SoylentNews

upstart writes:

Neither war nor bad sensors nor a nautical journey will keep this probe from building a 3D map of space:

It's been a decade since it was announced, but the Euclid mission to build a 3D map of the universe is finally getting close to launch with the spacecraft landing in Florida ahead of an expected July liftoff.

The Euclid mission will send the eponymous probe to Earth-Sun Lagrange point 2 – the same spot as NASA's James Webb Telescope, where it will observe more than a third of the sky and detect galaxies as distant as 10 billion light years away.

[...] The ultimate goal of the European Space Agency mission – which has enjoyed important contributions from NASA – is to build a 3D map of the large-scale structure of the universe to help scientists better understand the nature of dark matter and dark energy, which make up more than 95 percent of the mass and energy of the universe. Scientists with the European Space Agency also hope Euclid will help them better understand how the expansion of the universe has changed over time, and whether we have a complete understanding of gravity.

"No test of the [general theory of relativity] has been made with high precision over the large distances and times that Euclid will cover. This way, Euclid will reveal if general relativity breaks down at the largest scales. If it does, physicists will need to go back to the drawing board," ESA said.

[...] If everything in the clean room goes to plan, Euclid will then be mounted atop its Falcon 9 ride to begin the final leg of its journey. If the James Webb's trip to L2 is any indicator, Euclid should reach its destination in around 30 days.

Original Submission

owl writes:

https://www.righto.com/2022/11/a-bug-fix-in-8086-microprocessor.html

While reverse-engineering the 8086 from die photos, a particular circuit caught my eye because its physical layout on the die didn't match the surrounding circuitry. This circuit turns out to implement special functionality for a couple of instructions, subtlely changing the way they interacted with interrupts. Some web searching revealed that this behavior was changed by Intel in 1978 to fix a problem with early versions of the 8086 chip. By studying the die, we can get an idea of how Intel dealt with bugs in the 8086 microprocessor.

In modern CPUs, bugs can often be fixed through a microcode patch that updates the CPU during boot.1 However, prior to the Pentium Pro (1995), microprocessors could only be fixed through a change to the design that fixed the silicon.

Original Submission

TSMC May Charge 30 Percent More for Chips Made at its Arizona Fabs

upstart writes:

The chipmaker is looking for $15 billion in subsidies from the US government:

Setting up a chip factory in the US is an expensive endeavor, especially when compared to doing the same in Asia. Still, companies like TSMC are investing billions into doing just that, wooed by government subsidies and the idea of achieving a geographically diverse supply chain that is less prone to disruptions. As a result, chips made by the Taiwanese giant in the US could end up 30 percent more expensive than those made in its home country.

Back in December, TSMC announced it would pour no less than $40 billion into building two advanced chip factories in Arizona. In other words, the world's biggest chipmaker made the largest direct foreign investment in Arizona history and one of the largest in US history. When they become operational, the fabs will employ thousands of people and produce around 600,000 wafers per year.

The move represents a significant shift in mindset for TSMC leadership, which has traditionally been reluctant to establish chip manufacturing facilities in the US. One reason has to do with Taiwan's policy of keeping its most advanced technologies locally even as it builds external manufacturing arms for a more resilient global supply chain.

Another reason has to do with cost – setting up and operating a factory on American soil is up to 50 percent more expensive than in Taiwan. Then you'd need to source materials like rare earth metals and semiconductor-grade neon as well as a steady supply of water to keep the facility running at full capacity. And last but not least, you need a lot of skilled (and cheap) labor, something that is easier to find in Asia as opposed to North America and even Europe.

Naturally, this means TSMC will have extra costs in the US that will translate into higher prices for wafers made in Arizona as opposed to those made in Taiwan. Some industry insiders estimate that chips made in America on N5 and N4 process nodes will cost up to 30 percent more than those made in the Asian country.

It's not just the US that's more expensive than Taiwan for chip manufacturers, either. Chips made on older, more mature process technologies like N28, N22, N16, and N12 at TSMC's Japanese arm in Kumamoto are also 10-15 percent more expensive to make than similar chips made in the company's home country.

Just like the US, Japan is looking o reignite its once-dominant semiconductor industry. Last year, we learned the two countries kicked off a joint research and development arm for 2nm-class process technologies and beyond.

upstart writes:

Netflix Loses A Million Subscribers In Spain After Greedy Password Sharing Crackdown:

We've noted repeatedly how Netflix's password sharing crackdown is a stupid cash grab that alienates and annoys loyal customers, duplicates existing efforts to restrict "freeloaders," won't give the company the financial windfall it thinks, and just generally represents how the company has inevitably shifted from innovative disruptor to the kind of tone deaf cable giants it used to criticize.

The plan basically involves charging users an extra $2-$3 a month if it's found that someone is using your account outside of your home. The problem: Netflix has already been imposing blanket price hikes, and it already limits the number of simultaneously streams per account, forcing users to subscribe to more expensive tiers if they want to expand the limit.

While the crackdown isn't expected to hit U.S. subscribers until the end of the second quarter (aka soon), the effort has generally been a hot mess in the smaller countries Netflix first used as guinea pigs to test both the underlying tech and company messaging.

The company had to suspend the efforts in countries like Argentina, El Salvador, Guatemala, Honduras, and the Dominican Republic, after users were equal parts befuddled and annoyed. And in Spain, estimates are that the company saw a defection of more than one million subscribers largely due to the higher, unnecessary fees:

"There are of course inherent risks with clamping down on password sharing, particularly when back in 2017 Netflix was seen to be actively encouraging it. Some users were expected to be lost in the process but losing over 1 million users in a little over a month has major implications for Netflix and whether it decides to continue with its crackdown globally.

Interestingly, there is no strong demographic skew to those who cancelled, signaling a more outright rejection of the password sharing clampdown. In a worrying sign for the next quarter, 10% of remaining Netflix subscribers say they plan to cancel their plan in Q2 2023, which is well above the average seen in previous quarters."

---

Original Submission

SpaceX Completes First Fully Expendable Falcon Heavy Mission

SpaceX Completes First Fully Expendable Falcon Heavy Mission:

The triple-core rocket blasted off on Sunday evening to deliver three satellites to a high Earth orbit.

SpaceX's giant Falcon Heavy rocket successfully delivered three satellites to high Earth orbit on April 30. The launch marked the first time that none of the rocket's boosters were recovered.

Wild SpaceX Video Shows Hottest Reentry Yet of Reusable Rocket Fairing

upstart writes:

Wild SpaceX Video Shows Hottest Reentry Yet of Reusable Rocket Fairing:

The dramatic footage shows a Falcon Heavy fairing blazing through the atmosphere at speeds reaching Mach 15.

The most recent flight of SpaceX's Falcon Heavy was historic in that it was the rocket's first fully expendable mission, and it was also the first Falcon Heavy mission to include previously flown fairings. As new footage attests, the fiery return of these fairings was a sight to behold.

After several delays, the Falcon Heavy blasted off from Launch Complex 39A at NASA's Kennedy Space Center in Florida on Sunday, April 30 at 8:26 p.m. ET. The triple-core rocket successfully deployed its primary payload, the broadband ViaSat-3 Americas satellite, and two smaller satellites to geosynchronous Earth orbit (GEO).

Key to SpaceX's ongoing success is its devotion to reusability—it's a lot cheaper to recover your engines and reuse them than to produce new ones for every launch. For this mission, however, none of the rocket's three booster stages could be recovered, as they expelled all their fuel in the effort to get the payloads to GEO. That said, SpaceX did make the attempt to recover the rocket's fairings, so in that sense it wasn't a fully expendable mission.

An Anonymous Coward writes:

T-Mobile disclosed the second data breach of 2023 after discovering that attackers had access to the personal information of hundreds of customers for more than a month, starting late February 2023.

Compared to previous data breaches reported by T-Mobile, the latest of which impacted 37 million people, this incident affected only 836 customers. Still, the amount of exposed information is highly extensive and exposes affected individuals to identity theft and phishing attacks.

"In March 2023, the measures we have in place to alert us to unauthorized activity worked as designed and we were able to determine that a bad actor gained access to limited information from a small number of T-Mobile accounts between late February and March 2023," the company said in data breach notification letters sent to affected individuals just before the weekend, on Friday, April 28, 2023.

T-Mobile said the threat actors didn't gain access to call records or affected individuals' personal financial account info, but the exposed personally identifiable information contains more than enough data for identity theft.

While the exposed information varied for each of the affected customers, it could include "full name, contact information, account number and associated phone numbers, T-Mobile account PIN, social security number, government ID, date of birth, balance due, internal codes that T-Mobile uses to service customer accounts (for example, rate plan and feature codes), and the number of lines."

After detecting the security breach, T-Mobile proactively reset account PINs for impacted customers and now offers them two years of free credit monitoring and identity theft detection services through Transunion myTrueIdentity.

Original Submission

IBM to Stop Hiring for Jobs That AI Could Do

upstart writes:

Routine tasks like transferring employees between departments are likely to be fully automated:

American tech major IBM anticipates pausing hiring for positions that it believes artificial intelligence (AI) will eventually take over.

In an interview with Bloomberg, IBM CEO Arvind Krishna said the company will suspend or pause hiring for back-office functions such as human resources.

The company employs some 26,000 people in these non-customer-facing roles, Krishna said.

"I could easily see 30% of that getting replaced by AI and automation over a five-year period," he added.

[...] Routine tasks like transferring employees between departments or providing letters of employment verification are likely to be fully automated, said the company's chief.

Over the next ten years, he continued, it is likely that some HR functions related to workforce composition analysis and productivity will not be replaced.

upstart writes:

It looks like Microsoft is gearing up for a long and difficult fight with regulators for its $68.7 billion deal:

Microsoft is furious. Last week, a surprise decision from the UK's Competition and Markets Authority (CMA) left its $68.7 billion deal to acquire Activision Blizzard blocked in Britain, thanks to concerns about the future of cloud gaming.

Microsoft president Brad Smith was awake at 2AM that morning hastily writing a response from across the pond, according to Bloomberg. He spoke to the BBC a day later and called the UK regulator's decision the "darkest day" for Microsoft in its four decades of working in Britain. He went a step further and said "the European Union is a more attractive place to start a business" than the UK, a particularly stinging statement given the political issues around Brexit.

Now, Microsoft is bruised, angry, and plotting its next move. If Brad Smith's fighting talk is anything to go by, Microsoft will try to keep this deal alive. But the CMA's decision won't be an easy one to appeal.

[...] Meta's battle with the CMA over its Giphy acquisition shows what Microsoft might be in store for. Meta was originally ordered to sell Giphy in 2021 but appealed the ruling and was unsuccessful. Meta eventually had to comply with the UK competition watchdog and divest itself of social media GIF library Giphy. Viagogo's $4 billion takeover of StubHub was also partially blocked by the CMA, forcing the company to keep StubHub's US and Canadian operations but sell its UK and international businesses.

[...] The CMA said in September that it was concerned about the effects of Microsoft owning Activision Blizzard games on existing rivals and emerging entrants offering multi-game subscriptions and cloud gaming services. I tweeted at the time that all of the headlines around Call of Duty were just noise, and there would be bigger concerns around Microsoft's ability to leverage Windows and Azure, unlike its competitors, and how it could influence game distribution and revenue shares across the game industry with its Xbox Game Pass subscription.

upstart writes:

A new system was able to capture exact words and phrases from the brain activity of someone listening to podcasts:

A noninvasive brain-computer interface capable of converting a person's thoughts into words could one day help people who have lost the ability to speak as a result of injuries like strokes or conditions including ALS.

In a new study, published in Nature Neuroscience today, a model trained on functional magnetic resonance imaging scans of three volunteers was able to predict whole sentences they were hearing with surprising accuracy—just by looking at their brain activity. The findings demonstrate the need for future policies to protect our brain data, the team says.

Speech has been decoded from brain activity before, but the process typically requires highly invasive electrode devices to be embedded within a person's brain. Other noninvasive systems have tended to be restricted to decoding single words or short phrases.

This is the first time whole sentences have been produced from noninvasive brain recordings collected through fMRI, according to the interface's creators, a team of researchers from the University of Texas at Austin. While normal MRI takes pictures of the structure of the brain, functional MRI scans evaluate blood flow in the brain, depicting which parts are activated by certain activities.

[...] Romain Brette, a theoretical neuroscientist at the Vision Institute in Paris who was not involved in the experiment, is not wholly convinced by the technology's efficacy at this stage. "The way the algorithm works is basically that an AI model makes up sentences from vague information about the semantic field of the sentences inferred from the brain scan," he says. "There might be some interesting use cases, like inferring what you have dreamed about, on a general level. But I'm a bit skeptical that we're really approaching thought-reading level."

upstart writes:

Software vendors and the EU weren’t interested, so giving it away became the best option:

The European Organization for Nuclear Research (CERN) on Sunday celebrated the 30th anniversary of releasing the World Wide Web into the public domain.

As the World Wide Web Consortium's brief history of the web explains, in 1989 Tim Berners-Lee - then a fellow at CERN - proposed that the organization adopt "a global hypertext system." His first name for the project was "Mesh".

And as the Consortium records, in 1990 Berners-Lee set to work on "a hypertext GUI browser+editor using the NeXTStep development environment. He makes up 'WorldWideWeb' as a name for the program."

Berners-Lee's work gathered a very appreciative audience inside CERN, and soon started to attract attention elsewhere. By January 1993, the world had around 50 HTTP servers. The following month, the first graphical browser – Marc Andreessen's Mosaic – appeared.

Alternative hypertext tools, like Gopher, started to lose their luster.

On April 30, 1993, CERN signed off on a decision that the World Wide Web – a client, server, and library of code created under its roof – belonged to humanity (the letter was duly stamped on May 3).

"CERN relinquishes all intellectual property rights to this code, both source and binary form, and permission is granted for anyone to use, duplicate, modify and redistribute it" states a letter signed on that day by Walter Hoogland and Helmut Weber – at the time respectively CERN's director of research and director of administration.

upstart writes:

Pornhub protests Utah age verification law by blocking the state's access:

MindGeek, the owner of adult platforms such as Pornhub, has blocked everyone in Utah from accessing its sites in protest of the state's age verification law that has just come into effect.

Utah has been fighting against online pornography for years. It called porn a public health crisis in 2016 and previously proposed that all smartphones and tablets in the state automatically block pornography. An age verification law was eventually passed in March, requiring users visiting adult platforms deemed "harmful to minors" to verify their age before being allowed access. Axios writes that any companies that don't comply with the law will be liable if they're sued over minors accessing their content.

Now that the law has gone into effect, MindGeek has responded by blocking anyone in Utah who tries to access Pornhub. Those with Utah IPs will see only a video of adult performer Cherie DeVille, a member of the Adult Performer Advocacy Committee, explaining the reason for the block.

"As you may know, your elected officials in Utah are requiring us to verify your age before allowing you access to our website," DeVille says. "While safety and compliance are at the forefront of our mission, giving your ID card every time you want to visit an adult platform is not the most effective solution for protecting our users, and in fact, will put children and your privacy at risk."

DeVille adds that "mandating age verification without proper enforcement" drives users to other sites with fewer safety measures in place.

[...] Utah's new laws also extend to social media companies. From March 1 next year, those under 18 will require a parent's permission before opening an account on social media platforms. Companies must also give parents access to their kids' posts, messages, and responses; are barred "from using a design or feature that causes a minor to have an addiction to the company's social media platform;" and must block under-18s from using social media between 10:30 pm and 6:30 am.

Original Submission

The Morning After: the Godfather of AI Leaves Google Amid Ethical Concerns

upstart writes:

The Morning After: The Godfather of AI leaves Google amid ethical concerns:

Geoffrey Hinton, nicknamed the Godfather of AI, told The New York Times he resigned as Google VP and engineering fellow in April to freely warn of the risks associated with the technology. The researcher is concerned Google is giving up its previous restraint on public AI releases to compete with ChatGPT, Bing Chat and similar models. In the near term, Hinton says he's worried that generative AI could lead to a wave of misinformation. You might "not be able to know what is true anymore," he says. He's also concerned it might not just eliminate "drudge work," but outright replace some jobs – which I think is a valid worry already turning into a reality.

Cops Raid Swedish VPN Provider Only to Find Out There's No 'There' There

upstart writes:

Cops Raid Swedish VPN Provider Only To Find Out There's No 'There' There:

[...] So, it always gives me pleasure to learn that cops armed with court orders approached a privacy oriented tech company only to find out the stuff they wanted didn't actually exist at the place they searched. Due diligence is a thing, investigators. Your boilerplate is obviously false if you've claimed (based on "training and expertise") that the place you want to search contains the information you wish to obtain.

That's the case here. A Swedish VPN provider was raided by local law enforcement, but was unable to produce any of the information officers were searching for... something officers might have realized prior to the search if they'd bothered to read the terms of service. Here's Michael Kan with the details for PC World:

The company today reported that Swedish police had issued a search warrant two days earlier to investigate Mullvad VPN's office in Gothenburg, Sweden. "They intended to seize computers with customer data," Mullvad said.

However, Swedish police left empty-handed. It looks like Mullvad's own lawyers stepped in and pointed out that the company maintains a strict no-logging policy on customer data. This means the VPN service will abstain from collecting a subscriber's IP address, web traffic, and connection timestamps, in an effort to protect user privacy. (It's also why Mullvad VPN is among our most highly ranked VPN services.)

If the cops had run a search of Mullvad's website before running a physical search of its offices, it might have discovered the stuff they swore would be found there actually wouldn't be found on Mullvad's premises. It's not like it's that difficult to find:

There is a law to collect user data in India and other countries. Does this affect Mullvad?

Mullvad does not collect user data. Mullvad is based in Sweden and none of the Swedish regulations (https://mullvad.net/help/swedish-legislation/) can force VPN providers to secretly collect traffic-related data. We also have no servers, infrastructure or staff in India.

upstart writes:

Improvements to future versions of Python are set to speed it up, slim it down, and pave the way toward even better things:

Because Python is a dynamic language, making it faster has been a challenge. But over the last couple of years, developers in the core Python team have focused on various ways to do it.

At PyCon 2023, held in Salt Lake City, Utah, several talks highlighted Python's future as a faster and more efficient language. Python 3.12 will showcase many of those improvements. Some are new in that latest version, others are already in Python but have been further refined.

Mark Shannon, a longtime core Python contributor now at Microsoft, summarized many of the initiatives to speed up and streamline Python. Most of the work he described in his presentation centered on reducing Python's memory use, making the interpreter faster, and optimizing the compiler to yield more efficient code.

Other projects, still under wraps but already showing promise, offer ways to expand Python's concurrency model. This will allow Python to better use multiple cores with fewer of the tradeoffs imposed by threads, async, or multiprocessing.

[...] One long-dreamed way to solve this problem is to remove Python's GIL, or Global Interpreter Lock. The GIL synchronizes operations between threads to ensure objects are accessed by only one thread at a time. In theory, removing the GIL would allow true multithreading. In practice—and it's been tried many times—it slows down non-threaded use cases, so it's not a net win.

Core python developer Eric Snow, in his talk, unveiled a possible future solution for all this: subinterpreters, and a per-interpreter GIL. In short: the GIL wouldn't be removed, just sidestepped.

Subinterpreters is a mechanism where the Python runtime can have multiple interpreters running together inside a single process, as opposed to each interpreter being isolated in its own process (the current multiprocessing mechanism). Each subinterpreter gets its own GIL, but all subinterpreters can share state more readily.

upstart writes:

The European Union is writing legislation that would hold accountable companies that create generative AI platforms:

A proposed set of rules by the European Union would, among other things. require makers of generative AI tools such as ChatGPT,to publicize any copyrighted material used by the technology platforms to create content of any kind.

A new draft of European Parliament's legislation, a copy of which was attained by The Wall Street Journal, would allow the original creators of content used by generative AI applications to share in any profits that result.

The European Union's "Artificial Intelligence Act" (AI Act) is the first of its kind by a western set of nations. The proposed legislation relies heavily on existing rules, such as the General Data Protection Regulation (GDPR), the Digital Services Act, and the Digital Markets Act. The AI Act was originally proposed by the European Commission in April 2021.

The bill's provisions also require that the large language models (LLMs) behind generative AI tech, such as the GPT-4, be designed with adequate safeguards against generating content that violates EU laws; that could include child pornography or, in some EU countries, denial of the Holocaust, according to The Washington Post.

[...] But the solution to keeping AI honest isn't easy, according to Avivah Litan, a vice president and distinguished analyst at Gartner Research. It's likely that LLM creators, such as San Fransisco-based OpenAI and others, will need to develop powerful LLMs to check that the ones trained initially have no copyrighted materials. Rules-based systems to filter out copyright materials are likely to be ineffective, Liten said.