This Week in Modern Software logoWelcome to This Week in Modern Software, or TWiMS, our weekly roundup of the need-to-know news, stories, and events of interest surrounding software analytics, cloud computing, application monitoring, development methodologies, programming languages, and the myriad of other issues that influence modern software.

This week, our top story concerns Volkswagen’s headline-shattering software malfeasance.


TWiMS Top Story:
Volkswagen Emission Cheating Affects 11 Million Worldwide—The Washington Post

What it’s about: Software can be incredibly powerful, but that power can be misappropriated and even outright abused. The automaker Volkswagen is reeling from the revelations that it equipped some 11 million diesel vehicles worldwide with software—known as a “defeat device”—designed to cheat emissions tests used to measure environmental output and minimize pollution. As this excellent explainer from The New York Times notes, the software itself is astonishingly straightforward: It sensed when a vehicle was being tested and kicked in its emissions-control systems to pass. The problem? The software then turned those systems off during regular driving, meaning the vehicles are releasing much higher levels of pollutants than allowed. Speculation on the “why” behind the scandal includes the fact that the software enabled VW to lower fuel consumption by turning off the vehicle’s emissions controls.

cartoon car polluting air: this week in modern softwareWhy you should care: Just because you can do something with software does not necessarily mean you should. The fallout from the scandal has been massive and continues to grow by the day: Volkswagen’s CEO was forced to resign. Multiple employees have been suspended pending further investigation. The company lost $15 billion in market value on Monday alone, and faces billions of dollars in government penalties. The EPA will redo emissions testing processes and rules in the wake of the revelations, and will order VW to recall around 500,000 vehicles in the United States. And more consequences, government subpoenas among them, are expected in the coming weeks.

“With great power comes great responsibility,” as the saying goescompanies in all industries must apply the lesson to how they use software and data in their business. Properly managing that responsibility will only grow more essential over time as software impacts more and more aspects of business operations and daily life.

Further reading: 


Microsoft Office 2016 and the End of the Major Software UpgradeFast Company

What it’s about: Microsoft released Office 2016 this week, a modernized edition of the venerable suite of productivity apps businesses have relied upon for more than two decades. While there is plenty of new stuff around collaboration, mobility, and other areas—arriving to generally positive reviews, it appears—it may not be a particular feature or app that’s most significant here. Rather, Office 2016—as with Windows 10—delivers on Microsoft’s promise to retire the major, two-to-three-years-give-or-take release cycle in favor of faster, regular updates consistent with modern software delivery and user expectations. Even a version name like “2016” or “10” may soon be irrelevant, something Microsoft effectively declared with the release of Windows 10. Microsoft’s not getting out of the software business; rather, it’s getting out of the “major release” game. Fast Company’s Harry McCracken points out that the company will treat Windows, and now Office, as a service going forward: “It’ll evolve a little at a time, not in epoch-shifting reboots.”

Why you should care: When the once-undisputed king of business software implements such a historic change to how it delivers its flagship products, we should probably perk up and take notice. This evolution didn’t happen overnight, of course—CEO Satya Nadella has been talking up the strategy pretty much since day one of his tenure. But now two of Microsoft’s longtime rainmakers exist firmly in the modern software realm, and as a result are as relevant as ever. Better still, Microsoft has dropped the protectionist ecosystem mindset in favor an “Office everywhere” approach—something that McCracken and others note is now actually a differentiator for Office: You can now use it on just about any device or platform, a far cry from the not-so-distant days when, say, an iPad user couldn’t open their Excel spreadsheets. Software companies must evolve or face extinction: Microsoft clearly plans to be around for years to come.

Further reading:


Apple’s Rapid iOS 9 Adoption Shows the Future of ITComputerworld

What it’s about: Of course, Microsoft still faces stronger competition than ever, including in the enterprise market it has long dominated, where Apple, in particular, appears intent on gaining ground. Apple’s mainstream popularity, especially with the iPhone, remains enormous—as evident in the rapid adoption of iOS 9 upon its release last week. The company said more than half of iPhone, iPad, and iPod touch users had already downloaded the newest iOS as of last Saturday, following its Wednesday release. That puts it on course to be the most-downloaded software in Apple history. And iOS 9 adoption will certainly get a boost with Friday’s release of the iPhone 6s and 6s Plus, which Apple watchers expect will set new sales records—and which will undoubtedly enter corporate environments at a rapid-fire pace.

Why you should care: Developers don’t really need to be told why a new iPhone or iOS matters, per se. But in the bigger picture, Computerworld’s Jonny Evans offers a take on what iOS 9’s record-setting adoption could foretell of the future of IT. Among other conclusions: The massive popularity of iOS, along with other developments, could solve fragmentation and other hurdles of the “connected world” people envision when they talk about the Internet of Things and the future of the digital age. Evans also cites Tech.pinions’ Tim Bajarin’s recent post on why iOS could become the go-to operating system of enterprise IT when future generations of digital natives enter (and later ascend in) the workforce. “One could say the iPhone and iPad with iOS is laying the groundwork for Apple to eventually own, not only the consumer market but, over time, the business market too,” Bajarin writes.

Even if iOS doesn’t really become the go-to enterprise OS—a notion that seems to focus more on end users and their devices rather than backend IT—it’s still critical to Apple’s enterprise plans. Network World’s Jon Gold, for one, noted that iOS 9—rather than hardware or Apple’s recent high-profile partnerships with the likes of Cisco and IBM—is its more important enterprise weapon, perspective we covered here last week.

Further reading:


Will Wearables Reshape Sports and Athletic Performance?

What it’s about: This isn’t your household Fitbit: A startup in Boston called WHOOP has unveiled a new wearable wristband, à la Fitbit and other fitness-oriented devices, that’s going to do bit more than count your steps or your calories. The company’s founder tells that the device, which an athlete wears 24 hours a day, applies predictive analytics to the data it receives from the person’s body—up to 150 MB a day transferred via Bluetooth to a dashboard—to make better decisions about training, recovery, and even the athlete’s travel schedule. (WHOOP also says on its website that it’s working with the military.) Such promise comes at a great price: Each device runs between $500 to $5,000, according to ESPN, and teams might shell out up to $100,000 annually to monitor and analyze the collected data.

athlete with digital readouts: this week in modern softwareWhy you should care: Even if you’re not an Olympic sprinter or a budding soccer star, the WHOOP device may tell us something about the future of wearables and the apps that make them truly useful. For starters, as with mobile apps in general, the real money might be in the enterprise—the multibillion-dollar sports industry certainly qualifies. Moreover, it’s a reminder that the applications running on and behind wearables—in other words, what you can actually do with these devices—are what will make them truly valuable, rather than the devices themselves. This is something that’s increasingly evident in areas such as healthcare. In any context, from sports to the military to hospitals and beyond, it’s not simply enough to collect data—it’s what you do with it next that matters most.


The Countries Introducing Coding Into the CurriculumJAXenter

What it’s about: If you want to ensure a bright future for the next generation of software developers, you’d better start ’em young. Very young, in the case of England, which last year added a mandatory computing curriculum in its schools for children ages 5 through 16. England’s program includes learning programming language such as Java, among other components. Once chiefly the province of higher education, private training, or self-directed learning, software development and other computer science topics are beginning to make their way into primary and secondary school curriculums around the world, with “digital literacy” joining core fundamentals such as reading, writing, and math.

Why you should care: Well, if you’re in the United States, you might notice something about JAXenter’s piece on the early leaders incorporating programming skills into their schools: We’re not on the list. While coding and related computing skills have received plenty of attention from the White House and organizations such as in recent years, JAXenter notes England and other European nations are the leaders in school-based computing education from a young age, with countries in the Asia Pacific region also beginning to add programming to their school curricula. A recent Microsoft survey reported three out of four students in Asia Pacific want coding to be a core subject in school. JAXenter points out one pitfall, however, that any nation or school would do well to avoid: Limiting computing education to the “best” schools; high schools in Singapore, for example, have already begun piloting Python classes, but “it’s been noted that these schools all belong to the middle-to-upper tier of Singapore’s public education system.”


Software Is Smart Enough for SAT, but Still Far From IntelligentThe New York Times

What it’s about: If you’re a reader of a certain age, how do you think you would you fare if you took the SAT today? Think you could outscore a nascent artificial intelligence program? It might be harder than you think: Researchers at the Allen Institute for Artificial Intelligence at the University of Washington revealed last weekend that their GeoSolver A.I. program had correctly answered geometry questions pulled from the SAT, an exam long used by U.S. college admissions departments, at the level of an 11th-grade student. The program had not previously seen the questions. While it’s being hailed as a significant milestone for A.I., researchers, including GeoSolver’s developer, say it’s also a reminder that A.I. has a long way to go before it can actually match the power of the human brain.

digital brain: this week in modern softwareWhy you should care: Once the harrowing flashbacks of SAT anxiety subside, the breakthough reveals some truths about modern software and data. Among others, ambitious software projects—we’d say building a program that actually matches human intelligence qualifies—often require solving considerable problems, and that takes time. Moreover, not everyone will always agree on the nature of those problems or their solutions. For instance, Ali Farhadi, the A.I. researcher who designed GeoSolver, tells The New York Times that some of his colleagues have declared machine vision “a solved problem. My answer is, ‘Call me when you’ve solved this,’” noting that A.I. programs still can’t parse simple visual instructions that a young child can understand.

Similar debates abound, such as whether the 65-year-old Turing test is a valid measure of machine intelligence. Just as the SAT isn’t a singular measure of human intelligence, modern applications and data require multiple tools and methodologies to best monitor, measure, and analyze them. Gary Marcus, a cognitive scientist at New York University, tells the Times the same must be true for the future of A.I.: “There is no one measure of human intelligence. Why should there be just one A.I. test?”

Want to suggest something that we should cover in the next edition of TWiMS? Email us at


Tune In to the Future

Modern Software Podcast logoCan’t get enough modern software news and commentary? Be sure to check out our new monthly Modern Software Podcast. New Relic Editor-in-Chief Fredric Paul and guests discuss the most important things happening in the world of software analytics, cloud computing, application monitoring, development methodologies, programming languages, and more. Listen to the premiere episode.


Car, athlete, and digital brain images courtesy of

Kevin Casey writes about technology and business for a wide variety of publications and companies. He won an Azbee Award, given by the American Society of Business Publication Editors, for his InformationWeek story, “Are You Too Old for IT?” He’s also a former community choice honoree in the Small Business Influencer Awards. View posts by .

Interested in writing for New Relic Blog? Send us a pitch!