By EditorDavid from Slashdot's toxicity-reports department
Slashdot reader Lauren Weinstein writes:
Google has announced (with considerable fanfare) public access to their new "Perspective" comment filtering system API, which uses Google's machine learning/AI system to determine which comments on a site shouldn't be displayed due to perceived high spam/toxicity scores. It's a fascinating effort. And if you run a website that supports comments, I urge you not to put this Google service into production, at least for now.
The bottom line is that I view Google's spam detection systems as currently too prone to false positives -- thereby enabling a form of algorithm-driven "censorship" (for lack of a better word in this specific context) -- especially by "lazy" sites that might accept Google's determinations of comment scoring as gospel... as someone who deals with significant numbers of comments filtered by Google every day -- I have nearly 400K followers on Google Plus -- I can tell you with considerable confidence that the problem isn't "spam" comments that are being missed, it's completely legitimate non-spam, non-toxic comments that are inappropriately marked as spam and hidden by Google.
Lauren is also collecting noteworthy experiences for a white paper about "the perceived overall state of Google (and its parent corporation Alphabet, Inc.)" to better understand how internet companies are now impacting our lives in unanticipated ways. He's inviting people to share their recent experiences with "specific Google services (including everything from Search to Gmail to YouTube and beyond), accounts, privacy, security, interactions, legal or copyright issues -- essentially anything positive, negative, or neutral that you are free to impart to me, that you believe might be of interest."Read Replies (0)
By EditorDavid from Slashdot's browser-bugs department
An anonymous reader quotes BleepingComputer:
Google has gone public with details of a second unpatched vulnerability in Microsoft products, this time in Edge and Internet Explorer, after last week they've published details about a bug in the Windows GDI (Graphics Device Interface) component... The bug, discovered by Google Project Zero researcher Ivan Fratric, is tracked by the CVE-2017-0037 identifier and is a type confusion, a kind of security flaw that can allow an attacker to execute code on the affected machine, and take over a device.
Details about CVE-2017-0037 are available in Google's bug report, along with proof-of-concept code. The PoC code causes a crash of the exploited browser, but depending on the attacker's skill level, more dangerous exploits could be built... Besides the Edge and IE bug, Microsoft products are also plagued by two other severe security flaws, one affecting the Windows GDI component and one the SMB file sharing protocol shipped with all Windows OS versions...
Google's team notified Microsoft of the bug 90 days ago, only disclosing it publicly on Friday.Read Replies (0)
By EditorDavid from Slashdot's dashboard-donors department
An anonymous reader writes:
Two geeks are crowdfunding an open source car hacking tool that will allow builders to experiment with diagnostics, telematics, security, and prototyping. "Cars have become complicated and expensive to work with," they explain on a Kickstarter page. "Macchina wants to use open source hardware to help break down these barriers and get people tinkering with their cars again." After years developing a beta prototype, they announced a tiny plug-and-play device/development platform (that can also be hardwired under the hood) on an Arduino Due board with a 32-bit ARM microcontroller. They almost immediately reached their $25,000 funding goal, and with 24 days left to go they've already raised $41,672, and they're now also selling t-shirts to benefit the EFF's "Right to Repair" activism.
Challenging "the closed, unpublished nature of modern-day car computers," their M2 device ships with protocols and libraries "to work with any car that isn't older than Google." With catchy slogans like "root your ride" and "the future is open," they're hoping to build a car-hacking developer community, and they're already touting the involvement of Craig Smith, the author of the Car Hacker's Handbook from No Starch Press.
"The one thing that all car hobbyists can agree on is that playing with cars isn't cheap," argues the campaign page. "Open source hardware is the answer!"Read Replies (0)
By EditorDavid from Slashdot's or-by-not-using-Microsoft-products department
An anonymous reader quotes Computerworld:
If you want to shut out the overwhelming majority of vulnerabilities in Microsoft products, turn off admin rights on the PC. That's the conclusion from global endpoint security firm Avecto, which has issued its annual Microsoft Vulnerabilities report. It found that there were 530 Microsoft vulnerabilities reported in 2016, and of these critical vulnerabilities, 94% were found to be mitigated by removing admin rights, up from 85% reported last year. This is especially true with the browser, for those who still use Microsoft's browsers. 100% of vulnerabilities impacting both Internet Explorer and Edge could be mitigated by removing admin rights, Avecto reported... Windows 10 was found to have the highest proportion of vulnerabilities of any OS (395), 46% more than Windows 8 and Windows 8.1 (265 each). Avecto found that 93% of Windows 10 vulnerabilities could be mitigated by removing admin rights.
Of course, the stats are based on vulnerabilities announced in Microsoft Security Bulletins, but there's an overwhelming pattern. Turning off admin rights mitigated the vast majority of vulnerabilities, whether it was Windows Server (90%) or older versions of Microsoft Office (99%). And turning off admin rights in Office 2016 mitigated 100% of its vulnerabilities.Read Replies (0)
By EditorDavid from Slashdot's dominated-by-Detroit department
schwit1 quotes Autoblog:
Up until very recently the talk in Silicon Valley was about how the tech industry was going to broom Detroit into the dustbin of history. Companies such as Apple, Google, and Uber -- so the thinking went -- were going to out run, out gun, and out innovate the automakers. Today that talk is starting to fade. There's a dawning realization that maybe there's a good reason why the traditional car companies have been around for more than a century.
Last year Apple laid off most of the engineers it hired to design its own car. Google (now Waymo) stopped talking about making its own car. And Uber, despite its sky high market valuation, is still a long, long way from ever making any money, much less making its own autonomous cars. To paraphrase Elon Musk, Silicon Valley is learning that "Making rockets is hard, but making cars is really hard."
The article argues the big auto-makers launched "vigorous in-house autonomous programs" which became fully competitive with Silicon Valley's efforts, and that Silicon Valley may have a larger role crunching the data that's collected from self-driving cars. "Last year in the U.S. market alone Chevrolet collected 4,220 terabytes of data from customer's cars... Retailers, advertisers, marketers, product planners, financial analysts, government agencies, and so many others will eagerly pay to get access to that information."Read Replies (0)
By EditorDavid from Slashdot's full-Spectrum department
"New York subscribers have had to overpay month after month for services that Spectrum deliberately didn't provide," reports Backchannel -- noting these practices are significant because together Comcast and Charter (formerly Time Warner Cable) account for half of America's 92 million high-speed internet connections. An anonymous reader quotes Backchannel:
Based on the company's own documents and statements, it appears that just about everything it has been saying since 2012 to New York State residents about their internet access and data services is untrue...because of business decisions the company deliberately made in order to keep its capital expenditures as low as possible... Its marketing department kept sending out advertising claims to the public that didn't match the reality of what consumers were experiencing or square with what company engineers were telling Spectrum executives. That gives the AG's office its legal hook: Spectrum's actions in knowingly saying one thing but doing another amount to fraudulent, unfair, and deceptive behavior under New York law...
The branding people went nuts, using adjectives like Turbo, Extreme, and Ultimate for the company's highest-speed 200 or 300 Mbps download offerings. But no one, or very few people, could actually experience those speeds...because, according to the complaint, the company deliberately required that internet data connections be shared among a gazillion people in each neighborhood... [T]he lawsuit won't by itself make much of a difference. But maybe the public nature of the attorney-general's assault -- charging Spectrum for illegal misconduct -- will lead to a call for alternatives. Maybe it will generate momentum for better, faster, wholesale fiber networks controlled by cities and localities themselves. If that happened, retail competition would bloom. We'd get honest, straightforward, inexpensive service, rather than the horrendously expensive cable bundles we're stuck with today.
< article continued at Slashdot's full-Spectrum department
>Read Replies (0)
By EditorDavid from Slashdot's proprietary-consoles department
An anonymous reader quotes Motherboard:
The video game industry is lobbying against legislation that would make it easier for gamers to repair their consoles and for consumers to repair all electronics more generally. The Entertainment Software Association, a trade organization that includes Sony, Microsoft, Nintendo, as well as dozens of video game developers and publishers, is opposing a "right to repair" bill in Nebraska, which would give hardware manufacturers fewer rights to control the end-of-life of electronics that they have sold to their customers...
Bills making their way through the Nebraska, New York, Minnesota, Wyoming, Tennessee, Kansas, Massachusetts, and Illinois statehouses will require manufacturers to sell replacement parts and repair tools to independent repair companies and consumers at the same price they are sold to authorized repair centers. The bill also requires that manufacturers make diagnostic manuals public and requires them to offer software tools or firmware to revert an electronic device to its original functioning state in the case that software locks that prevent independent repair are built into a device. The bills are a huge threat to the repair monopolies these companies have enjoyed, and so just about every major manufacturer has brought lobbyists to Nebraska, where the legislation is currently furthest along... This setup has allowed companies like Apple to monopolize iPhone repair, John Deere to monopolize tractor repair, and Sony, Microsoft, and Nintendo to monopolize console repair...
Motherboard's reporter was unable to get a comment from Microsoft, Apple, and Sony, and adds that "In two years of covering this issue, no manufacturer has ever spoken to me about it either on or off the record."Read Replies (0)
By EditorDavid from Slashdot's open-source-Open-Source-guides department
An anonymous reader quotes InfoQ:
GitHub has recently launched its Open Source Guides, a collection of resources addressing the most common scenarios and best practices for both contributors and maintainers of open source projects. The guides themselves are open source and GitHub is actively inviting developers to participate and share their stories... "Open source is complicated, especially for newcomers. Experienced contributors have learned many lessons about the best way to use, contribute to, and produce open source software. Everyone shouldn't have to learn those lessons the hard way." Making a successful first contribution is not the exclusive focus of the guides, though, which also strives to make it easier to find users for a project, starting a new project, and building healthy open source communities. Other topics the guides dwell on are best practices, getting financial support, metrics, and legal matters.
GitHub's Head of Open Source says the guides create "the equivalent of a water cooler for the community."Read Replies (0)
By EditorDavid from Slashdot's hole-in-the-host department
An anonymous IT geek writes:
Cloudflare-hosted web sites have been leaking data as far back as September, according to Gizmodo, which reports that at least Cloudflare "acted fast" when the leak was discovered, closing the hole within 44 minutes, and working with search engines to purge their caches. (Though apparently some of it is still lingering...) Cloudflare CEO Matthew Prince "claims that there was no detectable uptick in requests to Cloudflare-powered websites from September of last year...until today. That means the company is fairly confident hackers didn't discover the vulnerability before Google's researchers did."
And the company's CTO also told Reuters that "We've seen absolutely no evidence that this has been exploited. It's very unlikely that someone has got this information... We do not know of anybody who has had a security problem as a result of this." Nevertheless, Fortune warns that "So many sites were vulnerable that it doesn't make sense to review the list and change passwords on a case-by-case basis." Some sites are now even resetting every user's password as a precaution, while site operators "are also being advised to wipe their sites' cookies and security certificates, and perform their own web searches to see if site data leaked." But I'd like to know what security precautions are being taken by Slashdot's readers?
Leave your own answers in the comments. How did you respond to Cloudbleed?Read Replies (0)
By EditorDavid from Slashdot's cognitive-Captain-Falcon department
"The AI is definitely godlike," one professional player told Quartz. "I am not sure if anyone could beat it." An anonymous reader quotes their report about an AI's showdown with the best players of Super Smash Bros. Melee:
Of 10 professionals that faced the bot, each one was killed more than they could kill the bot... But the bot was once only as good as a mere mortal. At first, Vlad Firoiu, creator and a competitive Smash player himself, couldn't train 'Phillip' to be as strong as the in-game bot, which he says even the worst players can beat fairly easily. Firoiu's solution? He started making the bot play itself over and over again, slowly learning which techniques fail and which succeed, called reinforcement learning. Then, he left it alone.
"I just sort of forgot about it for a week," said Firoiu, who coauthored an unreviewed paper with William F. Whitney, the NYU student [who helped him] on the work. "A week later I looked at it and I was just like, 'Oh my gosh.' I tried playing it and I couldn't beat it."
Business Insider points out that their AI read the players positions, velocities, and states directly from the game's memory, so the AI responds six times faster than a human player. To compensate it played as Captain Falcon, the game's slowest character, but there was one crucial glitch. "One particularly clever player found that the simple strategy of crouching at the edge of the stage caused the network to behave very oddly, refusing to attack and eventually KOing itself by falling off the other side of the stage."Read Replies (0)
By EditorDavid from Slashdot's hashing-things-out department
Google's researchers specifically cited Git when they announced a new SHA-1 attack vector, according to ZDNet. "The researchers highlight that Linus Torvald's code version-control system Git 'strongly relies on SHA-1' for checking the integrity of file objects and commits. It is essentially possible to create two Git repositories with the same head commit hash and different contents, say, a benign source code and a backdoored one,' they note." Saturday morning, Linus responded:
First off - the sky isn't falling. There's a big difference between using a cryptographic hash for things like security signing, and using one for generating a "content identifier" for a content-addressable system like git. Secondly, the nature of this particular SHA1 attack means that it's actually pretty easy to mitigate against, and there's already been two sets of patches posted for that mitigation. And finally, there's actually a reasonably straightforward transition to some other hash that won't break the world - or even old git repositories...
< article continued at Slashdot's hashing-things-out department
>Read Replies (0)
By EditorDavid from Slashdot's killed-in-Kansas department
Garmin engineer Srinivas Kuchibhotla was shot and killed at a local bar in Olathe, Kansas, the U.S. headquarters of Garmin. Co-worker Alok Madasani was also injured along with bystander Ian Grillot, who attempted to help the men. "The suspect in the shooting, Adam Purinton, was drinking at the bar in Olathe, Kansas, at about 7:15 p.m. that night," reports The Verge. "A witness said he yelled 'get out of my country' to two of the victims, reportedly saying the men, believed to originally be from India, were 'Middle Eastern.'" In 2015, Garmin employed 2,700 workers in Olathe and has plans to double this number, which the article notes has led to "increasing diversity" in the community.Read Replies (0)
By EditorDavid from Slashdot's unauthorized-flying-objects department
Between February and September of 2016, there were 1,274 reports of drones near airports -- versus just 874 for the same period in 2015, according to newly-released FAA research. "The report detailed more than 1,200 incidents of airplane pilots, law enforcement, air traffic controllers, and U.S. citizens reporting drones flying in places they shouldn't," writes Fortune. An anonymous reader quotes their report:
One of takeaway of the report was that while the FAA has received several reports from pilots that drones may have hit their aircraft, the administration was unable to verify any such claim. "Every investigation has found the reported collisions were either birds, impact with other items such as wires and posts, or structural failure not related to colliding with an unmanned aircraft," the FAA said in a statement... Although a drone hasn't smashed into an airplane yet, the FAA "wants to send a clear message that operating drones around airplanes and helicopters is dangerous and illegal. Unauthorized operators may be subject to stiff fines and criminal charges, including possible jail time," the FAA said.Read Replies (0)