Mikeal Rogers has died (h4x.zip | comments) Mikeal Rogers, a key figure in the Node.js community, has passed away from aggressive cancer. He was known for his contributions, including transforming developer evangelism into an act of hospitality, and for his role at DigitalOcean.

• oh gosh, I have so many great memories of Mikeal due to the wonderful javascript community, and am really sad to hear he has passed. I hope others were able to be delighted by spending time with him, as I was. maybe someone else who gets the inside joke will built a naanument to him.

Another Crack in the Chain of Trust: Uncovering (Yet Another) Secure Boot Bypass (binarly.io | comments) Binarly has disclosed a security vulnerability, designated CVE-2025-3052, that affects most Unified Extensible Firmware Interface (UEFI) devices. This vulnerability allows attackers to execute unsigned code prior to the operating system loading, thereby compromising the device’s security at a critical boot stage.

• > Because the attacker’s code executes before the operating system even loads, it opens the door for attackers to install bootkits and undermine OS-level security defenses. Excellent.
• > The root cause of this bug, once again, lies in the unsafe handling of NVRAM variables. Sheer incompetence, in other words.

"Localhost tracking" explained. It could cost Meta €32B (zeropartydata.es | comments) Recent discussions highlight that the practice of 'localhost tracking' could lead to significant financial repercussions for Meta, potentially costing the company up to €32 billion. This tracking method has raised concerns about privacy and regulatory compliance, sparking debates on its implications for digital advertising.

• This is quite an interesting read. But if Android does not allow listening to local host ports, how did meta achieve it?
• "The Meta Pixel script sends the _fbp cookie to the native Instagram or Facebook app via WebRTC (STUN) SDP Munging." Crazy to deploy a hack like this at the scale of Meta.

Ada Continues to Climb in June Tiobe Index and PYPL (ada-lang.io | comments) Ada programming language sees a rise in popularity as indicated by its climbing rankings in the TIOBE Index and the PYPL (PopularitY of Programming Languages) index for June. These trends suggest a growing interest in Ada, traditionally used in systems programming and critical applications, spotlighting its relevance in the current technology landscape.

• I've only ever seen Ada used in a military-related product, and there was one person on the team responsible for maintaining it. They didn't like doing it, but it was a legacy product, and they were the only one who had learned enough Ada to do it.
• This is nice, I like Ada, but why is this happening now, is there any explanation for the blip?

Apple has announced its final version of macOS for Intel (tedium.co | comments) Apple has revealed its final version of macOS for Intel processors, marking the end of an era for support that lasted two decades. With this announcement, the use of Hackintoshing—running macOS on non-Apple hardware—is likely to diminish significantly.

• The article claims T2Linux has "largely solved" the problem of Linux on T2 Macs. Is that true? I have a 2020 MBP with touch bar & 2 Thunderbolt ports. My reading of the T2Linux project sounded a lot more negative than that, but I'd love to be shown I'm wrong.

Fine-tuning LLMs is a waste of time (codinginterviewsmadesimple.substack.com | comments) The text critiques the efficiency and necessity of fine-tuning Large Language Models (LLMs) for tasks such as knowledge injection, suggesting that many practitioners misunderstand its value. It argues that resources spent on fine-tuning could be better utilized, indicating a need for a reevaluation of current practices in LLM usage.

• It's pretty frustrating to spend weeks on finetuning and end up with a model that says: "SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT SELECT ..."

It's the end of observability as we know it (and I feel fine) (honeycomb.io | comments) The blog post outlines the evolution of observability tools, highlighting a shift caused by advancements in large language models (LLMs), which are set to challenge and transform the existing paradigms in monitoring and analytics over the last decade.

• LLMs won't replace observability, but they absolutely change the game. Asking "why is latency spiking" and getting a coherent root cause in seconds is powerful. You still need good telemetry, but this shifts the value from visualizing data to explaining it.

The librarian immediately attempts to sell you a vuvuzela (kaveland.no | comments) Entering a vast library, one is greeted by a pushy librarian eager to sell vuvuzelas, mirroring the annoyance of misleading ads in internet searches. The library symbolizes the web, cluttered with low-quality content and SEO tactics, contrasting with past experiences of finding valuable information. This shift raises concerns about the effectiveness of modern search engines.

• So why aren't real libraries like this? Can you fund the internet the same way?
• > AIs are much better at responding to my intent, and they rarely attempt to sell me anything Yet. It's only a matter of time before AI becomes ad-riddled and enshittified.

The Gentle Singularity (samaltman.com | comments) Sam Altman warns we are entering a pivotal stage in AI development, with digital superintelligence on the horizon. Current systems outsmart humans in various tasks, amplifying productivity. By the 2030s, advancements will significantly transform creative and scientific output, potentially making intelligence and energy universally accessible. While challenges like job displacement persist, rapid technological progress could foster new economic opportunities and innovative policies.

• I started quickly reading the article without reading who actually wrote it. As I scanned over the things being said, I started to ask myself: Who wrote this? It's probably some AI proponent, someone who has a vested interest. I had to smile when I saw who it was.

Low-background Steel: content without AI contamination (jgc.org | comments) A blog entry details the creation of a site dedicated to 'low-background steel', designed to be a resource-free platform devoid of artificial intelligence (AI) influences. The author reflects on the necessity of such a clearinghouse for genuine content without the contamination of AI-generated material, emphasizing the site's relevance since its launch in March 2023.

• Ai generated content is inherently a regression to the mean and harms both training and human utility. There is no benefit in publishing anything that an AI can generate, just ask the question yourself. Maybe publish all AI content with <AI generated content> tags, but other than that it is a public nuisance much more often than a public good.

Malleable software: Restoring user agency in a world of locked-down apps (inkandswitch.com | comments) The essay critiques the current state of software as rigid tools that hinder user creativity and adaptability. Instead of empowering users to shape their environments, today's software often forces them to conform to predefined constraints. The authors propose a vision of 'malleable software' that allows users more control and customization, fostering a collaborative and creative computing landscape.

• Ink & Switch is doing great (some would say overdue) research that’s on the boundary of commercializability but outside the bounds of what the big corporates want to do with computers. Great to see them pushing work like this, building experiments, and talking about what they’ve learned.

AI Saved My Company from a 2-Year Litigation Nightmare (tylertringas.com | comments) After enduring over two years of litigation, Tyler Tringas's company, Calm Company Fund, managed to resolve the lawsuit favorably by leveraging artificial intelligence (AI). Tringas shares how AI not only streamlined their legal strategy but also provided a competitive edge against well-resourced adversaries, transforming a stressful experience into a learning opportunity.

• I think this article really buries the lead—what really allowed him to win was leverage. He says that’s an over beers info, not a blog post. But that’s actually what let him win—not learning from the LLM how to understand his lawyers arguments better.

Successful people set constraints rather than chasing goals (joanwestenberg.com | comments) Joan Westenberg reflects on the shift from conventional goal setting to establishing personal constraints, emphasizing the importance of limits for effective achievement. She relates a personal experience in which pursuing a promotion felt unfulfilling and suggests that creating boundaries can foster more meaningful progress.

• I have a google doc called "The No Project". Every day I try and add something that I have said "no" to. Projects, feature requests etc. I don't always have an entry, but it keeps No top of mind. Not exactly a constraint, but....
• I have limited myself to not setting any constraints.

Show HN: I made a 3D printed VTOL drone (tsungxu.com | comments) Tsung Xu's 3D-printed vertical take-off and landing (VTOL) drone represents a significant engineering feat, capable of flying up to 130 miles without prior experience in computer-aided design (CAD), 3D printing, or aerodynamics modeling. This project exemplifies innovative design in the hardware domain.

• This is incredibly impressive. I'd love to hear more about what relevant skills & knowledge you started with, and how you worked out what you'd need to know to complete the project. How much customisation did you need to make to Ardupilot? Is your drone's control unique, or somehow standardised?

Show HN: PyDoll – Async Python scraping engine with native CAPTCHA bypass (github.com/autoscrape-labs | comments) PyDoll is an asynchronous Python scraping engine designed for high efficiency, equipped with built-in CAPTCHA bypass capabilities. This productivity-enhancing tool aims to simplify the web scraping process, making it more robust against common challenges encountered in automated data extraction.

• Checking it out and I see you're using CDP. It's been a bit, but I'm pretty sure use of CDP can be detected. Has anything changed on that front, or are you aware and you're just bypassing with automated captcha handling?
• Hi, just wondering what you're thinking about how your tool might be abused.

Android 16 is here (blog.google | comments) Android 16 introduces enhanced productivity, improved security, and a range of new features aimed at optimizing user experience. This update continues Android's evolution, focusing on streamlining device performance and providing innovative functionalities for better management of tasks.

• The notification feature looks nice. I've pretty much exclusively used iOS, but honestly notifications is a weak point for iOS, in my opinion. I frequently have the 1 notification on my home screen, but once unlocked notifications are pretty much impossible to find again.

Magistral — the first reasoning model by Mistral AI (mistral.ai | comments) Mistral AI has launched 'Magistral', its inaugural reasoning model designed to enhance decision-making processes. This model integrates advanced reasoning capabilities, improving its utility in various applications, particularly in artificial intelligence solutions that require logical inference.

• Benchmarks suggest this model loses to Deepseek-R1 in every one-shot comparison. Considering they were likely not even pitting it against the newer R1 version (no mention of that in the article) and at more than double the cost, this looks like the best AI company in the EU is struggling to keep up with the state-of-the-art.

WWDC25: macOS Tahoe Breaks Decades of Finder History (512pixels.net | comments) macOS Tahoe, revealed at WWDC25, marks a significant shift in Apple's Finder interface, notably featuring a reversed Finder icon. This change diverges from decades of tradition, sparking discussions among users about its implications on usability and design aesthetics.

• The color change doesn't bother me nearly as much as the change in proportions. Insetting the face on the right and removing its curvature makes the smile look super weird and less "human" to me. I'm sure I'll get used to it, but it feels a little off.

Denuvo Analysis (connorjaydunn.github.io | comments) Denuvo is an advanced digital rights management (DRM) system widely recognized for its effectiveness in preventing piracy in video games. The analysis delves into Denuvo's unique semi-online mechanisms, such as hardware identification, license file exchanges, and integrity checks, detailing its protection methods against reverse engineering.

• Nothing a well-trained model won't be able to instantly solve. It's literally just grunt work, not rocket science. F DRMs though. Good news is those AAA games are rarely worth anyone's time anyways. Better spin up indies or classic games - a good SNES game is worth a hundred of those garbo AAA license rehashes.

We’re secretly winning the war on cancer (vox.com | comments) Cancer deaths in the U.S. have decreased by 33% since 1991, equating to 4 million lives saved. Advances in treatments, screening, and prevention, including vaccines like the HPV vaccine, have been pivotal. This change reflects a significant improvement in survival rates, turning cancer from a nearly certain death sentence to a manageable condition for many.

• My dad was diagnosed with multiple myeloma 2 years ago. His bone marrow transplant failed (frequent first line of defense) and he just finished CAR-T therapy a couple months ago. The initial side effects from the treatment were _bad_, but everything is looking good right now. CAR-T is really mindbogglingly insane cyberpunk stuff.