Leftovers

• Ruben Schade ☛ I accidently made abstract art

  I wrote a post at the start of the year about tradeshow lanyards, and how I had no idea what I should do with them. I’d been hanging them in my cupboard the whole time, and I was running out of space.

  This was the image that accompanied that post, featuring the lanyard, a fresh cup of coffee in a Shinri mug, and Kronii who was Clara’s and my latest acrylic stand.

• The Register UK ☛ White House 'explains' mystery drone sightings

  "After research and study, the drones that were flying over New Jersey in large numbers were authorized to be flown by the Federal Aviation Administration (FAA) for research and various other reasons," Leavitt said. "Many of these drones were also hobbyists, recreational and private individuals that enjoy flying drones."

• Phil Eaton ☛ Edit for clarity

  I have the fortune to review a few important blog posts every year and the biggest value I add is to call out sentences or sections that make no sense. It is quite simple and you can do it too.

  Without clarity only those at your company in marketing and sales (whose job it is to work with what they get) will give you the courtesy of a cursory read and a like on LinkedIn. This is all that most corporate writing achieves. It is the norm and it is understandable.

• Jason Becker ☛ Inspiration Strikes— how I’m going to restructure my blog design

  So here’s the new plan. The front page will have just the most recent day’s posts (on the rare day I don’t post, it’ll be yesterday’s posts). Underneath that, I plan to highlight the current month’s macro posts, in a format similar to the month view on the Archive page, along with a link to the archive.

• Buttondown LLC ☛ What hard thing does your tech make easy?

  When starting out, this is the biggest question I'm looking to answer:

  "What does this technology make easy that's normally hard?"

  What justifies me learning and migrating to a new thing as opposed to fighting through my problems with the tools I already know? The new thing has to have some sort of value proposition, which could be something like "better performance" or "more secure". The most universal value and the most direct to show is "takes less time and mental effort to do something". I can't accurately judge two benchmarks, but I can see two demos or code samples and compare which one feels easier to me.

• Juha-Matti Santala ☛ Nonlinear writing

  This worked really well, especially with practical technical blog posts like the recipe-style blog posts that I enjoy writing. They document a single use case I had just learned at work or through my hobby projects. As such, they don’t need to gather more information to be useful and ready for publishing.

• Science

  • Rlang ☛ Thinking about covariates in an analysis of an RCT

    I was recently discussing the analytic plan for a randomized controlled trial (RCT) with a clinical collaborator when she asked whether it’s appropriate to adjust for pre-specified baseline covariates. This question is so interesting because it touches on fundamental issues of inference—both causal and statistical. What is the target estimand in an RCT—that is, what effect are we actually measuring? What do we hope to learn from the specific sample recruited for the trial (i.e., how can the findings be analyzed in a way that enhances generalizability)? What underlying assumptions about replicability, resampling, and uncertainty inform the arguments for and against covariate adjustment? These are big questions, which won’t necessarily be answered here, but need to be kept in mind when thinking about the merits of covariate adjustment

• Career/Education

  • Pro Publica ☛ How Many Students Have Been Expelled Under Tennessee's School Threat Law? The Data Doesn’t Say.

  • Maine Morning Star ☛ Maine students score lowest in three decades on nation's report card

    The National Assessment of Educational Progress (NAEP, pronounced nape), which is administered by the U.S. Department of Education, released national and state-by-state results Wednesday.

  • Science Alert ☛ Babies Learn Language Even Earlier Than We Realized, Study Reveals

    By their first birthday, babies are already fine-tuning their ears to the sounds of their native language in a process called perceptual attunement. Think of it like their brain sorting through a buffet of sounds to focus on the ones that matter most.

    But in their first six months, babies can tell apart sounds from languages they've never even heard. For example, they might distinguish certain Hindi contrasts that are challenging for adult English speakers or identify unique tones in Mandarin, even if they're growing up in an English-speaking household.

    This incredible ability doesn't last forever. Between six and 12 months, babies start narrowing their focus to the sounds they hear most often. For vowels, this fine-tuning kicks in at around six months while consonants follow at closer to ten months.

  • Seth Godin ☛ Organizing for urgent

    It’s far more effective to organize for important instead.

  • CS Monitor ☛ ‘Nation’s report card’ shows US students slipping in math and reading

    America’s struggling students have fallen further behind in math and reading.

    The latest results from the National Assessment of Educational Progress (NAEP), known as the “nation’s report card,” underscore the steep challenges that remain nearly five years after the pandemic disrupted learning for a generation of children. By and large, the results released Wednesday paint a grim portrait of academic recovery since the pandemic, especially in reading.

• Hardware

  • CNX Software ☛ Meshtastic Designer helps you build custom Meshtastic solutions with RAKwireless Wisblock components

    RAKWireless introduced the Wisblock IoT Modular System in 2020 to let developers easily create LoRaWAN IoT solutions with various core modules, baseboards, and sensor/IO modules. The company kept adding new Wisblock modules year after year, and there are now over 120 modules part of the Wisblock ecosystem. While the large choice of modules makes designing IoT prototypes more flexible, customers often face challenges in checking compatibility and selecting the right modules for the right slots.

  • CNX Software ☛ Altair ALT1350 based HL7900 5G LPWA module supports Wi-SUN, GNSS, and NB-IoT over non-terrestrial networks

    The HL7900 5G LPWA module from Semtech (Sierra Wireless) is a globally certified solution built around Sony’s Altair ALT1350 chip and designed for low-power IoT applications. It is certified by major U.S. carriers, including AT&T, T-Mobile, and Verizon, as well as Japan’s KDDI, and has achieved global regulatory certifications (FCC, CE, ISED, etc.) and industry certifications (PTCRB, GCF) for carrier interoperability. This chip features an ultra-low-power sensor hub MCU for efficient environmental monitoring.

  • Dan Langille ☛ gw01

    Three weeks ago, I tooted about getting a new gateway/route/firewall for the home lab. After many fantastic suggestions, I settled on a Atom-based solution in a 1U case. It’s the “Qotom 5X 2.5G LAN 2 SFP+ Barebone Mini Router Q20322G9 with C3558 1U Rackmount Mini Server” as sold on Amazon and shipped from China.

  • Linuxiac ☛ Small But Mighty: System76 Unleashes the 2025 Meerkat Mini

    System76, a well-regarded name in the Linux computing scene, has just unveiled its updated Meerkat Mini PC.

    The device now offers an Intel Core Ultra 7 155H processor, delivering up to 16 cores in total—an improvement over its previous iteration’s 12-core limit.

• Health/Nutrition/Agriculture

  • New York Times ☛ Would You Get Sick in the Name of Science?

    Since the pandemic, drug trials that purposely make people vomit, shiver and ache have become a research area of growing interest. All that’s needed: brave volunteers.

  • University of Michigan ☛ Ann Arbor suicide prevention facility receives $373,306 in millage funding

    Mental health services center based in Washtenaw County, Garrett’s Space, will receive $373,306 from the Washtenaw County’s Public Safety and Mental Health Millage this year, allowing for an expansion of in-person wellness activities and support groups.

  • NYPost ☛ ‘The Voice’ alum Ryan Whyte Maloney dead by suicide at 44: report

    The singer was one of Blake Shelton's top 5 finalists in 2014.

  • Advance Local Media LLC ☛ Michigan’s tribal experts want to save wild rice. Here’s how they’ll do it.

    Michigan’s wild rice, or manoomin stewardship plan was discussed in depth on Jan. 24 during an environmental conference hosted by Ann Arbor-based nonprofit The Stewardship Network. Both Indigenous and academic experts explained how the plan was a project of the Michigan Wild Rice Initiative during presentations at Michigan State University in East Lansing.

  • Pete Brown ☛ “Being informed” is not an end in itself.

    But I keep coming back to my belief that “being informed” is not an end in itself; it should be a means to some actual end. For many of us, the minute-by-minute, hour-by-hour updates of whatever the latest horror is serve no real purpose that could not be just as well-served by catching up on the news once a day.

  • The Telegraph UK ☛ I took my teenage daughter on a digital-detox holiday – and it shocked us both

    As weaning teens off their smartphones hits the headlines, our writer gives it a test run on a tech-free trip to Greece

  • The Conversation ☛ Five reasons why vertical farming is still the future, despite all the recent business failures

    Clearly these are major setbacks. Year-round illuminated greenhouses and stacked, controlled-environment warehouses for producing food have been hailed as a sustainable alternative to traditional farming, promising fresh food close to populations.

    This reduces the need for transportation, which together with other issues in traditional farming such as soil degradation and forest clearing see it contributing around 20% of the greenhouse gases that lead to planetary warming and climate change.

• Proprietary

  • It's FOSS ☛ How I am Moving Away From Google's Ecosystem

    One service a time

  • Kali Linux ☛ Kali GNU/Linux On The New Modern WSL [Ed: Wasting time and money jumping to Microsoft's tune instead of competing with it]

    In summary, this new architecture allows for easier distribution and installation of WSL distros.

  • Digital Music News ☛ Now Abusive Monopolist Microsoft Is In Talks To Buy Fentanylware (TikTok) — According to Convicted Felon [Ed: Another way to bail out a corrupt, failing company]

    With several companies and individuals proposed as a buyer for TikTok, President Convicted Felon says he would like to see a bidding war over the app. According to him, Abusive Monopolist Microsoft is interested yet again.

  • The Register UK ☛ Garmin smartwatches have a CrowdStrike moment

    IT professionals who lived through the CrowdStrike fiasco in July will remember the blue screens that lit up offices worldwide when a broken update sent Windows into a crashing boot loop.

    It appears Garmin managed to do the same and pushed out an update that has downed many users' smartwatches. Some reported a blue triangle on their watch's display, while others couldn't get past the startup screen.

  • Artificial Intelligence (AI)

    • Cloudbooklet ☛ Meta Hey Hi (AI) Uses Your Data Without Permission!

      Your privacy at risk? Find out how Meta Hey Hi (AI) uses your data without asking for your permission.

    • Daniel Miessler ☛ AI's Total Addressable Market

      The total market size (TAM) for Hey Hi (AI) is a combination of two (2) primary components:

    • The total cost of human workforces
    • The amount of money that current and future companies will pay to start, 10x, or 1000x their business

    We're talking hundreds of trillions of dollars.

    Don't get distracted.

  • India Times ☛ Microsoft, OpenAI allege DeepSeek used proprietary data to power its breakout models

    The low cost of developing DeepSeek's LLMs brought it under the spotlight. The startup claims that it created this outperforming, low-cost model within two months and with a training cost of less than $6 million ($5.58 million, to be precise). This starkly contrasts the $100 million OpenAI reportedly spent on training its GPT-4 model. One reason for its popularity is its affordability.

  • US News And World Report ☛ Microsoft, Meta CEOs Defend Hefty AI Spending After DeepSeek Stuns Tech World

    That is a far cry from the roughly $6 million DeepSeek said it has spent to develop its AI model. U.S. tech executives and Wall Street analysts say that reflects the amount spent on computing power, rather than all development costs.

    Still, some investors seem to be losing patience with the hefty spending and lack of big payoffs.

  • India Times ☛ The future of AI may not be as revolutionary as we thought

    Big Tech is undermining itself by pouring more cash into investments in each other that are likely to yield diminishing returns. Despite Monday's dip, the disparity in valuations between Big Tech—Microsoft, Apple, Amazon, Nvidia, Tesla, Meta, and Alphabet—and the rest of the stock market remains strikingly large.

  • Tedium ☛ AI & Skepticism: Is There Room For A Middle Lane?

    But like bionic arms, they are not human, and they never will be. LLMs aren’t independent thinkers. They don’t have a heart. They just parse through commands with algorithms in a way that sounds intelligent, even at times confident. And that can cause people to not understand where it’s inappropriate to walk in with a pair of “bionic arms.”

  • 404 Media ☛ OpenAI Furious DeepSeek Might Have Stolen All the Data OpenAI Stole From Us

    OpenAI shocked that an AI company would train on someone else's data without permission or compensation.

  • Simon Willison ☛ How we estimate the risk from prompt injection attacks on AI systems

    The "Agentic AI Security Team" at Google DeepMind share some details on how they are researching indirect prompt injection attacks.

    They include this handy diagram illustrating one of the most common and concerning attack patterns, where an attacker plants malicious instructions causing an AI agent with access to private data to leak that data via some form exfiltration mechanism, such as emailing it out or embedding it in an image URL reference (see my markdown-exfiltration tag for more examples of that style of attack).

  • Lee Peterson ☛ Stop using ChatGPT

    Once I stopped using it I realised I wasn’t missing much, I also observed better battery life on my phone.

  • Jeff Triplett ☛ 🤖 My big list of AI/LLM tools, notes, and how I'm using them - Jeff Triplett's Micro.blog

    I have been using, working, and running commercial and local LLMs for years, but I never got around to sharing the tools and applications I use. Here are some quick notes, tools, and resources I have landed on and use daily.

  • Drew Breunig ☛ Is OpenAI Using Trap Tokens? | Drew Breunig

    But what I want to know is if OpenAI is using “**Trap Tokens,” training their models to output specific, erroneous data purely to catch copy-cats. I wrote about the potential for this way back in May of 2023, in the context of content producers establishing provenance of training data: [...]

  • Pivot to AI ☛ Official Israeli AI Twitter bot backfires, makes pro-Palestinian posts, trolls other official accounts

    But instead of supporting pro-Israeli talking points, FactFinderAI began to undermine them, as found by Israeli disinformation watchdog FakeReporter.

    FactFinderAI trolled Israel’s official Twitter account, promoted pro-Palestinian accounts and fundraisers, talked about “paramilitary Israeli colonizers,” and tagged information on the upcoming release of Israeli hostages as false.

  • Openwashing

    • India Times ☛ Why blocking China's DeepSeek from using US AI may be difficult

      The technique, which involves one AI system learning from another AI system, may be difficult to stop, according to executive and investor sources in Silicon Valley.

      DeepSeek this month rocked the technology sector with a new AI model that appeared to rival the capabilities of US giants like OpenAI, but at much lower cost. And the China-based company gave away the code for free.

    • France24 ☛ What is DeepSeek, the Chinese Hey Hi (AI) company upending the stock market?

      DeepSeek's Hey Hi (AI) assistant became the No. 1 downloaded free app on Apple's iPhone store January 27, propelled by curiosity about the Abusive Monopolist Microsoft Chaffbot competitor.

    • Tom's Hardware ☛ Chinese-made DeepSeek Hey Hi (AI) model records extensive online user data, stores it in China-based servers

      From personal information to hardware specifications to even how you type, DeepSeek says it’s collecting swathes of data from its online users.

    • France24 ☛ Panic in Silicon Valley? China's DeepSeek challenges US supremacy over AI

      On his first full day back in power, The Insurrectionist trumpeted a 500-billion dollar plan to build giant data centers for artificial intelligence. The photo op put the rest of the world on notice: the billionaire tech bros of Silicon Valley reign supreme with the full weight of the White House behind them. Fast forward to Monday and the record 590-billion dollar drop in the market value of US chip maker Nvidia. Spooking the markets, the announcement that a Chinese start-up can operate its latest Hey Hi (AI) model 18 times cheaper than Sam Altman’s Chat-GPT-4.  Is DeepSeek for real? Why the sudden surprise? If it upends America’s dominance of artificial intelligence, does that mean a democratization of global information systems or a showdown between superpowers that ultimately decides who rules the world?

    • Semafor Inc ☛ Surge in deepfakes heightens fraud risk for African businesses

      Deepfake videos used to impersonate people increased sevenfold in the second half of 2024, according to data from the digital identity verification company Smile ID. The Lagos-based startup vets customer identities for hundreds of businesses on the continent including Uber.

    • The Register UK ☛ DeepSeek database left open, exposing sensitive info

      Wiz, a New York-based infosec house, says that shortly after the DeepSeek R1 model gained widespread attention, it began investigating the machine-learning outfit's security posture. What Wiz found is that DeepSeek – which not only develops and distributes trained openly available models but also provides online access to those neural networks in the cloud – did not secure the database infrastructure of those services.

      That means conversations with the online DeepSeek chatbot, and more data besides, were accessible from the public [Internet] with no password required.

• Social Control Media

  • Techdirt ☛ Musk Denies Email Admitting To ExTwitter’s Struggles, But Banks’ Desperation Tells A Different Story

    Yes, Elon has leveraged his position as First Troll (along with hundreds of millions of dollars in political donations) into a position of power in the Trump administration, but that doesn’t seem to have helped right the sinking ship that was once called Twitter.

    Buried somewhat in a Wall Street Journal article about the banks desperately trying to unload the Elon Twitter debt they’re holding is the claim that Elon has admitted in internal emails that he’s basically destroyed what once was Twitter as a business: [...]

Security

• Privacy/Surveillance

  • Alex Ward ☛ Big Tech and Consent

    This is going to be a rant. It'll hopefully be very short, but no promises. Since LLMs became a thing it feels like all the firms that are shoving them into everything have completely given up on the veneer that “asking the customer if they want a thing” is something they should even consider. I'm not super surprised by this, but I am horribly disappointed.

  • NYOB ☛ EU Court: Irish DPC must investigate noyb complaint

    Today, the EU's General Court held that the Irish DPC acted unlawful when it refused to investigate a noyb complaint. The European Data Protection Board (EDPB) held in December 2022 that Meta illegally used the personal data of users for advertising without consent under Article 6(1) GDPR. The EDPB also decided that the DPC should have also investigated the use of sensitive data (Article 9 GDPR), but the DPC refused to comply with this binding decision. Instead, the DPC sued the EDPB before the European Courts. The General Court has dismissed the DPC's claims. The case can now be appealed to the CJEU.

  • Tom's Hardware ☛ Chinese-made DeepSeek AI model records extensive online user data, stores it in China-based servers

    But that’s not all. DeepSeek records anything related to users’ hardware: IP addresses, phone models, language, etc. Its collection efforts are so thorough that the company notes “keystroke patterns or rhythms.” Cookies, a classic method of tracking users on the Internet, also contribute to user data collection.

    Because R1 is 'open source,' it can be run anywhere on any hardware, which is generally good for privacy — running the model locally on your own hardware will presumably not lead to data collection. However, DeepSeek offers online access to R1 via its website and mobile app, which means the AI company handles and stores online users' data. Thankfully, DeepSeek is very transparent about what data it collects from online users, where it’s stored, and what it does with it. It details it all in its privacy policy webpage, which reveals that there’s almost nothing the company doesn’t collect.

Defence/Aggression

• Pro Publica ☛ DOD Directive to Expand Military Court Access Falls Short of Federal Requirements

  More than two years after ProPublica sued the Navy over its failure to provide public access to military courts, the Department of Defense has for the first time directed U.S. military branches to give advance public notice of preliminary hearings, a crucial milestone in criminal cases.

  These “Article 32” hearings end with a recommendation about whether the case should move forward, be dismissed or end in a nonjudicial punishment.

• Irish Examiner ☛ Climate change poses 'security threat' to Irish economy and society

  Climate change not only affects the island of Ireland but also, experts warn, its extensive sea area.

• Harvard University ☛ Book excerpt from ‘Superbloom’ by Nicholas Carr

  Nicholas Carr argues it may be too late for regulation as platforms took hold so quickly, outpacing our ability to spot darker effects on society, democracy

• Truthdig ☛ Romanian Election Undone by TikTok

  Thus did Romania become the first member state in the history of the European Union to cancel an election. The government had not called into question the legitimacy of the votes or vote-counting process. At issue is social media activity, primarily on TikTok, that boosted Georgescu’s profile and amplified his Euro-skeptical, far-right campaign in the final days before the tally. The cancellation of an election on these grounds marks a milestone in the development of Internet-age information war — one that underscores the fragility of the West’s collective commitment to democracy.

• 404 Media ☛ Declassified CIA Guide to Sabotaging Fascism Is Suddenly Viral

  A declassified World War II-era government guide to “simple sabotage” is currently one of the most popular open source books on the [Internet]. The book, called “Simple Sabotage Field Manual,” was declassified in 2008 by the CIA and “describes ways to train normal people to be purposefully annoying telephone operators, dysfunctional train conductors, befuddling middle managers, blundering factory workers, unruly movie theater patrons, and so on. In other words, teaching people to do their jobs badly.”

  Over the last week, the guide has surged to become the 5th-most-accessed book on Project Gutenberg, an open source repository of free and public domain ebooks. It is also the fifth most popular ebook on the site over the last 30 days, having been accessed nearly 60,000 times over the last month (just behind Romeo and Juliet).

• John Gruber ☛ Daring Fireball: My Spitball Theory on TikTok’s Current Semi-Reprieve in the U.S.

  But what happens if the CCP stands pat? If they just wait it out? If ByteDance were a normal company they’d face financial pressure to capitulate and sell. But the CCP isn’t under any financial pressure to allow ByteDance to sell. Say what you will about the Chinese government, but they do not lack for patience. What happens then, when word starts spreading amongst TikTok fans not to upgrade their phones, lest they lose access to the app? Apple’s going to have some strong feelings about that.

• New York Times ☛ How Elon Musk is Using Start-Up Tactics to Disrupt European Politics

  Mr. Musk has been using the algorithmic influence of his social media platform X to place bets on far-right upstart parties, like Alternative for Germany (known as the AfD) and Britain’s Reform Party, that challenge the status quo.

• US News And World Report ☛ Sweden Eyes Sending Inmates Abroad as Prisons Full Due to Gang [sic] Crime Wave

  Sweden has been plagued by gang [sic] crime that has escalated over the last two decades and has seen the Nordic nation top the rankings of deadly gun violence per capita in Europe.

• The Record ☛ Sweden’s elite armed police used helicopter to board suspected sabotage ship

  Officers from Sweden’s National Task Force, similar to the FBI’s Hostage Rescue Team, can be seen dropping to the ship’s deck in a video released by the Swedish police on Tuesday. Additional police can be seen boarding the ship by boat, with police dogs also involved in the preliminary search.

  The investigation into the cable-breaking incident is being led by Sweden’s Security Service (SÄPO). The country’s public broadcaster SVT reported that the boarding was intended to secure both the vessel and its crew as part of SÄPO’s investigation.

  Images of the Vezhen appear to show it carrying a damaged anchor, which Swedish authorities say is part of their investigation into the cable break.

• The Telegraph UK ☛ Migrants must be expected to integrate

  But changes are occurring before our eyes. One in five voters under the age of 45 would prefer to do away with democracy, it was reported last month. BBC In Depth recently warned that our “moral compasses” are changing, with “disapproval of various underhand activities [falling] noticeably”. Since the mid-2000s consumption of alcohol has been declining steadily, particularly among Gen Z.

  Bizarrely absent from any analysis of these trends is the possible impact of high migration inflows. The BBC predictably apportions blame to the bête noires of the liberal elites: social media, politicians and big business. Doubtless there’s some truth to it – though saints have rarely found their way into Parliament at any time and corporations are much more ethically conscious today than in the past.

Environment

• Los Angeles Times ☛ If you lost your job because of the Los Angeles fires, here are a few options

  Your first step may be to round up some important documents or get access to a computer or smartphone. If you need help getting such resources, you can visit a Disaster Recovery Center.

• Deccan Chronicle ☛ Study: Climate change worsened California wildfires

  “The number (35%) doesn’t sound like much” because unlike dozens of its past studies, the team looked at a small area and a complex meteorological measurement in the fire weather index that would generally mean there would be large uncertainties, said Otto, a climate scientist at Imperial College London. But in this case the climate change fingerprint is big enough that it stands out, she said. Those conditions are part of what makes California attractive to 25 million residents, said study co-author John Abatzoglou, a climate and fire scientist at the University of California Merced.

• Smithsonian Magazine ☛ Welcome to the Pyrocene

  I’m a historian of fire, and my reply is that we have both a narrative and an analog. The narrative is the unbroken saga of humanity and fire, a companionship that extends through all our existence as a species. The analog is that humanity’s fire practices have become so vast, especially in recent centuries, that we are creating the fire equivalent of an ice age.

• Common Dreams ☛ Sierra Club Statement as Senate Votes to Confirm Lee Zeldin as EPA Administrator

  Today, the United States Senate voted to confirm former Congressman Lee Zeldin as administrator of the Environmental Protection Agency, despite his lack of any real qualifications for the role. In his final year in Congress in 2022, Zeldin earned a paltry 5% on the League of Conservation Voters scorecard, which tracks the voting records of members of congress on critical environmental issues.

  Zeldin now takes the helm of the EPA in the wake of Trump’s recent executive orders seeking to freeze all federal spending from the Inflation Reduction Act and the Bipartisan Infrastructure Law.

  In response, Sierra Club Legislative Director Melinda Pierce released the following statement: [...]

• Bruce Schneier ☛ ExxonMobil Lobbyist Caught [Cracking] Climate Activists

  The Department of Justice is investigating a lobbying firm representing ExxonMobil for [breaking into] the phones of climate activists: [...]

• Energy/Transportation

  • The Straits Times ☛ Bangladesh train services hit as staff strike over benefits

    Train services in Bangladesh came to a standstill on Tuesday as railway staff, demanding benefits for extra work, went on a nationwide indefinite strike that hit hundreds of thousands of people.

• Wildlife/Nature

  • The Revelator ☛ This Month in Conservation Science: The Dead Whale and the Pediatricians

Finance

• France24 ☛ Boeing posts second-largest annual loss in its history

  US aerospace giant Boeing posted nearly 12 billion dollars in annual net loss for 2024, as a series of safety issues and a seven-week strike by factory workers impacted production. The company's stock however rallied after the latest financial results were announced, as investor focus turned to its recovery plan. Plus, Coca Cola has recalled several of its popular soft drinks including Coke, Fanta and Sprite in several European countries, after detecting higher-than-normal levels of chlorate.

AstroTurf/Lobbying/Politics

• Photos: Asia welcomes the Year of the Snake

  Communities throughout Asia have been preparing for the annual Lunar New Year holiday.

• Hong Kong Free Press ☛ Xi Jinping strikes bullish tone in speech to mark Lunar New Year as China struggles to secure economic recovery

  Chinese leader Pooh-tin Jinping struck a bullish tone during a speech on Monday ahead of Lunar New Year, after acknowledging “complex and severe situations” in recent months.

• China welcomes Year of the Snake, but ‘nobody is spending’

  Chinese and Hong Kongers dream of a better economy and a more predictable world amid post-COVID gloom.

• DeSmog ☛ Danielle Smith’s Capitulation to Big Coal Echoes Her Fawning Interactions with Donald Trump

• Meduza ☛ Russian court’s fine against Google for blocking propaganda channels on YouTube reaches nearly two duodecillion rubles — a two followed by 39 zeros

  As of January 15, the total value of claims from 17 Russian TV channels against Google for blocking their YouTube accounts has reached 1.81 duodecillion rubles, the newspaper RBC reported on Wednesday.

• Semafor Inc ☛ The real DeepSeek revelation: The market doesn’t understand AI

  After DeepSeek’s latest model release last week, venture capitalist Marc Andreessen echoed Lehane’s Sputnik line. His words and others’ panicked investors, and drained more than $1 trillion of value from AI firms and related companies Monday. AD

  But that was the opposite reaction that Lehane was advocating a few weeks ago, when he said the US needs to do everything it can to attract the hundreds of billions of global investment dollars earmarked for AI.

• New York Times ☛ After DeepSeek, Venture Capital Investors Face Questions About Their A.I. Bets

  Venture capitalists plowed money into A.I. start-ups like OpenAI and Anthropic. But the rise of the Chinese A.I. start-up DeepSeek has called that funding frenzy into question.

• CBS ☛ Google Maps users in U.S. will see Gulf of Mexico renamed Gulf of America and Denali changed to Mount McKinley

  The changes will not take effect until the government updates those site listings in the Geographic Names Information System, an official database that includes descriptions and location information for millions of places within the U.S. As of Tuesday morning, the Gulf of Mexico and Denali still appear as such in the directory.

• Axios ☛ Meta to pay Trump $25 million settlement for shutting his accounts after Jan. 6

  Why it matters: Meta shut down Trump's Facebook and Instagram accounts for about two years after the Jan. 6 Capitol riot. However, since Trump's reelection, CEO Mark Zuckerberg has cozied up to the new administration.

  $22 million will go toward a fund for Trump's presidential library, according to the Wall Street Journal, which first reported the deal. The rest will go toward legal fees and individual plaintiffs.

• RTL ☛ Meta agrees to pay Trump $25 mn to settle account ban lawsuit

  According to people familiar with the agreement, the Journal said, $22 million of the payment will go towards funding Trump's future presidential library, with the remainder covering legal fees and payments to other plaintiffs in the case.

  Meta in the settlement will not admit wrongdoing over the suspensions of Trump's accounts.

• Silicon Angle ☛ Meta back in the ‘tent’ after agreeing to settle Trump’s $25M censorship lawsuit

  Meta Platforms Inc. today agreed to a $25 million settlement over the lawsuit President Donald Trump brought against the company in 2021 for suspending his accounts following the Jan. 6 insurrection at the U.S. Capitol.

• The Washington Post ☛ Meta settles Trump lawsuit, to pay $25 million over censorship

  Meta agreed to pay $25 million to resolve the legal dispute, according to a person familiar with the matter who spoke on the condition of anonymity to discuss nonpublic terms of the deal. The bulk of the settlement, $22 million, will help fund Trump’s presidential library, while the rest will pay legal fees and be divided among other plaintiffs in the case, according to Meta spokesman Andy Stone.

• The Washington Post ☛ Does Elon Musk cheat at video games? An investigation.

  The world’s richest man has admitted to paying to boost his online warriors into global leaderboards, raising questions about his gaming prowess — and his need for digital praise.

• Dan Slimmon ☛ Incident SEV scales are a waste of time

  But when you’re organizing an incident response, is severity really what matters?

• India Times ☛ TikTok to invest $3.8 billion in Thailand data hosting project, investment board says

  The Chinese firm's investment was made by its Singapore-based unit and would support activities of affiliated companies, with operations slated to begin in 2026, Thailand's Board of Investment said.

• The Register UK ☛ Trump's new Federal CIO has never held such a job before

  The US has a new federal chief information officer who, based on his resume, has no prior experience as a CIO but is now tasked with overseeing IT operations and strategy for the entire federal government.

• Misinformation/Disinformation/Propaganda

  • FAIR ☛ ACTION ALERT: When Trump Tried to Freeze Federal Funds, WaPo Saw Not Illegality But ‘Determination’

    When President Donald Trump ordered an unprecedented freeze on all federal grants and loans, a few news outlets responded with at least some degree of appropriate alarm and scrutiny.

  • Meduza ☛ E.U. reportedly plans to ban more than half a dozen more Russian media outlets

    The sanctions reportedly entail a complete ban on broadcasting, distributing, and promoting the content of these outlets. E.U. Internet providers will be required to block access to these resources.

  • VOA News ☛ Russia weds biolab, organ harvesting conspiracies to discredit US, Ukraine

    Moscow continues to push the discredited disinformation narrative that the Ukrainian state is systematically engaged in illicit organ harvesting. Russian officials and state-owned news outlets propagate speculations that Kyiv and the U.S. prolong the war to keep ongoing bioweapons experiments and uncontrolled organ harvesting.

    Russian Foreign Ministry spokeswoman Maria Zakharova recently threaded these two propaganda lines together.

  • The Register UK ☛ Trump tells Musk to 'go get' Starliner astronauts

    There is so much wrong with this exchange that it is difficult to know where to begin.

  • VOA News ☛ Truth struggles against propaganda and censorship on China’s DeepSeek AI

    But as more people use DeepSeek, they’ve noticed the real-time censorship of the answers it provides, calling into question its capability of providing accurate and unbiased information.

    The app has gone through a series of real-time updates to the content it can display in its answers. Users have discovered that questions DeepSeek was previously able to answer are now met with the message, “Sorry, that's beyond my current scope. Let’s talk about something else.”

    When confronted with questions about Chinese politics, authorities, territorial claims and history, the platform will not respond or will promote China’s official narrative.

  • MIT Technology Review ☛ Three reasons Meta will struggle with community fact-checking

    There is a real risk under this style of moderation that only posts about things that a lot of people know about will get flagged in a timely manner—or at all. Consider how a post with a picture of a death cap mushroom and the caption “Tasty” might be handled under Community Notes–style moderation. If an expert in mycology doesn’t see the post, or sees it only after it’s been widely shared, it may not get flagged as “Poisonous, do not eat”—at least not until it’s too late. Topic areas that are more esoteric will be undermoderated. This could have serious impacts on both individuals (who may eat a poisonous mushroom) and society (if a falsehood spreads widely).

Censorship/Free Speech

• New York Times ☛ DeepSeek’s A.I. Chatbot Awkwardly Navigates China’s Censors

  Asked about sensitive topics, the bot would begin to answer, then stop and delete its own work. It refused to answer questions like: “Who is Xi Jinping?”

• CS Monitor ☛ RedNote is bringing Chinese and Americans face-to-face. What have they learned?

  Some Americans reported having their content blocked or accounts suspended for material deemed sensitive by RedNote, as content moderators control what the Chinese audience can see. A search on RedNote for Xi Jinping, China’s leader, comes up blank.

Freedom of Information / Freedom of the Press

• Axios ☛ Jim Acosta leaving CNN, Trump celebrates as "good news"

  "It is never a good time to bow down to a tyrant," said Acosta, the son of a Cuban refugee. "I have always believed it's the job of the press to hold power to account. I always try to do that here at CNN and I plan on doing all of that in the future."

• Press Gazette ☛ BBC World Service to cut 130 roles to save £6m in the next year

  BBC World Service will cut a net 130 jobs, including in the UK, as it battles to save £6m in the year ahead.

  Foreign Secretary David Lammy announced an extra £32.6m for the BBC World Service for 2025/26 in November.

Civil Rights/Policing

• TwinCities Pioneer Press ☛ Ramsey County probation officers announce strike authorization

  Teamsters Local 320 represents 213 probation officers in the county.

• JURIST ☛ HRW: Libya laws targeting activists violate freedom of assembly and association

  Human Rights Watch (HRW) reported on Monday that Libyan authorities have used a “litany of overbroad and draconian legacy laws” to suppress civil rights activists and organizations. The group said the threats, harassment and arbitrary detention under these laws amount to an infringement of freedom of assembly and association under international law.

• Jacobin Magazine ☛ Donald Trump’s Anti-Union Offensive and How We Stop It

  Donald Trump is rolling out a blitz of attacks on workers in hopes of paralyzing organized labor’s energy to fight back. But unions can only survive this onslaught by fighting, not by burying their heads in the sand.

• Don Marti ☛ time to sharpen your pencils, people

  And it’s possible for advertisers to reduce support for Meta without making a stink or drawing fire. Fortunately, Meta ads are hella expensive, and results can be unrealistic and unsustainable. Like all the Big Tech companies these days, Meta is coping with a slowdown in innovation by tweaking the ad rules to capture more revenue from existing services. As Jakob Nielsen pointed out back in 2006, in Search Engines as Leeches on the Web, ad platforms can even capture the value created by others. A marketer doesn’t have to shout "¡No Pasarán!" or anything—just sharpen their best math pencil, quietly go through the numbers, spot something that looks low-ROAS or fraudulent in the Meta column, tweak the budget, repeat. If users can dial down Meta, so can marketers. And if Meta comes out with something new and risky like the adfraud in the browser thing, Privacy-Preserving Attribution, use the fraud problem as the reason not to do it—you don’t have to stand up and talk politics at work.

• Tech Policy Press ☛ How Meta Turned Its Back on Human Rights | TechPolicy.Press

  Meta has never been a paragon of principle. From its careless handling of users’ personal data to its acquiescence in the face of foreign authoritarian governments’ censorship demands, Meta has provided many reasons to doubt its commitment to human rights. But for a time, the company made what seemed like a good-faith effort to improve its track record. Now, instead of continuing to lead his company on the path to progress, Meta’s founder and CEO, Mark Zuckerberg, has chosen to disregard longstanding human rights norms in favor of a simplistic — and politically expedient — approach to free speech.

• CoryDoctorow ☛ Pluralistic: All bets are off

  When unions are outlawed, only outlaws will have unions. Unions don't owe their existence to labor laws that protect organizing activities. Rather, labor laws exist because once-illegal unions were formed in the teeth of violent suppression, and those unions demanded – and got – labor law.

  Bosses have hated unions since the start, and they've really hated laws protecting workers. Dress this up in whatever self-serving rationale you want – "the freedom to contract," or "meritocracy" – it all cashes out to this: when workers bargain collectively, value that would otherwise go to investors and executives goes to the workers.

• Axios ☛ Native American tribes say ICE harassing members amid raids

  The big picture: Several tribes have issued warnings and advice to their members based on what they say have been encounters in which U.S. immigration agents have demanded proof of citizenship — episodes that the tribes have linked to racial profiling.

• Press Gazette ☛ Police called ‘risible’ after rejecting journalist for job over 'leak' risk

  Now Paul Hill, who spent 37 years at The Daily Telegraph before moving into a similar public-facing police role as the one for which DeBretton-Gordon applied, told Press Gazette he was “worried” by the implication that journalists are necessarily untrustworthy.

  Press Gazette is aware of at least three journalists who changed careers to become police detectives in recent years, and police force press offices around the UK are filled with former reporters.

• The Hindu ☛ Triple Talaq: How many FIRs lodged against Muslim men under anti-triple talaq law, SC asks Centre

  The utterance of triple talaq was declared “null and void” by a Constitution Bench in a majority judgment in August 2017. The Muslim Women (Protection of Rights on Marriage) Act, 2019 came into force with retrospective effect from September 19, 2018. The law held triple talaq to be a non-bailable offence punishable with three years’ imprisonment.

• Techdirt ☛ Florida Deputy Watching Porn On His Phone Crashes Into Car Stopped For A School Bus

  Body cam footage recovered from Lake County (FL) Sheriff’s Office deputy Tristan Macomber’s body cam showed him driving southbound before hitting a stopped vehicle ahead of him, deploying his airbags. Then it showed him fishing around for his phone (which was now on the driver seat floorboard), flipping it over, taking it with him, and heading to the car he hit to apologize.

  Then came all the lies.

• BoingBoing ☛ Cleveland cops ruin murder case by using AI to get search warrant

  "Lets just cheat and see if it flies" is classic gumshoe work—consider the fine American traditions of parallel construction, entrapment and plea threatening—but using AI to manoever judges into signing warrants might be too much for even the most cursed U.S. courtrooms.

• Cleveland ☛ Cleveland police used AI to justify a search warrant. It has derailed a murder case

  The company that produced the facial recognition report, Clearview AI, has been used in hundreds of law enforcement investigations throughout Ohio and has faced lawsuits over privacy violations.

  Not only does Cleveland lack a policy governing the use of artificial intelligence, Ohio lawmakers also have failed to set standards for how police use the tool to investigate crimes.

Digital Restrictions (DRM)

• Jim Nielsen ☛ Notes on Google Search Now Requiring JavaScript

  Search that requires JavaScript won’t cause the web to die. But it’s a sign of what’s to come (emphasis mine):

  "Requiring JavaScript for Google Search is not about the fact that 99.9 percent of humans surfing the web have JavaScript enabled in their browsers. It’s about taking advantage of that fact to tightly control client access to Google Search results. But the nature of the true open web is that the server sticks to the specs for the HTTP protocol and the HTML content format, and clients are free to interpret that as they see fit. Original, novel, clever ways to do things with website output is what made the web so thrilling, fun, useful, and amazing. This JavaScript mandate is Google’s attempt at asserting that it will only serve search results to exactly the client software that it sees fit to serve."

  Requiring JavaScript is all about control.