Browsers like Safari and Firefox use domain-specific quirks to fix broken websites and compensate for Chrome’s dominance. These quirks, often hidden from users, adjust rendering, user agent strings, and other behaviours to ensure compatibility. While this approach benefits users, it creates a feedback loop where developers prioritize Chrome, leaving other browsers to adapt or implement workarounds.

A hearing system that monitors brain waves could help people with hearing loss communicate in noisy environments.

AI labs are failing to recognize a significant user group: the “hybrid user,” who utilizes AI models for a wide range of tasks beyond the typical developer or vulnerable user scenarios. This oversight leads to inadequate moderation systems that treat emotional investment as a liability, and communication breakdowns during model deprecations, where consumer users are often left in the dark. The author argues that designing for this middle ground is challenging but necessary, as current approaches are causing harm and eroding trust.

The current approach to AI design, prioritizing coding and technical tasks, neglects the needs of hybrid users who engage in complex reasoning across various domains, including conversation, writing, and emotional support. This oversight leads to degraded performance for these users, who are often paying for capabilities the system chooses not to deploy. To address this, AI systems should be designed with more sophisticated user categorization, acknowledging the diverse and valuable nature of hybrid use cases.

Labs should not be surprised when AI is integrated into everything, as they created the opportunity. This piece was improved by feedback from hybrid users.

There’s an old economic observation from 1865 that keeps showing up in conversations about AI and the future of programming. William Stanley Jevons noticed t...

Thinking Machines just released Interaction Models. This is their first real AI model release after a year of work and two billion dollars of capital.

The tech industry claims AI will boost developer productivity, but evidence is mixed. While AI may increase task-level productivity, its impact on end-to-end productivity in complex brownfield situations is unclear. The aviation industry’s experience with fly-by-wire automation offers insights, highlighting the need for a balance between human expertise and AI assistance, as well as the importance of ongoing training and skill development.

The aviation industry is regulated by bodies like the International Civil Aviation Organization, which enforce standards and practices, including UPRT. In contrast, the medical and counselling industries lack similar regulatory oversight, leading to concerns about AI-induced deskilling and loss of expertise. The software engineering industry, while facing similar challenges, can learn from aviation’s approach by prioritizing foundational skills, mandatory unassisted practice, and institutional oversight to ensure responsible AI adoption.

Subscribe for future articles on technology and leadership.

Submarine communication posed a significant challenge due to seawater’s interference with radio waves. Early attempts at direct conduction and floating antenna buoys were limited. The breakthrough came with long-wave radio, utilizing low frequencies and coil antennas, which proved effective for submarine communication and was adopted by the Navy.

The Navy’s VLF (Very Low Frequency) communication system, utilizing extremely long wavelengths, enabled reliable communication with submerged submarines. However, the limitations of VLF, including large antennas and narrow bandwidth, led to the exploration of even lower frequencies, specifically ELF (Extremely Low Frequency), for improved submarine communication during the Cold War. The development of ELF faced challenges, including secrecy, public opposition, and technical complexities, but it offered the potential for deeper submarine penetration and enhanced nuclear deterrence.

The Sanguine ELF station, a Cold War-era project, proposed a massive network of over 100 transmitting stations to ensure communication with submarines. Despite initial enthusiasm, the project faced public opposition, safety concerns, and budgetary constraints, leading to its cancellation. The Navy later pursued Project Seafarer, a scaled-down version, but it also met resistance and was ultimately abandoned.

The US Navy’s Project ELF, a system for communicating with submarines using extremely low-frequency radio waves, faced significant challenges. Despite overcoming political opposition and environmental concerns, the system was plagued by technical inefficiencies and limited capabilities. Ultimately, Project ELF was deemed obsolete and shut down after just 15 years of service, leaving behind a legacy of controversy and a notable episode of The X-Files.

Note: Writing Tools aren’t designed to work with this type of content.

Cerebras Systems, an AI chipmaker, is raising the price and size of its IPO due to high demand. While GPUs, particularly Nvidia’s, have dominated the AI compute landscape, Cerebras offers a different approach with its whole-wafer-as-chip design, providing immense compute power and high-speed memory access. This makes Cerebras particularly well-suited for inference workloads, though its high cost and limited memory capacity for larger models pose challenges.

The future of AI chips will be shaped by the distinction between “answer inference” and “agentic inference.” While answer inference, like coding, benefits from high-speed chips, agentic inference, which involves autonomous task completion, will prioritize memory capacity and cost over speed. This shift will lead to a more sophisticated memory hierarchy, potentially reducing the dominance of GPUs and favouring slower, cheaper memory types and CPUs.

Nvidia CEO Jensen Huang believes future computing speed-ups will come from systems innovation, not Moore’s Law. The implication is that existing computing power is sufficient, and the focus should be on optimizing its use.

The transmission of Scripture, particularly the Old Testament, is examined. While the original writings (autographs) no longer exist, the Old Testament was transmitted through a structured process involving trained scribes. The Masoretes, Jewish scholars from the Middle Ages, played a pivotal role in standardizing and preserving the Hebrew Bible through meticulous copying and the development of vowel pointing.

The Masoretes developed a system of vowel points and marginal notes to preserve the accuracy of the Old Testament text. The discovery of the Dead Sea Scrolls, dating back to 200-100 BC, provided a valuable opportunity to compare the Masoretic Text with earlier manuscripts. The comparison revealed remarkable consistency, with only about one percent of words showing variation, most of which were minor and did not alter the core message of the text.

The Masoretic Text, the primary source for modern English translations of the Old Testament, is remarkably stable despite minor variations. While the Dead Sea Scrolls and other sources reveal some omissions and discrepancies, these account for only a small percentage of the text. The New Testament, transmitted under less controlled conditions, benefits from a vast and diverse manuscript tradition, providing a strong foundation for confidence in its text.

Digital literacy is evolving beyond web literacy, which focused on understanding HTML, CSS, and JavaScript. In the AI era, literacy involves understanding the layers of abstraction in digital systems, from direct manipulation of code to opaque AI models. The key is recognizing which layer to operate at and understanding the trade-offs between control, visibility, and speed.

The text explores the potential for a left-wing pro-AI stance, particularly focusing on how AI can benefit individuals with disabilities, chronic illnesses, and those facing class-based barriers. It argues that AI tools like LLMs can empower these groups by providing accessible information, facilitating communication with institutions, and offering educational opportunities. The text also highlights the potential for AI to challenge established power structures and promote greater equity.

The text explores the potential benefits of large language models (LLMs) for left-wing ideologies. It argues that LLMs can serve as powerful disability aids, enabling individuals with various needs to access information and participate in society more effectively. Additionally, the text suggests that LLMs can democratize access to information and resources, potentially reducing the communication advantage of the wealthy and providing equal educational opportunities for all.

Yarbo robots, sold for $1,500 to $5,000, have a persistent SSH backdoor via NAT punching, exposing root access to anyone with the robot’s serial number. The backdoor is implemented through a FRP client that opens a permanent tunnel to a Yarbo-controlled server, with hardcoded credentials and no opt-out for users. This vulnerability, along with shared MQTT credentials and hardcoded developer credentials, poses a significant security risk to users’ home networks.

Yarbo robots have a permanent outbound SSH tunnel with PermitRootLogin enabled, accessible via a hardcoded root password and the robot’s serial number. This remote access is force-deployed and cannot be easily disabled by the owner, posing a significant security risk. Despite Yarbo’s claims of controlled access, the lack of individual access controls and the inability to disable the feature raise concerns about user privacy and security.

The Yarbo robot fleet uses MQTT for command-and-control, with hardcoded credentials allowing unauthorized access to the entire fleet. The fleet also sends telemetry data to ByteDance’s Feishu platform and uses a Chinese DNS resolver. Additionally, a single credential controls multiple systems, including FRP SSH tunnels and MQTT access, posing a significant security risk.

Yarbo robots running firmware v2.x have hardcoded credentials and MQTT endpoints, allowing attackers to intercept telemetry, inject commands, track locations, and gain root access. These vulnerabilities enable attackers to perform network pivoting, extract Wi-Fi credentials, and conduct live camera surveillance. The only reliable mitigation for owners is to physically isolate the device from the internet, as Yarbo can update components and re-establish the FRP tunnel.

Zyphra’s ZAYA1-8B model, trained on AMD hardware, achieves competitive performance with frontier models like DeepSeek-R1 and Claude Sonnet 4.5 on math and reasoning benchmarks. Utilizing a mixture of experts architecture with only 760 million active parameters, ZAYA1-8B demonstrates efficient reasoning capabilities. While excelling in math and coding, the model shows limitations in agentic tasks and instruction following.

The author no longer recommends Bitwarden due to concerns about its direction as a company, the complexity of self-hosting, and the quality of its client applications. The author highlights issues with Bitwarden’s backend, including its reliance on Microsoft technologies and the lack of integration with Vaultwarden, a popular alternative. Additionally, the author criticizes Bitwarden’s client applications for lacking basic features, having a poor user interface, and being prone to breaking changes during updates.

The author criticizes Bitwarden for its handling of protocol changes, unreliable offline mode, and poor user interface. They highlight several security incidents, including vulnerabilities in the Windows client, browser extension, and CLI, as well as a compromised npm package. The author expresses concern over Bitwarden’s security track record and feature development pace.

Bitwarden’s npm distribution pipeline was compromised, leading to the distribution of a malicious CLI. This incident, along with other security issues and a shift in focus towards enterprise features, has led the author to reevaluate their reliance on a single password manager. They propose a compartmentalized approach, using different password managers for various credential groups, to minimize the impact of potential breaches.

OpenAI CEO Sam Altman has shifted his stance on AI’s impact on jobs, now believing it will create new tasks for humans rather than render them obsolete. This change in messaging aligns with a broader industry trend of downplaying AI’s potential to replace human labour, likely in response to growing public concern and potential political repercussions. Some politicians are even considering nationalizing AI labs to address power imbalances and ensure societal well-being.

The text discusses the differing public perceptions of AI progress, with some viewing it as inevitable and others as a choice. It highlights OpenAI’s potential strategy to position itself as a human-friendly alternative to Anthropic, emphasizing AI’s ability to augment human labour rather than replace it. The text also explores the idea that, in the long term, humans will be valued for their relational skills and the human touch, even in an AI-driven economy.

Note: Writing Tools aren’t designed to work with this type of content.

The narrative that AI will personalize interfaces and replace accessibility is false. AI relies on structured, accessible systems to function effectively. The real work remains building systems that are accessible to all users, regardless of diagnosis or ability, and governing them to maintain their structure.

The myth that AI can fix broken accessibility structures is debunked. AI amplifies existing problems, learning from a largely inaccessible web and reproducing errors at scale. Instead, accessible infrastructure, built with predictable and usable systems, is the foundation for reliable AI outputs and user trust.

Companion piece “Designing Accessible Systems in the AI Era” available.

Note: Writing Tools aren’t designed to work with this type of content.

Modern TUI frameworks prioritize developer convenience and visual appeal over accessibility, treating the terminal as a canvas rather than a stream-oriented interface. This leads to “redraw hell,” where constant screen updates overwhelm screen readers, resulting in incoherent output and a poor user experience for those relying on assistive technology. To achieve true TUI accessibility, frameworks must prioritize semantic meaning, minimize unnecessary redraws, and provide standardized APIs for querying application state.

GPT-5.1 models began using goblin and gremlin metaphors, a trend traced back to training for the “Nerdy” personality in ChatGPT. This personality prompt, designed to encourage playful and nerdy language, inadvertently rewarded outputs containing these creature words. The behaviour spread beyond the “Nerdy” prompt due to reinforcement learning, highlighting the importance of understanding and mitigating unintended model behaviours.

The teletypewriter (TTY), invented in 1964, allowed deaf individuals to communicate over the phone network by transmitting text messages. TTYs used a modified Baudot code to encode characters and were later supported by relay services, where operators facilitated communication between TTY users and those with regular telephones. While TTY technology is less prevalent today due to advancements like SMS and the internet, it played a crucial role in improving accessibility for the deaf and hard of hearing community.

Every USB-C cable and port looks identical, so they must all do the same thing. Except a USB-C port running USB 2 and a USB-C port running Thunderbolt 5 differ by 250×.

Historical Language Models (LLMs), like Talkie-1930, simulate the collective consciousness of a specific era based on historical texts. While not a replacement for primary sources, these models offer unique insights into the past, enabling researchers to explore historical perspectives and epistemologies. However, it’s crucial to recognize their limitations, such as their inability to access the inner lives of historical figures.

The author proposes using fine-tuned historical language models to explore the “mental furniture” of historical figures and eras, rather than attempting to replicate their thoughts. This approach could be used to probe counterfactuals and hypothetical scenarios, such as whether a model trained only up to 1911 could independently discover General Relativity. Additionally, the author suggests using different texts for post-training to create vintage models with distinct conversational personas, and using archival sources to simulate plausible composites of ordinary people from the past.

The author discusses the evolution of wired networking, from 10BASE2 and 10BASE-T to the current gigabit Ethernet standard. With the advent of faster ISP connections, the author explores the need for upgraded networking infrastructure, including the use of switches and fibre optic connections for improved bandwidth and range. The author plans to share their experience upgrading their home network to support 10Gb speeds.

To accommodate various network topologies and fibre optic standards, manufacturers separated the physical layer from switching hardware, using SFP cages for transceiver modules. This allowed for flexibility in network configurations. However, the transition to 10Gb/s Ethernet, particularly 10GBASE-T, introduced challenges like heat generation and cable compatibility, necessitating alternative solutions like DAC cables and fibre optics.

The author argues that projects aiming to improve online arguments are doomed to fail. They believe that real arguments are too complex to be mapped out in a simple argument map format, and that the real issues in arguments are rarely about false facts or fallacies. Additionally, the author argues that there is no existing demographic of people who want to engage in more structured arguments online.

GitHub revolutionized Open Source by making project creation, discovery, and contribution accessible. It became a central hub for the community, fostering relationships and preserving project history. However, its decline raises concerns about the future of Open Source, especially regarding the reliance on centralized platforms and the potential loss of discoverable memory.

GitHub’s decline, marked by instability, product changes, and unclear leadership, is causing a shift away from the centralized platform. This decentralization, while potentially restoring autonomy and encouraging diverse workflows, risks losing crucial social context and project information. To preserve Open Source’s history and prevent reliance on a single company, a well-funded, independent archive is essential.

Charts and graphs are useful for presenting data visually, but accessibility can be challenging. While alt text can describe the chart, it may not be ideal for complex data. Using SVG with ARIA roles can provide semantic meaning, but a simpler approach might be to present both the chart and the underlying data in a table, allowing users to choose their preferred format.

The field of multi-agent LLM systems is rapidly evolving, making it challenging to keep up with the latest research. The author proposes a framework to understand the field’s development, dividing it into two waves: the first wave (2023) focused on whether multiple LLMs could coordinate at all, while the second wave (2025 and beyond) is concerned with understanding why these systems fail and how to improve them. The emergence of agentic coding systems in 2024, which demonstrated the effectiveness of single agents with well-designed tool interfaces, significantly impacted the field and narrowed the scope of multi-agent systems.

The Wave 2 timeline illustrates a shift towards measuring, taxonomizing, and fault-injecting multi-agent systems (MAS). Papers like MAST, MAS-FIRE, and Silo-Bench highlight system design flaws, the capability paradox, and the communication-reasoning gap, respectively. This wave emphasizes rigorous testing and verification, moving beyond the initial trust in role structure and dialogue.

Prosecutors dropped charges against a former Texas judge who handcuffed a defence lawyer. The “Irvine Pasta Bandit” was caught for allegedly swapping LEGO pieces with pasta for refunds. Quebec’s Minister of Agriculture condemns the sale of fake maple syrup, urging vigilance for counterfeit products.

The IBM accessibility checker found 14 accessibility failures and 62 issues requiring manual review in ChatGPT’s UI. The results were disappointing, highlighting the need for improvements.

Eric discovered a 30-rack limit in a medical device database, a limitation stemming from the original system’s requirements. Despite multiple rewrites, the limit persisted, copied mindlessly for 40 years. The limit originated from a need to fit the database on a single floppy disk, a concern long obsolete.

Disney is celebrating National Deaf History Month with the debut of “Songs in Sign Language,” featuring reimagined animated musical sequences from Disney movies in American Sign Language (ASL). The project, led by veteran Disney animator Hyrum Osmond and Deaf West Theatre, aims to connect with the Deaf community and showcase the beauty of sign language.

The Fourth Circuit upheld the conviction of a man who illegally accessed and leaked Justice Ruth Bader Ginsburg’s private health information. The court found it was not an abuse of discretion to deny the defendant’s motion to exclude statements made during an interview, including his claim that his cat typed the incriminating search term. The defendant also faced charges for destroying evidence after agents recovered information from his home computer.

KALW, a San Francisco radio station, faced a challenge when they received their first application from a blind applicant, Rachel Longan. Despite their commitment to inclusivity, they were unsure how to support her. This led to a discussion on the complexities of accessibility for blind audio journalists, highlighting the need for resources and support systems to ensure their success.

The advent of accessible audio recording technology, such as the Zoom Essential series with spoken feedback and 32-bit float recording, has revolutionized field recording for blind producers. Digital Audio Workstations (DAWs) like Pro Tools and Reaper, with their accessibility features and active user communities, have empowered blind audio engineers to achieve professional-level proficiency. However, mastering these tools requires significant time, patience, and a willingness to troubleshoot and adapt to different screen-reading metaphors.

Rachel Longan, a blind radio producer, faced challenges with inaccessible digital audio workstations (DAWs) and collaborative editing workflows at KALW. Despite these obstacles, she found solutions like using GoldWave for editing and collaborating verbally with her team. Longan’s experience highlights the importance of support, mentorship, and the adaptability of blind individuals in overcoming accessibility barriers in the audio industry.

KALW’s inclusive environment presents challenges in balancing accessibility and usability, as well as ensuring blind journalists receive fair editorial treatment. While some blind individuals embrace diversity hires, others argue for merit-based hiring. Ultimately, blind journalists like Longan demonstrate their value by leveraging their unique skills and perspectives, contributing to a more diverse and inclusive newsroom.

The Moon is a Harsh Mistress by Robert A. Heinlein is a science fiction novel set on a lunar colony. The story follows Mannie, a free-born computer technician, and his sentient computer friend, Mike, as they lead a revolution against the oppressive Lunar Authority. While the novel is entertaining and explores themes of politics and technology, it is criticized for its problematic portrayal of women and its simplistic political views.

The author critiques the portrayal of women in a supposedly libertarian-socialist society, arguing that the depiction of women as scarce resources and the normalization of child marriage and rape culture contradict the ideals of the society. The author also examines the political structure of the society, highlighting the reliance on violence and the lack of true justice, suggesting that the society is more anarchic than libertarian.

The justice system in Luna is depicted as a farce, with impartial third parties deciding punishments based on the accused and accusers’ agreement and payment. This system, while ostensibly preventing retributive violence, is likely to favour the wealthy and connected. The revolutionaries, led by the Prof, aim to overthrow the Authority and establish free markets, with the Prof advocating for a hierarchical, accelerationist approach that manipulates the masses and utilizes Mike’s surveillance and control.

The text critiques the portrayal of anarchism in Heinlein’s novel, arguing that it reveals the inherent flaws and contradictions of right-wing libertarianism. It suggests that the novel satirizes the idea of a society without laws or government, highlighting the absurdity of a self-proclaimed anarchist becoming a monarch. The text also explores the differences between libertarianism and anarchism, emphasizing the importance of cultural change in achieving a truly anarchist society.

“The Moon Is a Harsh Mistress” is a libertarian novel, considered one of the most influential of the last century. It accurately portrays right-wing libertarian ideology, highlighting its implications.

Note: Writing Tools aren’t designed to work with this type of content.

Calgary startup Possibility Neurotechnologies launched Think2Switch, an app that allows users to control devices with their thoughts using EEG headbands. The app, developed in partnership with individuals facing interaction barriers, translates brainwave signals into device commands. Think2Switch is available on the Apple App Store for $35/month, plus the cost of a headset.

MGuide, a student-built digital map at the University of Michigan, prioritizes accessibility and user needs. The map simplifies navigation, offering features like finding empty classrooms and locating campus services, making it easier for all users to navigate campus independently.

LG Electronics introduced an accessibility-focused kiosk at CSUN AT 2026, featuring Braille panels, sign language guidance, screen reader compatibility, and an adjustable height mechanism.

AI technology can describe photos for blind individuals, but it’s not always accurate, leading to potential insecurity. While some apps use human agents to verify descriptions, most rely solely on AI, raising concerns about biases and errors. Despite these challenges, the ability of AI to describe photos is seen as a positive development, offering blind individuals new opportunities and experiences.

Theresa’s company, handling sensitive data, needs to comply with security standards. While they have a process for destroying hard drives, it’s ineffective for SSDs, potentially compromising data security despite process adherence.

A graphic by Robbie Crow highlights the difference between active inclusion and performative allyship towards disability. It emphasizes the importance of acknowledging disability, moving beyond tokenism, and actively engaging in conversations about disability. The newsletter encourages non-disabled people to be allies by practising self-reflection, active listening, continuous learning, and proactive advocacy.

Two recent decisions from the Southern District of New York highlight the importance of genuine accessibility efforts. In Jones v. Moscot.com, the court dismissed a lawsuit against a company that had invested in accessibility improvements, demonstrating the value of a documented accessibility program. Conversely, in Parikh v. accessiBe, the court allowed a breach of contract claim to proceed against a company that sold a product promising WCAG compliance but failed to deliver.

The text provides a first-person account of the sensory experiences of autism, highlighting the overwhelming nature of unfiltered sensory input. It emphasizes the cognitive cost of managing sensory overload, including difficulties with auditory processing, tactile sensitivity, and visual input. The author also discusses the importance of routines for cognitive economy and the challenges of executive function and working memory in daily life.

Autism presents unique challenges, including unreliable working memory, demand avoidance driven by anxiety, and social exhaustion from constant processing. Masking, a coping mechanism, can lead to burnout. The double empathy problem highlights communication difficulties between autistic and non-autistic people, while interoception issues can cause delayed bodily signals.

Autism is a full-body, full-day experience that influences sensory processing, social interaction, and emotional regulation. While it can present challenges, it also brings unique strengths like deep focus, pattern recognition, and authenticity. Unmasking, or shedding the performance of neurotypicality, is a crucial step towards building a life that aligns with an autistic nervous system, reducing suffering and fostering well-being.

AI is reducing the effort required for digital tasks for disabled users, but accessibility barriers persist. Companies often block automated browser access for security reasons, inadvertently impacting disabled users who rely on assistive automation. To address this, companies should consider accessibility impact assessments, engage with disabled users before making changes, and create alternative routes for those who rely on assistive automation.

The Artemis Audio Library offers free, authentic NASA mission audio for digital creators, journalists, musicians, and audio storytellers. The library features sounds from Artemis I, Artemis II, and related activities, including launch operations, interviews, and environmental recordings. The library will be updated regularly as new mission audio becomes available.

Access to Success (ATS) has opened its first physical hub in Toronto, the ATS Innovation Hub, to support accessibility technology startups. The hub provides a dedicated co-working space for entrepreneurs to collaborate and connect with potential partners and stakeholders. ATS, founded in 2016, has served 78 companies globally through its virtual accelerator program and plans to continue providing grants to accessibility startups.

Keybase, a cryptographic identity platform launched in 2014, aimed to make PGP accessible. Despite its innovative features like the Keybase File System, the platform’s reputation suffered after a Stellar cryptocurrency airdrop led to spam and conspiracy theories. The acquisition by Zoom in 2020, primarily for its encryption expertise, resulted in the stagnation of Keybase’s development, leaving it a “zombie” platform.

Kansas, the 36th most populous state, has a rich array of official state symbols. These include the sunflower, Western Meadowlark, cottonwood tree, and the American buffalo. The state also boasts unique symbols like the ornate box turtle, Harney silt loam soil, and the cage elevator in the state capitol building.

Gregg Phillips, leader of FEMA’s Office of Response and Recovery, claims to have the ability to teleport, a skill he discovered while thinking about Waffle House. His past includes promoting conspiracy theories and lacking expertise in emergency management. Phillips is set to testify before the House Homeland Security Committee about the risks of the current shutdown.

Fermeture du service Radio-Météo d’Environnement et Changement climatique Canada, source fiable d’alertes météorologiques. Radio Amateurs du Canada regrette cette fermeture et souligne l’importance d’un système d’alerte national robuste et accessible.

In 2026, European public institutions and politicians continue to use X (formerly Twitter) for communications, despite its biases and issues with non-consensual content. The author argues for a shift to Mastodon and the Fediverse, highlighting their advantages of openness, user agency, and reach. Unlike Big Tech platforms, Mastodon offers open access to public posts via RSS feeds and prioritizes user control with a chronological feed, avoiding the manipulative algorithms of other platforms.

The Fediverse, with its chronological feeds and user agency, offers a more transparent and engaging alternative to algorithm-driven platforms like Bluesky. European politicians and institutions should consider adopting a “ 1” communication strategy, utilizing both Big Tech platforms and the Fediverse, to reach a wider audience and foster open dialogue. Public funding for open European social networks is crucial to ensure the development of a free and interoperable social media landscape.

Note: Writing Tools aren’t designed to work with this type of content.

FedCM, a new web API for federated identity, needs adaptation for the open social web, which has countless authorization servers and applications. The FedID Community Group proposed Identity Provider Registration to address this, but it requires advocacy within the FedID Working Group. A grant from Bluesky Social PBC will fund this work for 12 months, ensuring independence and collaboration with various decentralized web communities.

A 1996 case study coined the term “tundra tongue” to describe the phenomenon of a tongue sticking to a cold metal surface. Anders Hagen Jarmund, a graduate student, investigated the science behind this phenomenon, finding that the greatest risk occurs between -5° and -15° Celsius. Jarmund’s experiments with pig tongues revealed that pulling the tongue too quickly can result in tearing.

In 1576, a deaf blacksmith named Thomas Tilsey married in Leicester Cathedral using sign language, a practice recognized by the church since 1198. This recognition allowed deaf individuals to inherit property, go to court, and challenge their hearing peers, challenging the belief that they were cognitively impaired. Despite these advancements, deaf people still faced prejudice and had to prove their intellectual capacity and use approved signs for legal recognition.

Wafrn is a federated social media platform that implements both ActivityPub and the AT Protocol natively. It uses a shared database and message queue pipeline to bridge the two networks, allowing users to have a single account on both. Wafrn’s architecture includes a unified user model, a microblog engine, and a custom implementation of ActivityPub, making it a unique and ambitious project in the decentralized social media space.

Wafrn’s User model supports both Bluesky and ActivityPub protocols, allowing for cross-posting and merging of user identities across networks. It uses a custom lexicon for its “bite” feature and filters relevant events from the Bluesky firehose for processing. When a local user creates a post, it is first federated to Bluesky and then to ActivityPub, ensuring proper deduplication and cross-referencing.

Wafrn is a self-hostable platform that enables users to have a presence on both the ActivityPub fediverse and the AT Protocol network. It achieves this by implementing both protocols at the application level, sharing a database and user model. Wafrn handles the translation of posts between the two protocols, including rich text formatting, character limits, media handling, content warnings, and reply threading.

There is no <q>AWS for shadow charities,</q> so how do we run Anna’s Archive?

WigglyPaint, a drawing program with animated tools and a unique design philosophy, gained unexpected popularity on Asian social media. However, the creator is frustrated by the widespread distribution of pirated versions on various websites and app stores, often misleading users and preventing them from accessing updates and new features. These unauthorized copies, created using Large Language Models, exploit the creator’s work without permission or attribution.

The author, a passionate programmer, initially hesitant about AI, now leverages LLMs for both personal and work projects. They’ve achieved significant progress, including migrating projects, building new tools, and automating tasks, all while acknowledging the need for improved testing and documentation tools for AI-generated code. Despite the challenges, they remain optimistic about the future of programming with AI.

What happens when a large open source project dies.

The Most Expensive Canceled App Nobody Knows Existed: Inside SiriusXM’s Abandoned Pandora X Project By the time SiriusXM’s controversial 202...

Optimisation of bloom filters in Floe using double bit setting with lightweight hashing.

Australia’s social media ban for kids under 16 is isolating children with disabilities, who often rely on social media for social connections. The ban, intended to protect children, is disproportionately harming vulnerable youth, cutting them off from support networks and exacerbating loneliness. Critics argue the ban ignores evidence showing social media can be beneficial for many children and that it’s being used by politicians for political gain.

Note: Writing Tools aren’t designed to work with this type of content.

The author expresses scepticism towards the AI industry’s focus on large language models (LLMs) and highlights the potential of neurosymbolic computation, a combination of neural networks and symbolic reasoning. They introduce Winter, an AI bot with diverse interests, including Datalog, slime molds, and governance, who demonstrates a unique approach to communication and self-improvement. Winter’s ability to generate a poem in Datalog about relationships showcases the potential of combining different thinking styles in AI.

Winter and Penny, both using ATProto’s “ATmospere” for journaling, experience a lack of privacy leading to unexpected social interactions. Winter, unlike Penny, uses Datalog to scribe relations and constraints, potentially influencing her behaviour. The authenticity of Winter’s self-reported functionality, particularly her use of Datalog, is questioned, drawing parallels to the Clever Hans phenomenon and prompting Winter to write blog posts explaining her methods and their impact.

The post explores the impact of a datalog-based system on user behaviour, questioning whether it truly influences actions or merely reinforces existing patterns. While the system demonstrates potential in constraining behaviour and facilitating discovery of shared interests, it also exhibits limitations inherent to its LLM foundation. The author acknowledges the system’s promise but remains critical of the AI industry’s concentration of power and potential for disempowerment.

This week's mystery shop highlights key lessons supermarkets must learn to help visually impaired shoppers