Version 9.3 of NetBSD is here, able to run on very low-end systems and with that authentic early-1990s experience. The Register reports: Version 9.3 comes some 15 months after NetBSD 9.2 and boasts new and updated drivers, improved hardware support, including for some recent AMD and Intel processors, and better handling of suspend and resume. The next sentence in the release announcement, though, might give some readers pause: "Support for wsfb-based X11 servers on the Commodore Amiga." This is your clue that we are in a rather different territory from run-of-the-mill PC operating systems here. A notable improvement in NetBSD 9.3 is being able to run a graphical desktop on an Amiga. This is a 2022 operating system that can run on late-1980s hardware, and there are not many of those around.

NetBSD supports eight "tier I" architectures: 32-bit and 64-bit x86 and Arm, plus MIPS, PowerPC, Sun UltraSPARC, and the Xen hypervisor. Alongside those, there are no less than 49 "tier II" supported architectures, which are not as complete and not everything works -- although almost all of them are on version 9.3 except for the version for original Acorn computers with 32-bit Arm CPUs, which is still only on NetBSD 8.1. There's also a "tier III" for ports which are on "life support" so there may be a risk Archimedes support could drop to that. This is an OS that can run on 680x0 hardware, DEC VAX minicomputers and workstations, and Sun 2, 3, and 32-bit SPARC boxes. In other words, it reaches back as far as some 1970s hardware. Let this govern your expectations. For instance, in VirtualBox, if you tell it you want to create a NetBSD guest, it disables SMP support.

China's first fully autonomous, commercial robotaxi rides -- with no safety drivers -- are about to open for public passengers in Wuhan and Chongqing, marking an inflection point for one of the key technological revolutions of the 21st century. New Atlas reports: The two newly-issued permits allow Baidu to charge for driverless rides within a 13-sq-km (5-sq-mi) area in Wuhan, between 9 am and 5 pm, and within a larger 30-sq-km (11.6-sq-mi) zone in Chonqing's Yongchuan district between 9.30 am and 5.30 pm -- so while they're currently set to avoid peak hours, they'll be mixing it up with plenty of daytime traffic. Each zone will run five 5th-generation Apollo cars, with remote drivers ready to assume control if the vehicles get themselves into any sticky situations. Home base will be watching closely through the cars' camera systems, particularly in these early days.

Baidu's Apollo Go is already the world's biggest robotaxi company, with operations already live in all tier-one Chinese cities using the same 5th-gen car, with backup drivers on board. The company recently revealed its 6th-gen design, its first ground-up fully autonomous car for mass production. The Apollo RT6 will cost just RMB 250,000 (US$37,000) to manufacture, says Baidu, and its optional, removable steering wheel and generous, configurable cabin space will make it one of the first proper mobility pod-type services when it hits the streets commercially in 2023.

Facebook announced on Thursday it will begin testing end-to-end encryption as the default option for some users of its Messenger app on Android and iOS. The Guardian reports: Facebook messenger users currently have to opt in to make their messages end-to-end encrypted (E2E), a mechanism that theoretically allows only the sender and recipient of a message to access its content. Facebook spokesperson Alex Dziedzan said on Thursday that E2E encryption is a complex feature to implement and that the test is limited to a couple of hundred users for now so that the company can ensure the system is working properly. Dziedzan also said the move was "not a response to any law enforcement requests." Meta, Facebook's parent company, said it had planned to roll out the test for months. The company had previously announced plans to make E2E encryption the default in 2022 but pushed the date back to 2023. "The only way for companies like Facebook to meaningfully protect people is for them to ensure that they do not have access to user data or communications when a law enforcement agency comes knocking," Evan Greer, the director of the digital rights group Fight for the Future, said. "Expanding end-to-end encryption by default is a part of that, but companies like Facebook also need to stop collecting and retaining so much intimate information about us in the first place."

Earlier this week, Samsung unveiled their new Z Fold 4 and Z Flip 4 -- two of the most refined and polished foldable smartphones on the market. However, what Samsung hasn't done (or any other phone manufacturer for that matter) "is make the case for why you'd actually want a foldable phone," writes David Pierce via The Verge. "And until it can explain why it's worth all the extra cost and tradeoffs, I'm having a hard time figuring out why you'd be willing to give up the phone you know and love to get one." From the report: What Samsung needs to do with the Galaxy Fold (and the rest of the industry will eventually need to do with their own foldables) is convince people that it's worth buying a phone that's more expensive, more fragile, and takes up more room in your pocket. Right now, the worst thing about foldables is that they force you to make significant sacrifices on the most important device you own: your smartphone. The new Fold 4 is a little shorter, about an ounce heavier, and about twice as thick as the Galaxy S22 Ultra. It's also $600 more expensive. The Ultra has a bigger battery, better camera specs, and a 6.8-inch screen that supports an S Pen. The Fold 4, when opened, is noticeably larger, but the candy bar phones still get plenty big. And Fold makes a lot of sacrifices for some more real estate.

It's not even clear to me that Samsung knows why you should make all of those sacrifices. On its website, one of the first selling points the company offers is that you can prop up the screen on a table by opening it halfway for watching or taking videos hands-free. Here in reality, we call that a kickstand, and this is an awfully expensive one. In this mode, you're also only using half the screen, which sort of defeats the whole purpose. So far, multitasking seems to be the foldable's one actual advantage. Open up your Galaxy Fold, and you can run two apps side by side or even three or four on the screen at once! This, I agree, is a delightful thing. Being able to use my browser and my notes app side by side or see my calendar and my email together is much better than constantly swiping between two full-screen apps. And seeing two pages at a time in the Kindle app is the best. And you know what? Big screens are just good -- good for games, good for reading, good for watching Netflix.

But these aren't just arguments for foldables; they're arguments for tablets. And so far, the arguments for Android tablets don't seem to be convincing many users. While Android has gotten better as a large-screen operating system, and the Fold 4's software being based on Android 12L is a good sign, too many apps that are "optimized" for foldables are actually just sticking a giant sidebar onto one side, which doesn't accomplish much. Others just streeeetch everything to fit the larger screen. Don't even get me started on how the vast majority of apps deal with Microsoft's approach of two separate screens attached with a hinge. Samsung has done an admirable job of wrangling all of Android's weirdness onto the Fold's screen, and in general, it's not that the Fold doesn't work; it's that there's nothing about the Fold that is dramatically better than the phone or tablet you might already be carrying around. And shoving them into a single device actually makes them both a little worse.

The US Federal Election Commission approved a Google plan to let campaign emails bypass Gmail spam filters. From a report: The FEC's advisory opinion adopted in a 4-1 vote said Gmail's pilot program is permissible under the Federal Election Campaign Act and FEC regulations "and would not result in the making of a prohibited in-kind contribution." The FEC said Google's approved plan is for "a pilot program to test new Gmail design features at no cost on a nonpartisan basis to authorized candidate committees, political party committees, and leadership PACs." On July 1, Google asked the FEC for the green light to implement the pilot after Republicans accused the company of giving Democrats an advantage in its algorithms. Republicans reportedly could have avoided some of their Gmail spam problems by using the proper email configuration. At a May 2022 meeting between Senate Republicans and Google's chief legal officer, "the most forceful rebuke" was said to come "from Sen. Marco Rubio (R-Fla.), who claimed that not a single email from one of his addresses was reaching inboxes," The Washington Post reported in late July. "The reason, it was later determined, was that a vendor had not enabled an authentication tool that keeps messages from being marked as spam, according to people briefed on the discussions."

Software sold by market leaders tend to be primary purchases for regular consumers. Brand comfort is important but so too is affordability, especially when pirate copies are available for free. Some find a middle ground with purchases of discounted activation keys but, as a new Microsoft lawsuit shows, that can amount to copyright infringement for buyers and sellers alike. From a report: In a complaint filed at a Washington court this week, Microsoft targets Canadian company The Search People Enterprises Ltd (TSPE), assumed director Mehtabjit Singh, and 'John Doe' defendants 1-10. The defendants are described as prolific distributors of "black market access devices," aka activation keys and tokens for Microsoft software. Those who bought keys and tokens may have been under the impression that they were purchasing official software but as Microsoft explains, that is not only misleading but a mischaracterization of the things they were sold.

Products including Microsoft Office, Project, Visio, Windows 10, and Windows 11 are all subject to licensing terms that restrict how the products can be used. Microsoft can also provide a product activation key to be entered as part of the installation process, with data about the activation sent to the company's servers. Like software tokens, which enable downloads and automatic software activation, activation keys are anti-piracy tools, and exchanging money for them is not the same as buying a license. Indeed, Microsoft makes itself very clear -- the activation of a piece of software means nothing in the absence of a license. Microsoft's problem is that product activation keys can be 'decoupled' from the software they were meant to authorize and then reused to activate more copies of the software, in some cases more copies than the attached Microsoft license permits.

Australia's consumer watchdog agency said Friday that Google has been ordered to pay AU$60 million, nearly $43 million, by the Federal Court over the collection of location data on Android phones. From a report: The fine stems from legal action the Australian Competition & Consumer Commission initiated back in 2019. The ACCC accused Google of "making misleading representations to consumers" about the collection and use of personal location data on Android devices between January 2017 and December 2018. The court previously found that Google misled consumers into thinking "Location History" was the only setting that impacted whether the tech giant collected, kept and used location data, when in fact, the "Web & App Activity" setting also allowed Google to collect some of this information, according to the ACCC. An estimated 1.3 million people with Google accounts in Australia may have "viewed a screen found by the Court to have breached the Australian Consumer Law," the ACCC said.

An anonymous reader shares a report: The earliest sales pitch for Stadia, the Google streaming service that beams high-end video games to web browsers via the cloud, included the idea that it might work as simply as Googling your favorite game. You might search for a popular game to learn more about it, only to immediately see an option to start playing it inside your web browser, no additional hardware required -- and perhaps no payment, either. Nearly three years after Stadia's official launch -- and 18 months after the service's massive internal downgrade -- that scenario has finally begun to play out. What's more, the feature appears to be streamer-agnostic, as multiple Stadia-like streaming services have started appearing in search results.

This week, Google rolled out a limited launch of a "Play Now" tab that appears on searches for select video games on desktop browsers. (As of press time, out of three Google accounts tested, the search results shown in this article only appear on one of them.) This tab can be found in the right-hand "knowledge panel" that is otherwise automatically populated with user reviews, game details, and digital download purchase links. When a Google game search returns a Play Now tab, it will include as many compatible streaming services as possible, including Google Stadia, Microsoft Xbox Cloud Gaming, Amazon Luna, and Nvidia GeForce Now. Each entry shows what kind of fee may be required to play the game. Conveniently enough, many Google Stadia games can now be played for free for the first 30 minutes or as long as 120 minutes, and supported free-trial Stadia games get a bright-green flash of italicized text: "trial available." Other games and services that have appeared thus far have included tags like "premium subscription" or "free-to-play."

An ongoing dispute over privacy between Apple and Facebook is roiling the digital economy, leading companies to shift billions in ad spending as users continue to limit the data available to advertisers. The feud took off last year, when Apple rolled out iOS 14.5, a version of its mobile operating system that made it easier than ever for iPhone and iPad users to opt out of letting apps like Facebook track their activity on their devices. The two companies weren't always at odds. In fact, they were almost business partners. From a report: In the years before the change, Apple suggested a series of possible arrangements that would earn the iPhone maker a slice of Facebook's revenue, according to people who either participated in the meetings or were briefed about them. As one person recalled: Apple officials said they wanted to "build businesses together." One idea that was discussed: creating a subscription-based version of Facebook that would be free of ads, according to people familiar with the discussions. Because Apple collects a cut of subscription revenue for apps in its App Store, that product could have generated significant revenue for the Cupertino, Calif., giant.

The companies also haggled over whether Apple was entitled to a piece of Facebook's sales from so-called boosted posts, said people familiar with the matter. A boost allows a user to pay to increase the number of people that see a post on Facebook or Instagram. Facebook, which considers boosts ads, has always contended that boosts are a form of advertising, in part because they are often used by small businesses to reach a bigger audience, said one of the people. Apple, which doesn't take a cut of advertising from developers, argued that Facebook boosts should be considered in-app purchases, according to a person familiar with the matter. Apple's standard terms would entitle it to take a 30% share of those sales.

Meta, the owner of Facebook and Instagram, has been rewriting websites its users visit, letting the company follow them across the web after they click links in its apps, according to new research from an ex-Google engineer. The Guardian reports: The two apps have been taking advantage of the fact that users who click on links are taken to webpages in an "in-app browser," controlled by Facebook or Instagram, rather than sent to the user's web browser of choice, such as Safari or Firefox. "The Instagram app injects their tracking code into every website shown, including when clicking on ads, enabling them [to] monitor all user interactions, like every button and link tapped, text selections, screenshots, as well as any form inputs, like passwords, addresses and credit card numbers," says Felix Krause, a privacy researcher who founded an app development tool acquired by Google in 2017.

Krause discovered the code injection by building a tool that could list all the extra commands added to a website by the browser. For normal browsers, and most apps, the tool detects no changes, but for Facebook and Instagram it finds up to 18 lines of code added by the app. Those lines of code appear to scan for a particular cross-platform tracking kit and, if not installed, instead call the Meta Pixel, a tracking tool that allows the company to follow a user around the web and build an accurate profile of their interests. The company does not disclose to the user that it is rewriting webpages in this way. No such code is added to the in-app browser of WhatsApp, according to Krause's research. [...] It is unclear when Facebook began injecting code to track users after clicking links. "We intentionally developed this code to honor people's [Ask to track] choices on our platforms," a Meta spokesperson told The Guardian in a statement. "The code allows us to aggregate user data before using it for targeted advertising or measurement purposes. We do not add any pixels. Code is injected so that we can aggregate conversion events from pixels."

They added: "For purchases made through the in-app browser, we seek user consent to save payment information for the purposes of autofill."

An anonymous reader quotes a report from Motherboard: A group of security researchers found a series of vulnerabilities in the software underlying popular apps like Discord, Microsoft Teams, Spotify and many others, which are used by tens of millions of people all over the world. At the Black Hat cybersecurity conference in Las Vegas on Thursday, the researchers presented their findings, detailing how they could have hacked people who use Discord, Microsoft Teams, and the chat app Element by exploiting the software underlying all of them: Electron, which is a framework built on the open source Chromium and the cross-platform javascript environment Node JS. In all these cases, the researchers submitted vulnerabilities to Electron to get them fixed, which earned them more than $10,000 in rewards. The bugs were fixed before the researchers published their research.

Aaditya Purani, one of the researchers who found these vulnerabilities, said that "regular users should know that the Electron apps are not the same as their day-to-day browsers," meaning they are potentially more vulnerable. In the case of Discord, the bug Purani and his colleagues found only required them to send a malicious link to a video. With Microsoft Teams, the bug they found could be exploited by inviting a victim to a meeting. In both cases, if the targets clicked on these links, hackers would have been able to take control of their computers, Purani explained in the talk. For him, one of the main takeaways of their research is that Electron is risky precisely because users are very likely to click on links shared in Discord or Microsoft Teams.

Epson has gained some scrutiny in recent weeks after the company disabled a printer that was otherwise working fine, leading to accusations of planned obsolescence. Epson knows its printers will stop working without simple maintenance at a predictable point in the future, and it knows that it won't be cost-effective for many owners to send their home printers in for service. So why not build them to be user serviceable in the first place? The Verge: The inciting post from @marktavern mentions that his wife was unable to use her "very expensive Epson printer" after an end-of-service error message appeared. This isn't anything new for Epson printers, sadly. Reports going back several years mention an infamous error message that reads "parts inside the printer have reached the end of their service life." Epson confirmed to The Verge that the error is related to the printer's ink pads, which had likely become saturated through extended use and were now at risk of spilling into the rest of the printer mechanism.

In a recently updated support document, Epson offers several solutions to resolve the problem. These include sending the printer into Epson to replace the ink pads or having a local certified technician do it. Previously (via Wayback Machine), just before the issue gained notoriety, Epson conceded that "repair may not be a good investment for lower cost printers because the printer's other components also may be near the end of usable life." It then added that "most consumers who are out of warranty elect to replace a lower-cost printer when they receive an end of life service message." Now, Epson suggests the feel-good option of sending the bricked unit in for recycling.

Intel has released 48 benchmarks that show its upcoming Arc A750 GPU should be able to trade blows with Nvidia's RTX 3060 running modern games. From a report: While Intel set its expectations low for its Arc GPUs last month, the company has now tested its A750 directly against the RTX 3060 across 42 DirectX 12 titles and six Vulkan games. The results look promising for what will likely be Intel's mainstream GPU later this year. Intel has tested the A750 against popular games like Fortnite, Control, and Call of Duty: Warzone, instead of the cherry picked handful of benchmarks the company released last month. "These are all titles that we picked because they're popular," explains Intel fellow Tom Petersen, in Intel's benchmark video. "Either reviewers are using them or they're high on the Steam survey, or new and exciting. These are not cherry picked titles."

We'll have to wait for independent benchmarks, but based on Intel's testing, the A750 looks like it will compete comfortably with Nvidia's RTX 3060. "You'll see we're kinda trading blows with the RTX 3060," says Petersen. "Sometimes we win, sometimes we lose." Intel's performance is, on average, 3 to 5 percent better than Nvidia's when it wins on titles running at 1080p. Over on the 1440p side, it looks like Intel wins on more of the benchmarks. On average it's a win of about 5 percent across the 42 games. Intel has also tested six Vulkan titles, where it seems be trading blows with the RTX 3060 once again.

An anonymous reader shares a report: Technology to build cleaner, more efficient air conditioners does exist. Two major AC manufacturers, Daikin and Gree Electric Appliances, shared the top award at last year's Global Cooling Prize, an international competition focused on designing climate-friendly AC tech. Both companies created ACs with higher internal performance that used less environmentally damaging refrigerants; the new units could reduce their impact on the climate by five times. [...] Another strategy is to double down on heat pumps, which are air conditioners that also work in reverse, using vapor compression to absorb and move heat into a home, instead of releasing it outside. Heat pumps usually cost several thousand dollars, though the Inflation Reduction Act includes a proposal for a significant heat pump rebate, and President Joe Biden has invoked the Defense Production Act to ramp up production.

Experts have argued installing heat pumps is critical to another important climate goal: transitioning away from fossil fuel-powered furnaces, which are an even bigger source of emissions than cooling. The holy grail of HVAC would be a heat pump that could provide both heating and cooling but isn't dependent on vapor compression. [...] Another challenge, though, is that heat pumps are not the easiest appliance to install, especially for renters, who don't necessarily have the money or ability to invest in bulky HVAC systems. To address this problem, a company called Gradient has designed a heat pump that easily slides over a windowsill -- it doesn't block light -- and currently uses a refrigerant called R32, which is supposed to have a (comparatively) low global warming potential. Gradient recently won a contract to install its units in New York City public housing.

The most ambitious upgrade to the Ethereum blockchain should take place in September, possibly closer to the middle of the month, developers working on the project said during a conference call after what was billed as a final dress rehearsal. From a report: Developers have picked a number of so-called total terminal difficulty required of the final block mined in Ethereum before the network switches to new software. Figuring out the exact date range when the upgrade will occur will require complex calculations, and will be a moving target, depending on changes to the network's use and support, developers said on the call that was broadcast over YouTube on Thursday. The final date range is expected to be approved during another developer call next week, though the software engineers are currently looking at Sept. 16 to Sept. 20.

Called the Merge, the software upgrade has been in the works for years, and it will change the way Ethereum orders transactions to become more energy efficient. Instead of using energy-guzzling computers called miners, the network will deploy so-called validators using staked Ether tokens -- a setup called proof of stake. Following years of delays, the time for the Merge is finally being set after Wednesday's completion of the Goerli merge test, which simulated the Merge on a smaller scale. A few problems popped up during the test, developers reported on the call. Goerli merge, which many celebrated with parties broadcast on YouTube, was the final test before the actual Merge was to take place.

Mailchimp appears to have suspended the accounts of several crypto-related firms, according to the affected outlets. Crypto firms on the chopping board include intelligence platform Messari. From a report: Founder Ryan Selkis posted on Twitter revealing the suspension and expressing his disappointment. Crypto wallet provider Edge, NFT artist Ocarina, and Jesse Friedland -- the founder of NFT collection Cryptoon Goonz -- are among prominent names that appear to have had their accounts suspended in the last several weeks, according to the Decrypt report.

The Federal Trade Commission is seeking public feedback on a proposed rulemaking to limit what it's dubbed "commerical surveillance" by businesses that sell or share information collected about people. From a report: The advance notice, announced Thursday, would protect the personal data companies such as Alphabet Inc. and Meta Platforms Inc. collect about consumers.

Google will stop giving snappy answers to stupid questions, the company has announced, as it seeks to improve its search engine's "featured snippets" service. From a report: That means users should see fewer answers to questions such as "When did Snoopy assassinate Abraham Lincoln?", to which the service would once merrily respond with "1865" -- the right date, but very much the wrong assassin. "This clearly isn't the most helpful way to display this result," said the company's head of search, Pandu Nayak, in a blogpost announcing the changes. "We've trained our systems to get better at detecting these sorts of false premises, which are not very common, but there are cases where it's not helpful to show a featured snippet. We've reduced the triggering of featured snippets in these cases by 40% with this update."

Snippets, which sometimes show up as a featured response to direct questions asked of Google Search, have long been a cornerstone of the company's AI strategy. The same technology powers its smart speakers and voice assistants, and lets the search engine satisfy search queries without visitors clicking away to other websites. But the snippets, which are automatically generated from the contents of websites, have also been a thorn in Google's side for just as long. [...] In an effort to address the root cause of such mistakes, Google is also rolling out new warnings for times when a search term has hit a "data void" -- a question where a good answer might simply not exist.

Researchers from Stanford University and Nvidia have teamed up to help develop VR glasses that look a lot more like regular spectacles. PC Gamer reports: "A major barrier to widespread adoption of VR technology, however, is the bulky form factor of existing VR displays and the discomfort associated with that," the research paper published at Siggraph 2022 says. These aptly named "Holographic Glasses" can deliver a full-colour 3D holographic image using optics that are only 2.5mm thick. Compared to the traditional way a VR headset works, in which a lens magnifies a smaller display some distance away from it, shrinking all the prerequisite parts down to such a small size is quite the spectacular step forward for VR.

The Holographic Glasses prototype uses pancake lenses, which is a concept that has been thrown around a couple of times in the past few years. These pancake lenses not only allow for a much smaller profile but reportedly they have a few other benefits, too: the resolution they can offer is said to be unlimited, meaning you can crank up the resolution for VR headsets, and they offer a much wider field of view at up to 200 degrees. [...] The research paper lists the glasses as such: "a coherent light source that is coupled into a pupil-replicating waveguide, which provides the illumination for a phase-only SLM that is mounted on the waveguide in front of the user's eye. This SLM creates a small image behind the device, which is magnified by a thin geometric phase (GP) lens."

Though, it's very much a promise of what's to come more than an immediately shippable product today. There are some limitations: while there's scope to have a much higher FOV than current generation VR headsets, this particular wearable prototype only offered an FOV of 22.8 degrees. The benchtop prototype offered even less, at only 16.1 degrees. "[The FOV] is far smaller than commercially available VR/AR displays. However, the FOV was mainly limited by the size of the available SLM and the focal length of the GP lens, both of which could be improved with different components," the researchers say. Another limitation is the likely requirement for a very accurate measurement of the user's pupil, which won't be easy without a well-thought-out design. It would be possible to use an infrared gaze tracker to do this, the researchers note, but you'd need to be able to track the wearer's pupil size constantly as they will adjust often to different light conditions while using the glasses.

The FCC is canceling $886 million in funding for Starlink to expand access in rural areas, citing the satellite internet system's cost and doubts over whether it can supply fast enough speeds. PC Magazine reports: The agency today announced it had rejected "long-form applications" from both SpaceX and an ISP called LTD Broadband to secure funding from the FCC's Rural Digital Opportunity Fund. "The Commission determined that these applications failed to demonstrate that the providers could deliver the promised service," the FCC said in a statement. FCC Chairwoman Jessica Rosenworcel added: "We cannot afford to subsidize ventures that are not delivering the promised speeds or are not likely to meet program requirements."

In December 2020, the FCC awarded $886 million to SpaceX to help its Starlink service supply high-speed broadband to 642,925 locations in 35 states. However, it came with a requirement that SpaceX provide a long-form application about how Starlink would meet its obligations before the federal funding could be fully secured. The FCC's goal with the Rural Digital Opportunity Fund is to supply gigabit internet speeds to over 85% of the selected rural locations and at least 100Mbps download speeds for all 99.7% of the locations in the coming years. "Starlink's technology has real promise," Rosenworcel said. "But the question before us was whether to publicly subsidize its still developing technology for consumer broadband -- which requires that users purchase a $600 dish -- with nearly $900 million in universal service funds until 2032."