By EditorDavid from Slashdot's tested-in-production department
Over a million lines of code -- in existence for over 10 years -- gets updates in six-week "sprints" using source control and bug-tracking systems. But now an anonymous reader writes:
In theory users report bugs, the developers "fix" the bugs, the users test and accept the fix, and finally the "fix" gets released to production as part of a larger change-set. In practice, the bug is reported, the developers implement "a fix", no one else tests it (except for the developer(s) ), and the "fix" gets released with the larger code change set, to production.
We (the developers) don't want to release "fixes" that users haven't accepted, but the code changes often include changes at all levels of the stack (database, DOAs, Business Rules, Webservices and multiple front-ends). Multiple code changes could be occurring in the same areas of code by different developers at the same time, making merges of branches very complex and error prone. Many fingers are in the same pie. Our team size, structure and locations prevent having a single gatekeeper for code check-ins... What tools and procedures do you use to prevent un-approved fixes from being deployed to production as part of the larger code change sets?
Fixes are included in a test build for users to test and accept -- but what if they never do? Leave your best answers in the comments. How woud you stop un-approved code changes from being deployed?Read Replies (0)
By EditorDavid from Slashdot's Ubering-under-the-influence department
While Uber's bookings doubled last year, the company still showed a net lost of $2.8 billion. And now, "California regulators are recommending that Uber pay a $1.13 million fine for not investigating rider complaints that drivers were working intoxicated." An anonymous reader writes:
California "requires ride-hailing companies to have a zero-tolerance policy for driving under the influence of alcohol or drugs," notes Reuters -- and yet Tuesday's order reports that investigators "found no evidence that (Uber) followed up in any way with zero-tolerance complaints several hours or even one full day after passengers filed such complaints." Investigators from the state's Public Utilities Commission are asking the full commission to examine their findings,
"To confirm the policy, regulators analyzed selected complaints against drivers who received three or more complaints," Reuters reports. Though Uber has sometimes suspended drivers within one hour of customer complaints -- 22 times -- they've apparently received 2,047 drug- or alcohol-related complaints between August 2014 and August of 2015. "The company said drivers were banned from working in 574 of those complaints, according to the order. But regulators then reviewed 154 complaints, and determined that the company failed to promptly suspend drivers in 149 complaints. The company also failed to investigate 133 complaints, and did not suspend a driver or investigate 113 complaints, the order shows... In at least 25 instances, Uber failed to suspend or investigate a driver after three or more complaints, the order states."
< article continued at Slashdot's Ubering-under-the-influence department
>Read Replies (0)
By EditorDavid from Slashdot's a-textbook-example department
"The University System of Maryland has awarded 21 "mini grants" to university faculty to "help them expand open education resources," reports OpenSource.com. Recipients of the grants are also given time off to prepare courses that use open textbooks, and will receive personalized support and training on effective course design.
An anonymous reader writes:
"Although our faculty view textbooks as essential, some of our students see them as a luxury they cannot afford," said Community College of Baltimore County President Sandra Kurtinitis. "Having access to open educational resources will provide some financial relief for our students as well as contribute to their academic success." The cost of textbooks has risen 812% since 1978, the school system said in an announcement, "outpacing even the cost of medical services and new housing. Nationally, students spend an average of $1,200 a year on textbooks."
The Maryland Open Source Textbook initiative started in 2013 "to provide a state-wide opportunity for faculty to explore the promise of open education resources to reduce students' cost of attendance while maintaining, or perhaps even improving, learning outcomes." Since then it's helped replace traditional textbooks in over 60 different courses at 14 public institutions across the state, resulting in a cumulative cost savings of over $1 million for 3,500 students. "In addition to saving students money, faculty have gained the ability to adapt and customize their instructional materials to ensure they are aligned with their pedagogical methods to best meet their students' needs," the school system reports. "In follow up surveys with students participating in the MOST initiative, 93% reported that the open educational resource content they used was the same or better quality than traditional textbooks."Read Replies (0)
By EditorDavid from Slashdot's seeing-the-attraction department
"Predicting magnets is a heck of a job, and their discovery is very rare," said a mechanical engineering professor at Duke University. But after years of work synthesizing various predictions, material scientists "predicted and built two new magnetic materials, atom-by-atom, using high-throughput computational models." An anonymous reader quotes Phys.org:
The success marks a new era for the large-scale design of new magnetic materials at unprecedented speed. Although magnets abound in everyday life, they are actually rarities -- only about 5% of known inorganic compounds show even a hint of magnetism. And of those, just a few dozen are useful in real-world applications because of variability in properties such as effective temperature range and magnetic permanence...
In a new study, materials scientists from Duke University provide a shortcut in this process. They show the capability to predict magnetism in new materials through computer models that can screen hundreds of thousands of candidates in short order. And, to prove it works, they've created two magnetic materials that have never been seen before.
"The first alloy is particularly interesting," reports the International Business Times, "because it contains no rare-earth materials, which are both expensive and difficult to acquire." But a Duke mechanical engineering professor points out that "It doesn't really matter if either of these new magnets proves useful in the future. The ability to rapidly predict their existence is a major coup and will be invaluable to materials scientists moving forward."Read Replies (0)
By EditorDavid from Slashdot's digital-DIY department
"It turns out that the NES Classic Edition is just a little Linux-powered board inside a cute case," writes Andrew Cunningham at Ars Technica, "and it's totally possible to build your own tiny Linux-powered computer inside a cute case without spending much more than $60." An anonymous reader writes:
Andrew used a $42 Raspberry Pi 3 Model B -- "it's relatively cheap and relatively powerful, and it can easily handle anything from the original PlayStation on down" -- plus an $8 case, and a microSD card. He also purchased a pair of gamepads -- there's several options -- and reports that "Putting our little box together is ridiculously easy, and you ought to have no problem with it even if you've never opened up a PC tower in your life."
"Making retro game consoles is a fairly common use case for the Pi, so there are a few different operating system choices out there," Andrew reports, and he ultimately chose the Linux-based RetroPie OS, which includes a number of emulators. Basically the process boils down to dropping a RetroPie boot image onto the SD card, putting it into the Pi, and then plugging it into your display and connecting your controllers -- plus configuring some menus. "The default quality of the emulation looks just as good as it does on the NES Classic Edition," and "the emulators for these older systems are all advanced enough that things should mostly run just like they did on the original hardware... I've been having a ton of fun with mine now that it's all set up, and its flexibility (plus the quality of those USB gamepads) has made it my favorite way to play old games, outpacing my Apple TV, the pretty but not-living-room-friendly OpenEmu, and the old hacked Wii I still have sitting around."
< article continued at Slashdot's digital-DIY department
>Read Replies (0)
By EditorDavid from Slashdot's let's-do-the-time-warp department
USA Today asked Steve Wozniak to predict what the world will look like in 2075 -- one hundred years after the founding of Apple. An anonymous reader writes:
"He's convinced Apple, Google and Facebook will be bigger in 2075," according to the article -- just like IBM, which endured long past its founding in 1911. Pointing to Apple's $246.1 billion in cash and marketable securities, Wozniak says Apple "can invest in anything. It would be ridiculous to not expect them to be around... The same goes for Google and Facebook."
Woz predicted portable laptops back in 1982, and now says that by 2075, we could also see new cities built from scratch in the deserts, with people wearing special suits to protect them from the heat. AI will be ubiquitous in all cities, as consumers interact with smart walls to communicate -- and to shop -- while home medical devices will allow self-diagnosis and doctor-free prescriptions. And according to the article, Woz "is convinced a colony will exist on the Red Planet. Echoing the sentiments of Amazon CEO Jeff Bezos, whose Blue Origin start-up has designs on traveling to Mars, Wozniak envisions Earth zoned for residential use and Mars for heavy industry." (Though he doesn't have high hopes that we'll ever meet aliens.)
Woz is promoting the Silicon Valley Comic Con next weekend. (Not coincidentally, its theme is "The Future of Humanity: Where Will We Be in 2075?") During the interview, Woz pointed at a colleague's iPhone, smiled broadly and said it "shows you how exciting the future can be."Read Replies (0)
By EditorDavid from Slashdot's predicting-the-future department
The CTO of Textio is describing the "already happening" AI disruption that no one's noticed, arguing that voice-activated assistants are "just one small part of what AI is about -- and not the part that will matter the most for the enterprise companies that actually buy almost $4 trillion in software and services each year." An anonymous reader writes:
Jensen Harris describes "the less-flashy flavor of AI that is changing the nature of work itself: headless AI...the application of artificial intelligence to vastly improve internal business processes. It is fully transforming the crucial machinery of business -- processes like hiring, lead generation, financial modeling, and information security. Legacy software has become a commodity in all of these areas, and purpose-built AI solutions will get a larger and larger wallet share of these huge enterprise cost centers."
Combining machine intelligence with learning loops, these constantly-evolving algorithms are "where the money is," since headless AI "doesn't try to replace people; it gives them superpowers" -- for example, predicting the future. Harris ultimately argues that headless AI are delivering "radical productivity leaps that they haven't seen from software in decades... In the near future, every core business function will have been transformed by AI -- hiring, sales, security, marketing, finance, manufacturing...everything... Legacy software will get squeezed down into a smaller portion of the IT wallet as the most valuable services become the native AI platforms -- just as form-based desktop software got squeezed out by the cloud in the last generation... the real enterprise revolution is happening in the companies that are using headless AI to transform their core businesses."
< article continued at Slashdot's predicting-the-future department
>Read Replies (0)
By EditorDavid from Slashdot's weasels-on-the-web department
An anonymous reader writes:
Three doctoral students at Stony Brook University spent eight months analyzing internet scammers who pose as remote tech support workers (usually pretending to be from Microsoft of Apple). Their research revealed more than 25,000 scam domains and thousands of different scam phone numbers. "Although victims of these scams can be anywhere, the researchers found that 85.4% of the IP addresses in these scams were located across different regions of India," reports On The Wire, "with 9.7% located in the United States and 4.9% in Costa Rica. Scammers typically asked users for an average of $291, with prices ranging from $70 to $1,000."
The researchers even called 60 of the con artists to study their technique, and concluded most were working in large, organized call centers. They use remote access tools, and in fact two popular tools were used in 81% of the scams, according to the paper. "We found that, on average, a scammer takes 17 minutes, using multiple social engineering techniques mostly based on misrepresenting OS messages, to convince users of their infections..."Read Replies (0)
By EditorDavid from Slashdot's doctor-doctor department
First, "A new study finds that nearly 9 in 10 people who go for a second opinion after seeing a doctor are likely to leave with a refined or new diagnosis from what they were first told," according to an article shared by Slashdot reader schwit1:
Researchers at the Mayo Clinic examined 286 patient records of individuals who had decided to consult a second opinion, hoping to determine whether being referred to a second specialist impacted one's likelihood of receiving an accurate diagnosis. The study, conducted using records of patients referred to the Mayo Clinic's General Internal Medicine Division over a two-year period, ultimately found that when consulting a second opinion, the physician only confirmed the original diagnosis 12 percent of the time. Among those with updated diagnoses, 66% received a refined or redefined diagnosis, while 21% were diagnosed with something completely different than what their first physician concluded.
But in a related story, Slashdot reader sciencehabit writes that four machine-learning algorithms all performed better than currently-used algorithm of the American College of Cardiology, according to newly-published research, which concludes that "machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others."
"I can't stress enough how important it is," one Stanford vascular surgeon told Science magazine, "and how much I really hope that doctors start to embrace the use of artificial intelligence to assist us in care of patients."Read Replies (0)
By EditorDavid from Slashdot's see-you-for-10-seconds-in-court department
"Saying it had 'nothing to hide,' the company behind Snapchat released an unredacted version of a lawsuit filed against it by a former employee that claims investors and advertisers were misled about usage data." And one allegation -- about a meeting with the company's 25-year-old CEO about flawed user metrics and low adoption in India in Spain -- is particularly embarrassing.
Pompliano, who had just been hired away from Facebook, contends that he presented methods to address the issue, but that Evan Spiegel, the company's CEO, abruptly cut him off. "This app is only for rich people," Spiegel said, according to Pompliano. "I don't want to expand into poor countries like India and Spain"... Pompliano claims that Spiegel then met with two other executives and determined that "Mr. Pompliano presented a risk to Snapchat's IPO."
It may have been a flip remark, but the lawsuit also alleges two data analysts confided to Pompliano that Snapchat had "an institutional aversion to looking at user data," where its efforts showed "utter incompetence". The former employee -- who was fired after three weeks -- alleges that Snapchat inflated the rate of completed registrations and the number of users who stayed longer than seven days.
Snap originally said the lawsuit should remain redacted because it contained damaging trade secrets that would help its competitors, but now Snap attorneys are accusing Pompliano and his attorneys of "just making things up... The simple fact is that he knows exactly nothing about Snap's current metrics." Variety reports that Pompliano's attorney "said that Snap withdrew its effort to seal the complaint because the company knew it would lose."Read Replies (0)
By EditorDavid from Slashdot's spaces-instead-of-tabs department
Researchers recently surveyed 2,200 software developers to calculate the distribution of unhappiness throughout the profession, and to identify its top causes, "incorporating a psychometrically validated instrument for measuring (un)happiness." An anonymous reader quotes Motherboard:
Daniel Graziotin and his team found their survey subjects via GitHub. Contact information was found by mining archived data for past public GitHub events, where email addresses are apparently more plentiful. They wound up with 33,200 records containing developer locations, contact information, and employers. They took a random sampling from this dataset and wound up with about 1,300 valid survey responses... According to survey results released earlier this month, software developers are on average a "slightly happy" group of workers...
Survey responses were scored according to the SPANE-B metric, a standard tool used in psychology to assess "affect," defined as total negative feelings subtracted from total positive feelings. It ranges from -24 to 24. The mean score found in the developer happiness survey was 9.05. Slightly happy. The minimum was -16, while the maximum was 24. So, even in the worst cases, employees weren't totally miserable, whereas in the best cases employees weren't miserable at all.
The paper -- titled "On the Unhappiness of Software Developers" -- found that the top cause of unhappiness was being stuck while solving a problem, followed by "time pressure," bad code quality/coding practices, and "under-performing colleague."
And since happiness has been linked to productivity, the researchers write that "Our results, which are available as open data, can act as guidelines for practitioners in management positions and developers in general for fostering happiness on the job...unhappiness is present, caused by various factors and some of them could easily be prevented."Read Replies (0)
By EditorDavid from Slashdot's way-back-machines department
We've recently seen stories about old computers and sys-ops resurrecting 1980s BBS's, but now an anonymous reader has a question for all Slashdot readers:
Whenever I meet geeks, there's one question that always gets a reaction: Do you remember your first home computer? This usually provokes a flood of fond memories about primitive specs -- limited RAM, bad graphics, and early versions of long-since-abandoned operating systems. Now I'd like to pose the same question to Slashdot's readers.
Use the comments to share details about your own first home computer. Was it a back-to-school present from your parents? Did it come with a modem? Did you lovingly upgrade its hardware for years to come? Was it a Commodore 64 or a BeBox?
It seems like there should be some good stories, so leave your best answers in the comments. What was your first home computer?Read Replies (0)
By EditorDavid from Slashdot's militarized-zone department
First, an anonymous reader quotes Inverse:
On Saturday, the North Korean military paraded an unprecedented array of weapons through Kim Il-sung Square in the center of Pyongyang... "We're totally floored right now," Dave Schmerler of the Middlebury Institute of International Studies in Monterey, California, tells the Wall Street Journal. "I was not expecting to see this many new missile designs." Schmerler tells The Journal that the large missiles -- the "frankenmissiles," as he calls them -- in the parade appear to be hybrids of the North Korean KN-08 and KN-14 missiles, both of which are ICBMs.
But at least one arms control expert noted that while the parade included ICBM-sized canisters, "what's inside is anyone's guess" -- and there's still mixed results for the country's missile program. "An attempted missile launch by North Korea on Sunday failed, US and South Korean defense officials told CNN... At this point, US military officials don't believe the missile had intercontinental capabilities, a US defense official told CNN." The official said there was limited data -- because the missile blew up so quickly -- prompting CNN.com to run the story under the headline "Show of Strength a Flop."
Update: Slashdot reader Dan Drollette is a science writer/editor and foreign correspondent for Bulletin of the Atomic Scientists, and contacted us earlier today to share his recently-published analysis "to delve into what has been happening lately...and to discredit some common tropes in the media, such as the idea that 'North Korea is about to collapse,' 'China has a lot of influence over North Korea,' 'North Korea can credibly threaten the United States right now,' 'North Korea has no reason to feel threatened,' or 'The North can be completely denuclearized.'"Read Replies (0)
By EditorDavid from Slashdot's red-lights-in-the-steering-wheel department
Using LIDAR sensors, Cadillac mapped 160,000 miles of U.S. highways "within five centimeters of accuracy" to give its hands-free-on-the-highway cars the ability to better anticipate the roads ahead -- and to know when a human driver should take over. An anonymous reader writes:
"The car can see farther than the sensors on the car with the map..." says the chief engineer for Cadillac's new "Super Cruise" hands-free driving option for highways, "so if we have a sharp curve, we can anticipate that." The system also gives Cadillac's vehicles a safety check not available to Tesla, which can't stop drivers from using Tesla's semi-autonomous Autopilot even when they're not on a highway. "We know where the car is because of the LIDAR map and the other data in the car," says a product communications manager at Cadillac. "Therefore we have the ability to geofence it." In addition,
The Verge reports that if drivers look away for more than 30 seconds, "the car will know thanks to an infrared camera attached to the top of the steering column. Eyes closed? The car will know and start a sequence of alerts to get the driver's focus back on the road. It can even see through UV-blocking sunglasses." While the camera doesn't record or store data, it will flash a strip of red LED lights embedded in the top of the steering wheel "if the driver is caught not paying attention."
Cadillac plans to create and transmit an updated map every year, and will also regularly update its map by "constantly" checking the database from the Transportation Department, and deploying own trucks to draw new maps of construction areas.Read Replies (0)
By EditorDavid from Slashdot's town-hall-tempests department
Wisconsin congressman F. James Sensenbrenner Jr. defended his decision to help repeal broadband privacy rules by telling a constituent, "Nobody's got to use the Internet." An anonymous reader quotes the 73-year-old congressman:
"And the thing is that if you start regulating the Internet like a utility, if we did that right at the beginning, we would have no Internet... Internet companies have invested an awful lot of money in having almost universal service now. The fact is is that, you know, I don't think it's my job to tell you that you cannot get advertising for your information being sold. My job, I think, is to tell you that you have the opportunity to do it, and then you take it upon yourself to make that choice... That's what the law has been, and I think we ought to have more choices rather than fewer choices with the government controlling our everyday lives."
"The congressman then moved on to the next question," reports The Washington Post, but criticism of his remarks appeared on social media. One activist complained that the congressman's position was don't use the internet if you don't want your information sold to advertisers -- drawing a clarification from the congressman's office. "Actually he said that nobody has to use the Internet. They have a choice. Big difference."Read Replies (0)
By EditorDavid from Slashdot's vulnerabilities-from-Vimeo department
An anonymous reader quotes ThreatPost:
A popular version of the open source Magento ecommerce platform is vulnerable to a zero-day remote code execution vulnerability, putting as many as 200,000 online retailers at risk... According Bosko Stankovic, information security engineer at DefenseCode, despite repeated efforts to notify Magento, which began in November 2016, the vulnerability remains unpatched despite four version updates since the disclosure. Affected versions of the Magento Community Edition software include v. 2.1.6 and below. DefenseCode did not examine Magento Enterprise, the commercial version of the platform, but warns both share the same underlying vulnerable code... The remote code execution (RCE) vulnerability is tied to the default feature in Magento Community Edition that allows administrators to add Vimeo video content to product descriptions.
DefenseCode says the exploit can be mitigated by enforcing Magento's "Add Secret Keys To URLS" feature, warning in a paper that the hole otherwise "could lead to remote code execution and thus the complete system compromise including the database containing sensitive customer information such as stored credit card numbers and other payment information." Magento has confirmed the exploit, says they're investigating it, and promises they'll address it in their next patch release.Read Replies (0)