Linux fréttir

Zorro shares a report from The Mercury News: Not only did Facebook inflate ad-watching metrics by up to 900 percent (Warning: source may be paywalled, alternative source), it knew for more than a year that its average-viewership estimates were wrong and kept quiet about it, a new legal filing claims. A group of small advertisers suing the Menlo Park social media titan alleged in the filing that Facebook "induced" advertisers to buy video ads on its platform because advertisers believed Facebook users were watching video ads for longer than they actually were. That "unethical, unscrupulous" behavior by Facebook constituted fraud because it was "likely to deceive" advertisers, the filing alleged. The latest allegations arose out of a lawsuit that the advertisers filed against Mark Zuckerberg-led Facebook in federal court in 2016 over alleged inflation of ad-watching metrics. "Suggestions that we in any way tried to hide this issue from our partners are false," the company told The Wall Street Journal. "We told our customers about the error when we discovered it -- and updated our help center to explain the issue." "The plaintiffs are seeking class-action status to bring other advertisers into the legal action, plus unspecified damages," reports The Mercury News. "They also want the court to order a third-party audit of Facebook's video-ad metrics."

Read more of this story at Slashdot.

'No evidence' vulnerability was abused, though, we're told

Tumblr today reveal it has fixed a security bug in its website that quietly revealed private details of some of its bloggers.…

An anonymous reader shares a report: Amazon cracked down on fake reviews two years ago by prohibiting shoppers from getting free products directly from merchants in exchange for writing reviews. It was a major turning point for the world's largest online retailer, which had previously seen "incentivized reviews" as a key way for consumers to discover new products. Amazon changed course because it realized some merchants were using such reviews to game its search algorithm, undermining faith in the customer feedback that helps drive e-commerce. Amazon instead used its "Vine" program, in which Amazon serves as a middleman between prolific Amazon reviewers and vendors eager for exposure. Amazon would still allow freebies in exchange for feedback so long as there was no direct contact between its retail partners and reviewers, theoretically lessening the chance of quid-pro-quo. Amazon would select shoppers eligible for the program, and Amazon vendors would pay a fee and provide free products to participate. But there was an important group excluded from the Vine program: independent merchants who supply about half the goods sold on the site. Now those excluded merchants and review watchdogs are alleging Amazon is guilty of the review manipulation the company said it was trying to prevent. Amazon uses Vine extensively to promote a fast-growing assortment of its own private-label products, distributing free samples to quickly accumulate the reviews needed to rise in search results and boost shopper faith in making a purchase. It gives Amazon a big advantage when introducing its own brands over third-party merchants who are more vulnerable to Amazon's private-label competition than prominent brands already in stores.

Read more of this story at Slashdot.

Code-hosting biz tries out a more automated delivery process

Analysis At its annual developer conference on Tuesday, GitHub unveiled a way to automate software deployment workflows called Actions. It sounds rather underwhelming, given all the different automation tools available, but the executives discussing it brimmed with Apple-level enthusiasm. Sam Lambert, head of platform for GitHub, called it "truly remarkable."…

An anonymous reader quotes a report from Bloomberg: Apple on Wednesday began allowing users in the U.S. to download a copy of all of the data that they have stored with the company from a single online portal. U.S. users will be able to download data such as all of their address book contacts, calendar appointments, music streaming preferences and details about past Apple product repairs. Previously, customers could get their data by contacting Apple directly. In May, when Apple first launched the online privacy portal, it only allowed U.S. users to either correct their data or delete their Apple accounts.

Read more of this story at Slashdot.

People's Facebook posts might predict whether they are suffering from depression, researchers reported this week. From a report: The researchers found that the words people used seemed to indicate whether they would later be diagnosed with depression. The findings offer a way to flag people who may be in need of help, but they also raise important questions about people's health privacy, the team reported in the Proceedings of the National Academy of Sciences. People who were later clinically diagnosed with depression used more "I" language, according to Johannes Eichstaedt of the University of Pennsylvania and his colleagues. They also used more words reflecting loneliness, sadness and hostility. "We observed that users who ultimately had a diagnosis of depression used more first-person singular pronouns, suggesting a preoccupation with the self," they wrote. That is an indicator of depression in some people. The team recruited 683 people who visited an emergency room for their study and asked to see their Facebook pages. Most were not depressed, but 114 had a depression diagnosis in their medical records.

Read more of this story at Slashdot.

Leaky AWS S3 bucket fingered by infosec bods

Now, now, America. Don't go overboard. Again.…

In the early days of what ultimately became Waymo, Google's self-driving car division (known at the time as "Project Chauffeur"), there were "more than a dozen accidents, at least three of which were serious," according to a new article in The New Yorker . From a report: The magazine profiled Anthony Levandowski, the former Google engineer who was at the center of the Waymo v. Uber trade secrets lawsuit. According to the article, back in 2011, Levandowski also modified the autonomous software to take the prototype Priuses on "otherwise forbidden routes." Citing an anonymous source, The New Yorker reports that Levandowski sat behind the wheel as the safety driver, along with Isaac Taylor, a Google executive. But while they were in the car, the Prius "accidentally boxed in another vehicle," a Camry. As The New Yorker wrote: "A human driver could easily have handled the situation by slowing down and letting the Camry merge into traffic, but Google's software wasn't prepared for this scenario. The cars continued speeding down the freeway side by side. The Camry's driver jerked his car onto the right shoulder. Then, apparently trying to avoid a guard rail, he veered to the left; the Camry pinwheeled across the freeway and into the median. Levandowski, who was acting as the safety driver, swerved hard to avoid colliding with the Camry, causing Taylor to injure his spine so severely that he eventually required multiple surgeries." This was apparently just one of several accidents in Project Chauffeur's early days.

Read more of this story at Slashdot.

Millions of dollars in funding and billions of dollars in valuations have made scooters the next big thing since the last big thing. From a report: When Michael Ramsey, an analyst for technology research firm Gartner, started in February to put together his 2018 "hype cycle" report for the future of transportation, he had plenty of topics to choose from: electric vehicles, flying cars, 5G, blockchain, and, of course, autonomous vehicles. But one type of transportation is conspicuously absent from the results of the report: electric scooters. "At the time, outside of California, these scooters were really not that common," Ramsey said. "That's how much has happened." As for autonomous vehicles, which have enjoyed years of hype as the next big thing, Ramsey labeled them sliding into "the trough of disillusionment," which Ramsey described as "when expectations don't meet the truth." In a matter of months, electric scooter startups have gone from tech oddity to global phenomenon. In some cities, hundreds of scooters suddenly showed up on streets from companies including Bird and Lime, leaving municipalities to figure out how to handle the sudden influx of two-wheeled travelers. The concept behind the scooters is simple: A user can grab any available scooter, unlock it with an app, ride to their destination, and leave the scooter there for someone else to use. Even by the hyper-growth expectations of Silicon Valley, the rise of scooter companies has been dizzying. Scooters can be found in more than 125 cities in the U.S. and more than 10 across the globe. In the year after their launch, both Lime and Bird said their scooters had been used for more than 10 million rides.

Read more of this story at Slashdot.

Grubby Grubbs' grifting days are gone

A programmer who wrote and sold software that backdoored PCs so they could be remotely controlled has been jailed for 30 months – and forced to give up his stash of 114 Bitcoins.…

A security researcher from Colombia has found a way of gaining admin rights and boot persistence on Windows PCs that's simple to execute and hard to stop -- all the features that hackers and malware authors are looking for from an exploitation technique. From a report: What's more surprising, is that the technique was first detailed way back in December 2017, but despite its numerous benefits and ease of exploitation, it has not received either media coverage nor has it been seen employed in malware campaigns. Discovered by Sebastian Castro, a security researcher for CSL, the technique targets one of the parameters of Windows user accounts known as the Relative Identifier (RID). The RID is a code added at the end of account security identifiers (SIDs) that describes that user's permissions group. There are several RIDs available, but the most common ones are 501 for the standard guest account, and 500 for admin accounts. Castro, with help from CSL CEO Pedro Garcia, discovered that by tinkering with registry keys that store information about each Windows account, he could modify the RID associated with a specific account and grant it a different RID, for another account group. The technique does not allow a hacker to remotely infect a computer unless that computer has been foolishly left exposed on the Internet without a password. But in cases where a hacker has a foothold on a system -- via either malware or by brute-forcing an account with a weak password -- the hacker can give admin permissions to a compromised low-level account, and gain a permanent backdoor with full SYSTEM access on a Windows PC.

Read more of this story at Slashdot.

So. Much. Glue.

Torx twirlers iFixit celebrated the release of Microsoft's Surface Pro 6 by ripping the thing apart only to find that its still pretty much unrepairable.…

Andrew Liszewski, writing for io9: Casting anyone other than Harrison Ford in the role of Han Solo just feels like sacrilege, but since Ford is now 76 years old, playing a younger version of himself would be all but impossible. Or at least impossible if you rely on the standard Hollywood de-aging tricks like makeup and CG. Artificial intelligence, it turns out, does a pretty amazing job at putting Ford back into the role of Solo. The YouTube channel "derpfakes" has been posting videos that demonstrate the impressive, and at times frightening, capabilities of image processing using artificial intelligence. Using a process called deep learning, an AI analyzes a large collection of photos of a given person, creating a comprehensive database of them in any almost any position and pose. It then uses that database to intelligently perform an automatic face replacement on a source clip, in this case replacing actor Alden Ehrenreich's face with Harrison Ford's.

Read more of this story at Slashdot.

They don't *seem* so different if you check digits (though your real world results may vary)

Micron's consumer SSD brand Crucial has pushed out an NVMe minicard flash drive that looks almost identical to Intel's 600p.…

An anonymous reader shares a report: Sellers have been modifying lower end NVIDIA graphics cards and selling them more powerful cards online. In a recent version of the GPU-Z graphics card information utility, TechPowerUp has added the ability to now detect these fake NVIDIA cards. This new feature allows buyers of cards to detect if the card is actually a relabled NVIDIA G84, G86, G92, G94, G96, GT215, GT216, GT218, GF108, GF106, GF114, GF116, GF119, or GK106 GPU by displaying an exclamation point where the NVIDIA logo would normally appear and also prepends the string "[FAKE]" before the card's name.

Read more of this story at Slashdot.

Cloud comms upstart touts more authentic engagement through artificial intelligence

At its Signal show in San Francisco, California, today, cloud comms biz Twilio intends to debut a chatbot platform called Autopilot for creating chatty software agents suitable for integration with a variety of services, including interactive voice response (IVR) systems, SMS, chat, Alexa, Slack, and Google Assistant.…

A Seattle-area startup is aiming to take on giants such as Google and change the way we do email with a new physical personal email server. From a report: Helm today unveiled its $499 device that lets consumers send and receive email from their own domain, in addition to saving contacts and calendar events. It's a bold bet that aims to provide comfort at a time when privacy and security issues related to personal data hosted by big tech companies in the cloud are top of mind. The idea comes from Giri Sreenivas and Dirk Sigurdson, two entrepreneurs who already sold a security startup and raised a $4 million seed round from top venture capital firms last year. The device is about the size of a router and looks like an upside-down book placed on a table. It connects to a home network and pairs with a mobile app that lets users create their own domain name, passwords, and recovery keys. Helm support standard protocols and works with regular email clients such as Outlook or the Mail app, with encryption protecting connection between the device and the apps.

Read more of this story at Slashdot.

Laboratories frequently "reclassify" genetic mutations. But there is no reliable system for telling patients or doctors that the results of their genetic tests are no longer valid. From a report: The results of a genetic test may seem final -- after all, a gene mutation is present or it is not. That mutation increases the risk of a disease, or it does not. In fact, those findings are not as straightforward as they might seem, and the consequences may have grave implications for patients. While a person's genome doesn't change, the research linking particular bits of DNA to disease is very much in flux. Geneticists and testing labs constantly receive new information that leads them to reassess genetic mutations. As a result, a mutation seen as benign today may be found dangerous tomorrow. And vice versa. But there is no good way to get the new information to doctors and patients. The result: The gene test you had a few years ago might yield a startlingly different result now. The problem affects a minority of patients, mostly people with unusual mutations. The more common disease-causing mutations -- like those that predispose you to breast or colon cancer -- are so well studied that their meaning is not in doubt. In a recent study, researchers at Myriad Genetics, a diagnostic company, reviewed data on 1.45 million patients who had genetic tests from 2006 to 2016 to see if the results originally reported still held true. The lab issued new reports for nearly 60,000 of them, meaning the old results had been superseded by new data. But many patients who carry mutations that have been reclassified remain in the dark.

Read more of this story at Slashdot.

But that's not as bad as Barbie flinger Mattel or, indeed, Oracle

It has been a bumper year for Microsoft CEO Satya Nadella, whose compensation soared to $25.8m in fiscal 2018. Quite a bit more than the median salary of the rest of the company, which hovered at an admittedly none-too-shabby $167,689.…

Not quite the native experience Google's shooting for

Version 70 of the Chrome browser has begun to slither onto Windows 10, bringing with it Google's desktop take on Progressive Web App functionality.…