Slashdot.org

Syndicate content Slashdot
News for nerds, stuff that matters
Updated: 51 min 14 sec ago

Cerebras Systems Unveils a Record 1.2 Trillion Transistor Chip For AI

Mon, 08/19/2019 - 16:20
An anonymous reader quotes a report from VentureBeat: New artificial intelligence company Cerebras Systems is unveiling the largest semiconductor chip ever built. The Cerebras Wafer Scale Engine has 1.2 trillion transistors, the basic on-off electronic switches that are the building blocks of silicon chips. Intel's first 4004 processor in 1971 had 2,300 transistors, and a recent Advanced Micro Devices processor has 32 billion transistors. Samsung has actually built a flash memory chip, the eUFS, with 2 trillion transistors. But the Cerebras chip is built for processing, and it boasts 400,000 cores on 42,225 square millimeters. It is 56.7 times larger than the largest Nvidia graphics processing unit, which measures 815 square millimeters and 21.1 billion transistors. The WSE also contains 3,000 times more high-speed, on-chip memory and has 10,000 times more memory bandwidth.

Read more of this story at Slashdot.

Newt Gingrich Trying To Sell Trump on a Cheap Moon Plan

Mon, 08/19/2019 - 15:45
WindBourne writes: Newt Gingrich and an eclectic band of NASA skeptics are trying to sell President Donald Trump on a reality show-style plan to jump-start the return of humans to the moon -- at a fraction of the space agency's estimated price tag. The proposal, whose other proponents range from an Air Force lieutenant general to the former publicist for pop stars Michael Jackson and Prince, includes a $2 billion sweepstakes pitting billionaires Elon Musk, Jeff Bezos and other space pioneers against each other to see who can establish and run the first lunar base, according to a summary of the plan shared with POLITICO. That's far less taxpayer money than NASA's anticipated lunar plan, which relies on traditional space contractors, such as Boeing and Lockheed Martin, and is projected to cost $50 billion or more. Backers of the novel approach have briefed administration officials serving on the National Space Council, several members of the group confirmed, though they declined to provide specifics of the internal conversations.

Read more of this story at Slashdot.

Bernie Sanders Wants To Ban Facial Recognition Use By Police

Mon, 08/19/2019 - 15:04
Democratic presidential candidate Senator Bernie Sanders (I-VT) wants to put an end to police use of facial recognition software. Sanders called for the ban as part of a criminal justice reform plan introduced over the weekend ahead of a two-day tour of South Carolina. From a report: The plan also calls for the ban of for-profit prisons and would revoke the practice of law enforcement agencies benefiting from civil asset forfeitures. Sanders kicked off his campaign by saying "I'm running for president because we need to understand that artificial intelligence and robotics must benefit the needs of workers, not just corporate America and those who own that technology."

Read more of this story at Slashdot.

Hacker Releases First Public Jailbreak for Up-to-Date iPhones in Years

Mon, 08/19/2019 - 14:21
Apple has mistakenly made it a bit easier to hack iPhone users who are on the latest version of its mobile operating system iOS by unpatching a vulnerability it had already fixed. From a report: Hackers quickly jumped on this over the weekend, and publicly released a jailbreak for current, up-to-date iPhones -- the first free public jailbreak for a fully updated iPhone that's been released in years. Security researchers found this weekend that iOS 12.4, the latest version released in June, reintroduced a bug found by a Google hacker that was fixed in iOS 12.3. That means it's currently relatively easy to not only jailbreak up to date iPhones, but also hack iPhone users, according to people who have studied the issue. "Due to 12.4 being the latest version of iOS currently available and the only one which Apple allows upgrading to, for the next couple of days (till 12.4.1 comes out), all devices of this version (or any 11.x and 12.x below 12.3) are jail breakable -- which means they are also vulnerable to what is effectively a 100+ day exploit," said Jonathan Levin, a security researcher and trainer who specializes in iOS, referring to the fact that this vulnerability can be exploited with code that was found more than 100 days ago. Pwn20wnd, a security researcher who develops iPhone jailbreaks, published a jailbreak for iOS 12.4 on Monday.

Read more of this story at Slashdot.

An Ode To Microsoft Encarta

Mon, 08/19/2019 - 13:41
Scott Hanselman: Microsoft Encarta came out in 1993 and was one of the first CD-ROMs I had. It stopped shipping in 2009 on DVD. I recently found a disk and was impressed that it installed just perfectly on my latest Window 10 machine and runs nicely. Encarta existed in an interesting place between the rise of the internet and computer's ability to deal with (at the time) massive amounts of data. CD-ROMs could bring us 700 MEGABYTES which was unbelievable when compared to the 1.44MB (or even 120KB) floppy disks we were used to. The idea that Encarta was so large that it was 5 CD-ROMs (!) was staggering, even though that's just a few gigs today. Even a $5 USB stick could hold Encarta - twice! My kids can't possibly intellectualize the scale that data exists in today. We could barely believe that a whole bookshelf of Encyclopedias was now in our pockets. I spent hours and hours just wandering around random articles in Encarta. The scope of knowledge was overwhelming, but accessible. But it was contained - it was bounded. Today, my kids just assume that the sum of all human knowledge is available with a single search or a "hey Alexa" so the world's mysteries are less mysteries and they become bored by the Paradox of Choice. In a world of 4k streaming video, global wireless, and high-speed everything, there's really no analog to the feeling we got watching the Moon Landing as a video in Encarta - short of watching it live on TV in the 1969! For most of us, this was the first time we'd ever seen full-motion video on-demand on a computer in any sort of fidelity - and these are mostly 320x240 or smaller videos!

Read more of this story at Slashdot.

Developers Accuse Apple of Anti-Competitive Behavior With Its Privacy Changes in iOS 13

Mon, 08/19/2019 - 13:02
A group of app developers have penned a letter to Apple CEO Tim Cook, arguing that certain privacy-focused changes to Apple's iOS 13 operating system will hurt their business. From a report: In a report by The Information, the developers were said to have accused Apple of anti-competitive behavior when it comes to how apps can access user location data. With iOS 13, Apple aims to curtail apps' abuse of its location-tracking features as part of its larger privacy focus as a company. Today, many apps ask users upon first launch to give their app the "Always Allow" location-tracking permission. Users can confirm this with a tap, unwittingly giving apps far more access to their location data than is actually necessary, in many cases. In iOS 13, however, Apple has tweaked the way apps can request location data. There will now be a new option upon launch presented to users, "Allow Once," which allows users to first explore the app to see if it fits their needs before granting the app developer the ability to continually access location data. This option will be presented alongside existing options, "Allow While Using App" and "Don't Allow." The "Always" option is still available, but users will have to head to iOS Settings to manually enable it. The app developers argue that this change may confuse less technical users, who will assume the app isn't functioning properly unless they figure out how to change their iOS Settings to ensure the app has the proper permissions.

Read more of this story at Slashdot.

Twitter is Blocked in China, But Chinese State News Agency is Buying Promoted Tweets To Portray Hong Kong Protestors as Violent

Mon, 08/19/2019 - 12:21
Chinese state-run news agency Xinhua is promoting tweets attacking the protestors and claiming they do not have wider support. From a report: Twitter is being criticized for running promoted tweets by China's largest state news agency that paint pro-democracy demonstrations in Hong Kong as violent, even though the rallies, including one that drew an estimated 1.7 million people this weekend, have been described as mostly peaceful by international media. Promoted tweets from China Xinhua News, the official mouthpiece of the Chinese Communist Party, were spotted and shared by the Twitter account of Pinboard, the bookmarking service founded by Maciej Ceglowski, and other users. The demonstrations began in March to protest a now-suspended extradition bill, but have grown to encompass other demands, including the release of imprisoned protestors, inquiries into police conduct, the resignation of current Chief Executive of Hong Kong Carrie Lam and a more democratic process for electing Legislative Council members and the chief executive. UPDATE: Twitter is now blocking state-run media outlets from advertising on its platform.

Read more of this story at Slashdot.

Small Companies Play Big Role in Robocall Scourge, But Remedies Are Elusive

Mon, 08/19/2019 - 11:41
The billions of illegal robocalls inundating Americans are being facilitated largely by small telecom carriers that transmit calls over the internet, industry officials say, but authorities are at odds over what -- if anything -- they can do to stop them. From a report: These telecom carriers typically charge fractions of a cent per call, making their money on huge volume. Their outsize role in the robocall scourge has become apparent as large telecom companies get better at tracing robocalls to their source, spurring calls for regulators to hold them accountable. "There are definitely repeat offenders who keep showing up as the sources of illegal robocalls," said Patrick Halley, a senior vice president at USTelecom, a trade association of telecom companies that runs a robocall-tracing group. "Carriers that knowingly allow the origination of billions of illegal robocalls should be held accountable." U.S. regulators have conflicting interpretations of their ability to take the companies to court, however. And carriers aren't explicitly required to try to differentiate between legal and illegal robocalls, further clouding enforcement.

Read more of this story at Slashdot.

Wireless Carrier Throttling of Online Video Is Pervasive: Study

Mon, 08/19/2019 - 11:01
U.S. wireless carriers have long said they may slow video traffic on their networks to avoid congestion and bottlenecks. But new research shows the throttling happens pretty much everywhere all the time. From a report: Researchers from Northeastern University and University of Massachusetts Amherst conducted more than 650,000 tests in the U.S. and found that from early 2018 to early 2019, AT&T throttled Netflix 70% of the time and Google's YouTube service 74% of the time. But AT&T didn't slow down Amazon's Prime Video at all. T-Mobile throttled Amazon Prime Video in about 51% of the tests, but didn't throttle Skype and barely touched Vimeo, the researchers say in a paper [PDF] to be presented at an industry conference this week.

Read more of this story at Slashdot.

Degrading Tor Network Performance Only Costs a Few Thousand Dollars Per Month

Mon, 08/19/2019 - 10:21
Threat actors or nation-states looking into degrading the performance of the Tor anonymity network can do it on the cheap, for only a few thousands US dollars per month, new academic research has revealed. An anonymous reader writes: According to researchers from Georgetown University and the US Naval Research Laboratory, threat actors can use tools as banal as public DDoS stressers (booters) to slow down Tor network download speeds or hinder access to Tor's censorship circumvention capabilities. Academics said that while an attack against the entire Tor network would require immense DDoS resources (512.73 Gbit/s) and would cost around $7.2 million per month, there are far simpler and more targeted means for degrading Tor performance for all users. In research presented this week at the USENIX security conference, the research team showed the feasibility and effects of three types of carefully targeted "bandwidth DoS [denial of service] attacks" that can wreak havoc on Tor and its users. Researchers argue that while these attacks don't shut down or clog the Tor network entirely, they can be used to dissuade or drive users away from Tor due to prolongued poor performance, which can be an effective strategy in the long run.

Read more of this story at Slashdot.

The Latest Claim To Satoshi Nakamoto is the 'Stupidest One Yet'

Mon, 08/19/2019 - 09:40
An anonymous reader shares a report: For years, Faketoshis have been fighting to claim the Bitcoin throne, trying to make us all believe they were responsible for the cryptocurrency's creation. But things took a different turn this weekend after an unknown person(s) decided it was time to reveal their identity as the 'real' Satoshi Nakomoto in a three-part blog post series. Possibly exhausted by peoples' previous attempts to do the same, and having noticed several significant inconsistencies in the person's writing, it didn't take long for Bitcoin Twitter to react and call Faketoshi's claims into question. Further reading: How the NSA Identified Satoshi Nakamoto (2017); Bizarre New Theories Emerge About Bitcoin Creator Satoshi Nakamoto (2019); The CIA 'Can Neither Confirm Nor Deny' It Has Documents on Satoshi Nakamoto (2018); Craig Wright Claims He's Satoshi Nakamoto, the Creator Of Bitcoin (2016); Former Bitcoin Developer Shares Early Satoshi Nakamoto Emails (2017); He Says He Invented Bitcoin and Is Suing Those Who Doubt Him (2019); Elon Musk Says He Is Not Bitcoin's Satoshi Nakamoto (2017); Satoshi Nakamoto Found? Not So Fast (2014); Bitcoin Releases Version 0.3 (2010).

Read more of this story at Slashdot.

Fearing Data Privacy Issues, Google Cuts Some Android Phone Data For Wireless Carriers

Mon, 08/19/2019 - 09:00
Alphabet' Google has shut down a service it provided to wireless carriers globally that showed them weak spots in their network coverage, Reuters reported Monday, citing people familiar with the matter, because of Google's concerns that sharing data from users of its Android phone system might attract the scrutiny of users and regulators. From the report: The withdrawal of the service, which has not been previously reported, has disappointed wireless carriers that used the data as part of their decision-making process on where to extend or upgrade their coverage. Even though the data were anonymous and the sharing of it has become commonplace, Google's move illustrates how concerned the company has become about drawing attention amid a heightened focus in much of the world on data privacy. Google's Mobile Network Insights service, which had launched in March 2017, was essentially a map showing carriers signal strengths and connection speeds they were delivering in each area. The service was provided free to carriers and vendors that helped them manage operations. The data came from devices running Google's Android operating system, which is on about 75% of the world's smartphones, making it a valuable resource for the industry. [...] Nevertheless, Google shut down the service in April due to concerns about data privacy, four people with direct knowledge of the matter told Reuters. Some of them said secondary reasons likely included challenges ensuring data quality and connectivity upgrades among carriers being slow to materialize.

Read more of this story at Slashdot.

The US Army Wants To Microwave Drones in Midair

Mon, 08/19/2019 - 05:34
"The U.S. Army, as part of a broad counter-unmanned aerial systems strategy, is pushing forward with the U.S. Air Force to develop a high-powered microwave weapon," reports Popular Mechanics: Microwave radiation can disrupt or destroy electronic equipment exposed to them, "cooking" internal circuits much in the same way a fork or other metal objects placed in a microwave oven will cause the oven's electronics to melt down. Here's 2018 footage of a Raytheon HPM system tested at Fort Sill in 2018. The Pentagon has researched high powered microwave weapons for years, but the threat of drone swarms may have presented it with the perfect threat. The military is preparing for the eventuality of facing swarms of suicide drones on the battlefield, each carrying an explosive payloads or prepared to make a suicide attack. Current anti-drone weapons include jammers, shotguns, nets, and even birds, but many of these weapons are only effective against one or a small number of drones at once, and not the dozens or more drones envisioned in the worst drone swarm scenarios.... Microwave radiation doesn't care about rain and other inclement weather, it doesn't rely on individual shots of ammunition, and as long as the electrical generator powering is powered on, it will continue to "fire"... The weapon's broad firing arc means it could take out many drones at once, defeating enemy drone swarms. The joint Army/Air Force microwave weapon prototype "should be operational by 2022."

Read more of this story at Slashdot.

PayPal Builds 'Zoid' JavaScript Library To 'Make IFrames Cool Again'

Mon, 08/19/2019 - 02:34
"Earlier this year I gave a talk at FullStack conference in London about making iFrames cool again," writes a lead engineer at PayPal. In a nutshell: iframes let you build user experiences into embeddable 'cross-domain components', which let users interact with other sites without being redirected. There are a metric ton of awesome uses for that other than tracking and advertizing. Nothing else comes close for this purpose; and as a result, I feel we're not using iframes to their full potential. There are big problems, though... My talk went into how at PayPal, we built Zoid to solve some of the major problems with iframes and popups: - Pre-render to avoid the perception of slow rendering - Automatically resize frames to fit child content - Automatically resize frames to fit child content - Pass down any kind of data and functions/callbacks as props (just like React), and avoid the nightmare of cross-domain messaging between windows. - Make iframes and popups feel like first class (cross-domain) components. Zoid goes a long way. But there are certain problems a mere javascript library can not solve. This is my bucket list for browser vendors, to make iframes more of a first class citizen on the web... Because fundamentally: the idea of cross-domain embeddable components is actually pretty useful once you start talking about shareable user experiences, rather than just user-tracking and advertizing which are obviously pills nobody enjoys swallowing. He acknowledges that he "really likes" the work that's been done on Google Chrome's Portals (which he earlier described as "like iframes, but better, and worse.") "I just hope iframes don't get left behind."

Read more of this story at Slashdot.

Massive Ransomware Attack Hits 23 Local Texas Government Offices

Sun, 08/18/2019 - 23:34
Long-time Slashdot reader StonyCreekBare shared this press release from the Texas Department of Information Resources (Dir) press release as of August 17, 2019, at approximately 5:00 p.m. central time: On the morning of August 16, 2019, more than 20 entities in Texas reported a ransomware attack. The majority of these entities were smaller local governments... At this time, the evidence gathered indicates the attacks came from one single threat actor. Investigations into the origin of this attack are ongoing; however, response and recovery are the priority at this time. It appears all entities that were actually or potentially impacted have been identified and notified. Twenty-three entities have been confirmed as impacted. Responders are actively working with these entities to bring their systems back online. The State of Texas systems and networks have not been impacted.

Read more of this story at Slashdot.

A New Idea For Fighting Rising Sea Levels: Iceberg-Making Submarines

Sun, 08/18/2019 - 20:34
To address the affects of global warming, a team of designers "propose building ice-making submarines that would ply polar waters and pop out icebergs to replace melting floes," reports NBC News: "Sea level rise due to melting ice should not only be responded [to] with defensive solutions," the designers of the submersible iceberg factory said in an animated video describing the vessel, which took second place in a recent design competition held by the Association of Siamese Architects. The video shows the proposed submarine dipping slowly beneath the ocean surface to allow seawater to fill its large hexagonal well. When the vessel surfaces, an onboard desalination system removes the salt from the water and a "giant freezing machine" and chilly ambient temperatures freeze the fresh water to create the six-sided bergs. These float away when the vessel resubmerges and starts the process all over again. A fleet of the ice-making subs, operating continuously, could create enough of the 25-meter-wide "ice babies" to make a larger ice sheet, according to the designers. Faris Rajak Kotahatuhaha, an architect in Jakarta and the leader of the project, said he sees the design as a complement to ongoing efforts to curb emissions. "Experts praised the designers' vision but cast doubt on the project's feasibility."

Read more of this story at Slashdot.

Stack Overflow Touts New Programming Solutions Tool That Mines Crowd Knowledge

Sun, 08/18/2019 - 18:45
Stack Overflow shares a new tool from a team of researchers that "takes the description of a programming task as a query and then provides relevant, comprehensive programming solutions containing both code snippets and their succinct explanations" -- the Crowd Knowledge Answer Generator (or CROKAGE): In order to reduce the gap between the queries and solutions, the team trained a word-embedding model with FastText, using millions of Q&A threads from Stack Overflow as the training corpus. CROKAGE also expanded the natural language query (task description) to include unique open source software library and function terms, carefully mined from Stack Overflow. The team of researchers combined four weighted factors to rank the candidate answers... In particular, they collected the programming functions that potentially implement the target programming task (the query), and then promoted the candidate answers containing such functions. They hypothesized that an answer containing a code snippet that uses the relevant functions and is complemented with a succinct explanation is a strong candidate for a solution. To ensure that the written explanation was succinct and valuable, the team made use of natural language processing on the answers, ranking them most relevant by the four weighted factors. They selected programming solutions containing both code snippets and code explanations, unlike earlier studies. The team also discarded trivial sentences from the explanations... The team analyzed the results of 48 programming queries processed by CROKAGE. The results outperformed six baselines, including the state-of-art research tool, BIKER. Furthermore, the team surveyed 29 developers across 24 coding queries. Their responses confirm that CROKAGE produces better results than that of the state-of-art tool in terms of relevance of the suggested code examples, benefit of the code explanations, and the overall solution quality (code + explanation). The tool is still being refined, but it's "experimentally available" -- although "It's limited to Java queries for now, but the creators hope to have an expanded version open to the public soon." It will probably be more useful than Stack Roboflow, a site that uses a neural network to synthesize fake Stack Overflow questions.

Read more of this story at Slashdot.

A Major Cyber Attack Could Be Just As Deadly As Nuclear Weapons

Sun, 08/18/2019 - 17:39
"As someone who studies cybersecurity and information warfare, I'm concerned that a cyberattack with widespread impact, an intrusion in one area that spreads to others or a combination of lots of smaller attacks, could cause significant damage, including mass injury and death rivaling the death toll of a nuclear weapon," warns an assistant Professor of Computer Science, North Dakota State University: Unlike a nuclear weapon, which would vaporize people within 100 feet and kill almost everyone within a half-mile, the death toll from most cyberattacks would be slower. People might die from a lack of food, power or gas for heat or from car crashes resulting from a corrupted traffic light system. This could happen over a wide area, resulting in mass injury and even deaths... The FBI has even warned that hackers are targeting nuclear facilities. A compromised nuclear facility could result in the discharge of radioactive material, chemicals or even possibly a reactor meltdown. A cyberattack could cause an event similar to the incident in Chernobyl. That explosion, caused by inadvertent error, resulted in 50 deaths and evacuation of 120,000 and has left parts of the region uninhabitable for thousands of years into the future. My concern is not intended to downplay the devastating and immediate effects of a nuclear attack. Rather, it's to point out that some of the international protections against nuclear conflicts don't exist for cyberattacks... Critical systems, like those at public utilities, transportation companies and firms that use hazardous chemicals, need to be much more secure... But all those systems can't be protected without skilled cybersecurity staffs to handle the work. At present, nearly a quarter of all cybersecurity jobs in the US are vacant, with more positions opening up than there are people to fill them. One recruiter has expressed concern that even some of the jobs that are filled are held by people who aren't qualified to do them. The solution is more training and education, to teach people the skills they need to do cybersecurity work, and to keep existing workers up to date on the latest threats and defense strategies.

Read more of this story at Slashdot.

XKCD Author Challenges Serena Williams To Attack A Drone

Sun, 08/18/2019 - 16:38
In just 16 days XKCD author Randall Munroe releases a new book titled How To: Absurd Scientific Advice for Common Real-World Problems. He's just released an excerpt from the chapter "How to Catch a Drone," in which he actually enlisted the assistance of tennis star Serena Williams. An anonymous reader writes: Serena and her husband Alexis just happened to have a DJI Mavic Pro 2 with a broken camera -- and Munroe asked her to try to smash it with tennis balls. "My tentative guess was that a champion player would have an accuracy ratio around 50 when serving, and take 5-7 tries to hit a drone from 40 feet. (Would a tennis ball even knock down a drone? Maybe it would just ricochet off and cause the drone to wobble! I had so many questions.) "Alexis flew the drone over the net and hovered there, while Serena served from the baseline..." His blog has the rest of the story, and Munroe has even illustrated the experiment, promising that the book also contains additional anti-drone strategies, an analysis of other sports projectiles, and "a discussion with a robot ethicist about whether hitting a drone with a tennis ball is wrong."

Read more of this story at Slashdot.

Why Am I Receiving Unordered Boxes From Amazon?

Sun, 08/18/2019 - 16:04
It's an unexpected surprise that's been popping up "all over the country," according to the Better Business Bureau. People are receiving boxes of unordered merchandise from Amazon. The companies, usually foreign, third-party sellers that are sending the items are simply using your address and your Amazon information. Their intention is to make it appear as though you wrote a glowing online review of their merchandise, and that you are a verified buyer of that merchandise. They then post a fake, positive review to improve their products' ratings, which means more sales for them. The payoff is highly profitable from their perspective... The fake online review angle is only one way they benefit...they also are increasing their sales numbers. After all, they aren't really purchasing the items since the payment goes right back to them.... Then there is the "porch pirate" angle. There have been instances where thieves used other people's mailing addresses and accounts, then watched for the delivery of the package so they can steal it from your door before you get it... The fact that someone was able to have the items sent to you as if you purchased them indicates that they probably have some of your Amazon account information. Certainly, they have your name and address and possibly, your phone number and a password. The company either hacked your account themselves or purchased the information from a hacker. The BBB notes that although it's strange to receive boxes of unordered merchandise, "You are allowed to keep it. The Federal Trade Commission says you have a legal right to keep unordered merchandise." "The bigger issue is: What do you do about your information having been obtained by crooks?"

Read more of this story at Slashdot.

Comment