerNews and in articles detailing AI’s code generation flaws. <a href="https://app.daily.dev/feed-by-ids?id=JjrjV7tSv&amp;id=ZeB3UcRxG&amp;id=alpyX7cTa&amp;id=m1Q81Q3xQ&amp;id=mY0RAoEVz&amp;id=x9XQt7sEo" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Good to know</h2>
<ul>
<li><strong>Wine 11.0 stable release brings major improvements</strong>: Wine 11.0 is out with 6,300 code changes, a new NTSYNC kernel module for better sync, and a complete WoW64 architecture overhaul for 32-bit apps on 64-bit systems. <a href="https://app.daily.dev/posts/5LDn1X8et" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Wine now enables direct Photoshop installation on Linux</strong>: New Wine patches allow direct installation of Adobe Photoshop 2021 and 2025 on Linux by fixing XML parsing issues, a capability not seen since 2018. <a href="https://app.daily.dev/feed-by-ids?id=WW9lhhDTb&amp;id=jcudSc8Dd" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Critical vulnerability in n8n automation platform</strong>: A critical unauthenticated remote code execution vulnerability (CVE-2026-21858) in n8n allows attackers to read arbitrary files and gain admin access, affecting an estimated 100,000 servers. <a href="https://app.daily.dev/posts/4XP5JDLv1" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Google’s Antigravity expands beyond coding</strong>: Google’s Antigravity, powered by Gemini 3 Flash, is now competing with Anthropic’s Co-work by adding productivity features like creating Google Slides and organizing files directly from the IDE. <a href="https://app.daily.dev/posts/HVlI4ECqg" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Jakarta EE 12 development and community events update</strong>: The Jakarta EE Platform project is progressing towards Milestone 2 for Jakarta EE 12, with a target date of January 27, 2026, and several community conferences scheduled. <a href="https://app.daily.dev/posts/YrhkKGKoN" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>ChaosBSD: A new BSD for experimental hardware</strong>: ChaosBSD is a new FreeBSD fork designed as a testing ground for experimental, broken, or incomplete drivers and hardware support that aren’t ready for upstream FreeBSD. <a href="https://app.daily.dev/posts/gu1ft7jZB" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>Coinbase scaled hiring to 150 engineers per month</strong>: Coinbase achieved massive hiring scale by centralizing and standardizing their entire recruitment process, treating it like an engineering problem with data-driven optimization. <a href="https://app.daily.dev/posts/Sapiwmv2G" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
<h2>Bullshit police</h2>
<ul>
<li><strong>AI companies measuring success by electricity consumption</strong>: Apparently, AI companies are now touting their ‘scale’ by how much electricity they consume. As Justin Searls points out, this is about as useful as steel companies bragging about CO2 emissions. It creates perverse incentives where inefficiency is celebrated, and when more efficient AI comes along, they’ll have to find a new, equally meaningless metric. <a href="https://app.daily.dev/posts/OHMKEG36O" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>GitHub contribution graphs are meaningless</strong>: Someone finally said it: your GitHub contribution graph, LinkedIn profile, and LeetCode challenges are terrible indicators of developer worth. The author argues these metrics fail to capture real value, especially across different job markets. It’s almost like real human assessment and trust are more important than a green square grid. <a href="https://app.daily.dev/posts/WQGtqJsDu" target="_blank" rel="noopener nofollow">Read more</a></li>
<li><strong>AI coding agents are ‘good enough’ to question learning fundamentals</strong>: A developer, after 20 years of Vim/Neovim, is now wondering if learning tools like awk and sed still makes sense because AI coding agents are ‘good enough.’ This perfectly encapsulates the modern tech dilemma: trading the joy of craft for efficiency, only to realize you’re just pushing buttons while a machine does the actual thinking. <a href="https://app.daily.dev/posts/vFzyyNK3z" target="_blank" rel="noopener nofollow">Read more</a></li>
</ul>
