Last year California’s Digital Age Assurance Act (AB 1043) was signed into law, requiring among other things that operating system providers implement an API for age verification purposes. With the implementation date of January 1, 2027 slowly encroaching this now has people understandably agitated. So what are the requirements, and what will its impact be, as it affects not only OS developers but also application stores and developers?
The required features for OS developers include an interface at account setup during which the person indicates which of the four age brackets they fit into. This age category then has to be used by application developers and application stores to filter access to the software. Penalties for non-compliance go up to $2,500 per affected child if the cause is neglect and up to $7,500 if the violation was intentional.
As noted in the Tom’s Hardware article, CA governor Newsom issued a statement when signing the unanimously passed bill, saying that he hopes the bill gets amended due to how problematic it would be to implement and unintended effects. Of course, the bigger question is whether this change requires more than adding a few input fields and checkboxes to an OS’ account setup and an API call or two.
When we look at the full text of this very short bill, the major questions are whether this bill has any teeth at all. From reading the bill’s text, we can see that the person creating the account is merely asked to provide their birth date, age or both. This makes it at first glance as effective as those ‘pick your age’ selection boxes before entering an age-gated part of a website. What would make this new ‘age-verification feature’ any more reliable than that?
Advertisement
Although the OS developer is required to provide this input option and an API feature of undefined nature that provides the age bracket in some format via some method, the onus is seemingly never put on the user who creates or uses the OS account. Enforcement as defined in section 1798.503 is defined as a vague ‘[a] person that violates this title’, who shall have a civil action lawsuit filed against them. What happens if a 9-year old child indicates that they’re actually 35, for example? Or when a user account is shared on a family computer?
All taken together, this bill looks from all angles to add a lot of nuisance and potential for catching civil lawsuit flak for in particular FOSS developers, all in order to circuitously reimplement the much beloved age dropdown selection widget that’s been around since at least the 1990s.
They could give this bill real teeth by requiring that photo ID is required for registering an (online-only) OS account, much like with the recent social media restrictions and Discord age-verification kerfuffle, but that’d run right over the ‘privacy-preserving’ elements in this same bill.
In the 1970s, the USSR had an undersea cable connecting a major naval base at Petropavlovsk to the Pacific Fleet headquarters at Vladivostok. The cable traversed the Sea of Okhotsk, which, at the time, the USSR claimed. It was off limits to foreign vessels, heavily patrolled, and laced with detection devices. How much more secure could it be? Against the US Navy, apparently not very secure at all. For about a decade starting in 1972, the Navy delivered tapes of all the traffic on the cable to the NSA.
Top Secret
You need a few things to make this a success. First, you need a stealthy submarine. The Navy had the USS Halibut, which has a strange history. You also need some sort of undetectable listening device that can operate on the ocean floor. You also need a crew that is sworn to secrecy.
That last part was hard to manage. It takes a lot of people to mount a secret operation to the other side of the globe, so they came up with a cover story: officially, the Halibut was in Okhotsk to recover parts of a Soviet weapon for analysis. Only a few people knew the real mission. The whole operation was known as Operation Ivy Bells.
The Halibut
The Halibut is possibly the strangest submarine ever. It started life destined to be a diesel sub. However, before it launched in 1959, it had been converted to nuclear power. In fact, the sub was the first designed to launch guided missiles and was the first sub to successfully launch a guided missile, although it had to surface to launch.
Advertisement
Oddly enough, the sub carried nuclear cruise missiles and its specific target, should the world go to a nuclear war, was the Soviet naval base at Petropavolvsk.
By 1965, the sub had been replaced for missile duty by newer submarines. It was tapped to be converted for “special operations.” Under the guise of being a deep-sea recovery vehicle, the Halibut received skids to settle on the seabed, side thrusters, specialized anchors, and a host of electronic equipment, including “the Fish” a 12-foot-long array of cameras, sonar, and strobe lights weighing nearly two tons. The “rescue vehicle” on its stern didn’t actually detach. It was a compartment for deploying saturation divers.
An early mission was Operation Sand Dollar. Halibut found the wreck of the Soviet K-129, which the US would go on to recover in another top secret mission, looking for secrets and Soviet technology.
When it came time to deploy the listening device on an underwater cable, Halibut was perfect. It could park a safe distance away, deploy saturation divers, and recover them. If you want to see more about the Halibut, check out the [Defence Central] video below.
Advertisement
The Listening Device
A later undersea wire tap device (Soviet photograph)
This wasn’t a hidden microphone in a briefcase. It was a 20-foot, six-ton pressure vessel parked on the ocean floor. Details are murky, but there was another part, probably smaller, that clamped around the cable. Working inductively, it didn’t pierce the cable for fear the Soviets would notice that. In addition, if they raised the cable for maintenance, the device was made to break away and sink to the bottom.
Needless to say, tapping a cable on the ocean floor isn’t easy. First, they had to locate the cable. Luckily, there were signs at either end telling fishing vessels to avoid the area. That helped, but they still had to search for the 5-inch wide cables. They found them at least 400 feet below the surface, some 120 miles offshore.
Saturation diving was a relatively new idea at the time, and the Navy’s SeaLab experiments had given them several years of experience with the technology. While commercial saturation dives started in 1965, it was still exotic technology in 1971. The first mission simply recorded a bit of data on the submarine and returned it. Once it was proven, the sub returned with the giant tap device and installed it.
It took four divers to position the big tap. Even then, you couldn’t just leave it there. The device used tapes and required service once a month. So Halibut or another sub had to visit each month to swap tapes out. We couldn’t find out what the power source for the bug was, so they probably had to change the batteries, too.
Advertisement
The Soviets didn’t consider the cable to be at risk for eavesdropping, so much of the traffic on the cable was in the clear. It was a gold mine of intelligence information, and many credit the information gained as crucial to closing the SALT II treaty talks.
Secondary Mission
Most of the crews participating in Operation Ivy Bells didn’t have clearance to know what was going on. Instead, they thought they were on a different secret mission to retrieve debris from Soviet anti-ship missiles.
To keep the story believable, the crew actually did recover a large number of parts from the subject Soviet missiles. Turns out, analysis of the debris did reveal some useful information, so two spy missions for the price of one.
Presumably, the assumption would be that if the Soviets heard a sub was scavenging missile parts, it might qualify as a secret, but it would hardly be a surprise. They couldn’t have imagined the real purpose of the submarine.
Advertisement
Future Taps
Later undersea taps were created that used radioisotope batteries and could store a year’s data between visits that tapped other Soviet phone lines. Submarines Parche, Richard B. Russel, and Seawolf saw duty with some of these other taps as well as taking over for Halibut when it retired four years after the start of Operation Ivy Bells.
The original Okhotsk tap would have operated for many more years if it were not for [Ronald Pelton]. A former NSA employee, he found himself bankrupt over $65,000 of debt. In 1980, he showed up at the Soviet embassy in Washington and offered to sell what he knew.
He knew a number of things, including what was going on with Operation Ivy Bells. That data netted him $5,000 and, overall, he got about $35,000 or so. Oh, he also got life in prison when, in 1985, a Soviet defector revealed he had been the initial contact for [Pelton].
The Soviets didn’t immediately act on [Pelton’s] intel, but by 1981, the Americans knew something was up. A small fleet of ships was parked right over the device. The USS Parche was sent to retrieve it, but they couldn’t find it. Today, it (or, perhaps, a replica) is in the Great Patriotic War Museum in Moscow.
Curious about split keyboards, but overwhelmed by the myriad options for every little thing? You should start with [thehaikuza]’s excellent Beginner’s Guide to Split Keyboards.
Your education begins with the why, so you can skip that if you must, but the visuals are a nice refresher on that front.
He then gets into the types of keyboards — you got your standard row-staggered rectangles that we all grew up on, column-staggered, and straight-up ortholinear, which no longer enjoy the popularity they once did.
At this point, the guide becomes a bit of a Choose Your Own Adventure story. If you want a split but don’t want to learn to change much if at all about your typing style, keep reading, because there are definitely options.
But if you’re ready to commit to typing correctly for the sake of ergonomics, you can skip the Alice and other baby ergo choices and get your membership to the light side. First are features — you must decide what you need to get various jobs done. Then you learn a bit about key map customization, including using a non-QWERTY layout. Finally, there’s the question of buying versus DIYing. All the choices are yours, so go for it!
Need something ultra-portable for those impromptu sessions at the coffee shop (when you can actually find a table)? You can’t get much smaller than the 28-key Koumori by [fata1err0r81], which means “bat” in Japanese. Here’s the repo.
This unibody beauty runs on an RP2040 Zero using QMK firmware. That 40 mm Cirque track pad has a glass overlay, which is a really nice touch. It’s actually a screen protector for a smart watch, and the purple bit is some craft vinyl cut to size.
Protecting that glass overlay is a case with a handle and a magnetic lid. Both the PCB and the case were designed in Ergogen, which as you know, I really like to see people using.
As you might have guessed, those are Kailh V1 choc switches with matching key caps. If you want a bat for your pocket, the build guide is simple, and there aren’t even any microscopic parts involved.
Advertisement
The Centerfold: [arax20]’s Been Workin’ On the Railroader
Okay, before you do anything, go check out the image gallery to see this baby glowing and being worn like a katana or something. Yeah.
So [arax20] built this as a gift for an ex. She likes the ergonomics of splits, but didn’t want cables between the halves and feels the space between is otherwise wasted. Really? There’s so much you can put there, from cats to mice to coffee mugs.
Do you rock a sweet set of peripherals on a screamin’ desk pad? Send me a picture along with your handle and all the gory details, and you could be featured here!
Historical Clackers: the Mysterious Rico
Frustratingly little is known about the Rico, a 1932 index machine out of Nuremburg, Germany. But the Antikey Chop has over a dozen books on typewriters, and only two have any mention of the Rico: Adler’s Antique Typewriters, From Creed to QWERTY, and Dingwerth’s Kleines Lexikon Historischer Schreibmaschinen.Image via The Antikey Chop
Adler calls it a “pleasant toy typewriter with indicator selecting letters from a rectangular index”, saying nothing more descriptive. Dingwerth’s volume both dates the Rico and lists the maker as Richard Koch & Co. of Nuremburg.
The Rico was ambitiously declared the No. A1 model, though there is no evidence of any other model in existence. It was made mostly of stamped tin, though the type element was made of brass. The type element looked like a tube cut in half lengthwise, and worked in a similar fashion to the Chicago typewriter with its type sleeve.
Advertisement
There are some interesting things about the Rico nonetheless. The platen could not accommodate paper wider than 4″, for one thing. There is also no inking system to speak of. Weirder still, this oversight isn’t mentioned in the original instructions. Most people just taped a couple inches of typewriter ribbon between the element and the platen and called it good .
To use the thing, you would move the center lever to the character you wanted. The lever has a pin in the bottom, and each character has a dimple in it for the pin to sit. The lever on the left side was used to pivot the carriage toward the type element in order to print. In total, the Rico typed 74 characters plus Space.
Finally, Someone’s Made a Braille Keyboard, and It’s Inexpensive
Once upon a time, New Jersey high schooler Umang Sharma saw an ad for a Braille keyboard. The price? A cool seven grand. For a keyboard. No problem, he thought. I can build my own.
Image via NJ.com
The astute among you will notice that there’s a Logitech keyboard in the picture, with what look like key cap hats. That is exactly what’s happening here. Sharma starts with a standard keyboard base, one that is usually either donated or was previously discarded.
He then focuses on the most important accessibility layer, which is tactile Braille key caps that are both readable and durable. In 2022, Sharma launched the non-profit Jdable to bring affordable, accessible design to people with disabilities.
Advertisement
He designed the key caps himself, and uses a combination of 3D printing and other materials to create them in bulk. They’re printed using a combination of PETG for toughness, TPU for grippiness, and resin for definition. The key caps are attached to the standard set with a strong adhesive.
Sharma has a team of student volunteers that help him build the keyboards and distribute them, and they have reached nearly 1,000 blind or visually-impaired students in the U.S. and abroad.
A lounge area at OpenAI’s new Bellevue office. (Trevor Tondro Photo for OpenAI)
OpenAI officially opened its new engineering office in downtown Bellevue, Wash., on Thursday, unveiling a retro-modern, wood-paneled space for its 250 employees in the region — with enough room in the tower to ultimately accommodate as many as 1,400 people.
It’s already the ChatGPT and Codex maker’s biggest office outside its San Francisco headquarters, and a sign of the AI industry’s impact on the Seattle area.
“This is a monumental day for OpenAI and Bellevue,” said Vijaye Raji, OpenAI’s CTO of applications, as he cut the ceremonial ribbon with Bellevue Mayor Mo Malakoutian.
OpenAI CTO of Applications Vijaye Raji (left) and Bellevue Mayor Mo Malakoutian prepare to cut the ribbon at the opening of OpenAI’s new Bellevue office. (GeekWire Photo / Todd Bishop)
The office puts OpenAI within close proximity of two of its biggest investors and partners: Microsoft in nearby Redmond and Amazon in Bellevue and Seattle. The opening comes less than a week after Amazon announced a $50 billion investment in the company.
It marks the latest milestone in OpenAI’s rapid expansion. The company first arrived in Bellevue in 2024, seeking to tap the region’s engineering talent pool. Last month, OpenAI scaled up, signing a lease to boost its footprint to nearly 300,000 square feet in City Center Plaza.
OpenAI currently occupies two floors with the ability to add 10 more as it grows.
Advertisement
The Bellevue office includes teams working on infrastructure, ChatGPT, research, and advertising, in addition to partnerships, an early sign of its expansion beyond engineering.
Statsig, the Bellevue startup Raji founded in 2021, forms the nucleus of the new office. OpenAI acquired the company for $1.1 billion last year, bringing Raji aboard as a key technical leader.
The space is built around a sweeping wood-clad central staircase connecting its two current floors, and lounge-like common areas designed for informal gatherings, including a library (yes, there are a few books) and a game room. Those were deliberate choices to encourage the kinds of connections that remote work can’t replicate, Raji said in an interview at the event.
A staircase connects the two floors of OpenAI’s new Bellevue office. The space was designed by Rapt Studio and built by general contractor BnBuilders. (Trevor Tondro Photo)
Malakoutian, the Bellevue mayor, called the opening “a vote of confidence” in the city, which has specifically courted AI companies as part of a broader economic development push.
In a recent interview with GeekWire, Malakoutian said companies are drawn to predictable permitting, modern infrastructure, and quality of life, offering a competitive edge in recruiting. A light rail line connecting the Eastside to Seattle across Lake Washington opens this month.
Advertisement
Elon Musk’s xAI is creating an engineering center a short walk away. Cloud and AI infrastructure company Crusoe opened a Bellevue office last year. Companies including Snap, Anduril, Shopify, Snowflake, Uber, and Databricks have signed new or expanded leases in the city.
Gov. Bob Ferguson, appearing via recorded video, noted that the region ranks among the top in the country for AI talent, saying it’s “very well-positioned to become a global hub for AI.”
The library at OpenAI’s new Bellevue office. (Trevor Tondro Photo)
Matt McIlwain, managing director at Madrona Venture Group, which was an early investor in Statsig, called the new office an example of a “virtuous cycle” of local founders building startups that attract larger employers. He credited Raji for pushing to build a critical mass for OpenAI in Bellevue, which has been “more on its front foot” than Seattle in courting tech companies.
But given ongoing tax debates in the state, in which McIlwain and others in the tech community have been vocal, he questioned whether lawmakers appreciate the dynamic.
“The folks in Olympia clearly do not understand that flywheel,” he said.
Advertisement
For Raji, the opening is the latest chapter in a larger story. The region has been his home for 23 years, starting when Microsoft recruited him to the area. He later joined Facebook’s Seattle office and helped it grow locally from a handful of employees to 5,000 as its regional leader.
In that way, the OpenAI expansion is part of a familiar pattern.
“You can see the sequence,” Raji said, crediting the region’s talent pool and growth. “So it’s only natural that now, with all the AI investments, this area is again back in the center.”
The former chief financial officer for Seattle-area retail software company Fabric was sentenced in U.S. District Court in Seattle Thursday to two years in prison for a wire fraud scheme that involved the misuse of $35 million from his former employer.
Nevin Shetty, 42, of Mercer Island, Wash., was found guilty last November, after a nine-day jury trial, of four counts of wire fraud.
“The loss had significant and severe effects on the company,” Judge Tana Lin told Shetty at the sentencing hearing, saying that his actions cost the jobs of 60 people. “You almost put the company out of business. … You were playing with money that wasn’t yours.”
The United States Attorney’s office for the Western District of Washington was seeking a nine-year sentence, according to a sentencing memorandum ahead of Thursday’s court action.
Assistant U.S. Attorney Philip Kopczynski wrote to the court that “Shetty’s serious crime deserves stern punishment,” calling it “a calculated scheme motivated by greed and meticulously carried out over many months.”
Advertisement
Shetty was ordered to pay $35,000,100 and will be on supervised release for three years after prison. Judge Lin also imposed a special condition that he not serve as an officer or director of a company without prior permission from the probation office.
Shetty joined Fabric as CFO in March 2021. The company, led at the time by several former Amazon executives, had just raised $43 million in new funding, and Shetty helped draft a policy governing how the money raised should be invested conservatively while the company worked to grow its business. Four months later, Fabric raised another $100 million and in February 2022 raised a $140 million Series C round to reach a valuation of $1.5 billion.
Prosecutors said Shetty diverted funds in early 2022 to his own cryptocurrency business, HighTower Treasury, without authorization. Although he helped create the company’s policy limiting investments to low-risk accounts, he secretly moved the money into high-yield decentralized finance platforms that promised 20% returns.
According to records, Shetty’s plan was to pay his employer 6% interest and keep the rest of the profits through HighTower. In the first month, he and a partner made about $133,000, but by May 2022, the crypto investments had collapsed, wiping out nearly all $35 million.
Advertisement
After confessing to colleagues, Shetty was fired and the company reported the theft to the FBI.
An anonymous reader quotes a report from 80 Level: Microsoft has officially confirmed development of its next-generation Xbox console, currently known internally as Project Helix. While concrete details remain limited, early information suggests the company is positioning the device as a hybrid between a traditional console and a gaming PC, capable of running both Xbox titles and PC games. The codename was revealed recently by new Xbox CEO Asha Sharma, who reaffirmed Microsoft’s continued commitment to dedicated gaming hardware despite speculation that the company might shift entirely toward cloud or platform-based ecosystems. According to Sharma, Project Helix represents the next step in Xbox’s console strategy.
Although official specifications have not yet been announced, early reports indicate the system will likely rely on a new AMD system-on-chip combining Xbox hardware with PC-style architecture. The device is expected to emphasize high performance while maintaining compatibility with existing Xbox game libraries. […] If the concept holds, Project Helix could mark a significant shift in how console ecosystems are structured, moving away from tightly closed hardware platforms toward something closer to a unified PC-console environment. Sharma wrote in a post on X: “Great start to the morning with Team Xbox, where we talked about our commitment to the return of Xbox, including Project Helix, the code name for our next generation console. Project Helix will lead in performance and play your Xbox and PC games. Looking forward to chatting about this more with partners and studios at my first GDC next week!”
The Department of Justice describes LeakBase as a clearinghouse for hacked databases. Alongside the “hundreds of millions” of account credentials, it offers credit and debit card numbers, banking details, and other personally identifiable information that can fuel account takeovers and fraud. Read Entire Article Source link
Iron is essential to biological function, and iron-based nanomaterials may become valuable tools in the long-term effort to develop cancer treatments. Researchers at Oregon State University have engineered a new “nanoagent” using an iron-based metal-organic framework (MOF) structure and demonstrated its ability to destroy cancerous cells in laboratory experiments. Read Entire Article Source link
On the surface, the change might look like semantics. But it reflects a broader evolution inside Apple Silicon that moves beyond the traditional mobile-processor model where “big” cores handle demanding work and “little” cores handle background tasks for efficiency’s sake. By renaming its cores (in pure Apple’s marketing style), the… Read Entire Article Source link
Huang’s explanation was brief, but the implications are broad. Nvidia, whose products have become indispensable to generative AI infrastructure, sits in a position few companies have ever occupied: both supplier and shareholder to the firms building the software atop its hardware. That arrangement, once mutually reinforcing, now appears increasingly tangled. Read Entire Article Source link
The episode 1 premiere began with the shock news of Monica Dutton (Kelsey Asbille) dying offscreen since we last saw her in Yellowstone, despite her marriage to Kayce (Luke Grimes) having been dissolved.
Surely there’s still plenty more surprises to come, but when does Marshals: A Yellowstone Story episode 2 arrive on CBS and Paramount+?
Advertisement
What time can I watch Marshals: A Yellowstone Story episode 2 on Paramount+?
Marshals 1×02 Promo “Zone of Death” (HD) Luke Grimes Yellowstone spinoff – YouTube