Welcome to This Week in Modern Software, or TWiMS, our weekly analysis of the most interesting and important news, stories, and events in the world of modern software and analytics.
This week, our top story concerns Internet Explorer’s fall from its long-time spot atop the browser rankings.
TWiMS Top Story:
Internet Explorer, the Second-Place Web Browser—ZDNet
What it’s about: As Windows 10 approaches its first birthday, Microsoft has already started celebrating, announcing this week that the operating system now runs on 300 million active devices worldwide. (Interestingly, nowhere in its blog post announcing the 300-million mark does Microsoft refer to Windows 10 as an OS; rather, it’s an “online service.”) But there may not have been many high-fives in Redmond this week about another Microsoft milestone: For the first time since the heady days of the first dot-com bubble, Internet Explorer is no longer the world’s most popular desktop web browser. NetMarketShare data for April showed Google’s Chrome garnering 41.7% of the browser market, edging out IE, which dropped to 41.4%. No one else is really close: As ZDNet’s Steven J. Vaughan-Nichols writes, the “real loser” in the latest NetMarketShare data is Mozilla’s Firefox, once the most-promising challenger to IE’s dominance, which fell to 9.8%. Multiple outlets pointed out that, by other web-tracking counts, IE has actually not been the top browser for a while; StatCounter has had Chrome in the lead since 2012. (Different data and methodologies account for the differences.) But NetMarketShare’s findings are the final straw, as it is one of the most commonly cited sources of browser usage.
Why you should care: Here’s a brief list of things that didn’t exist the last time that IE wasn’t the top dog in the browser market: Facebook, the iPhone, public shares of Google, Skype, Minecraft—you get the idea. When it first ascended to the throne, IE’s competition was Netscape. With Microsoft prioritizing its newer Edge browser in Windows 10 and for the foreseeable future, the end of IE’s dominance is likely final. (Vaughn-Nichols points out, though, that NetMarketShare counts Edge usage as IE.) With the rise of mobile, chatbots, and other massive changes, the browser game isn’t as high-stakes as it once was, but this is still a major transition for the web and for the people who build websites and services.
- Google Chrome Sneaks Past Internet Explorer to Become Top Browser—CNET
- Google Chrome Dethrones Internet Explorer to Become Undisputed Browser Champion—PCWorld
- Google Chrome Edges Microsoft’s IE as Top Browser—USA TODAY
- Windows 10 Now on 300 Million Active Devices—Free Upgrade Offer to End Soon—Windows Experience Blog
What it’s about: It is apparently time to add another business to the ever-growing list of industries disrupted by software and technology: gas stations. Bloomberg this week profiles a group of new startups that aspire to become the “Uber for gasoline.” After a few taps of an app on your mobile device, such companies as Filld, WeFuel, Yoshi, Purple, and Booster Fuels will deliver fuel directly to your vehicle’s tank at the office, at home, wherever—no need to pull into a gas station and fill up at the pump. Of course, the Bloomberg story also highlights an inherent potential risk in software disruption: driving around with hundreds of gallons of gasoline on the back of a retrofitted pickup truck isn’t always legal.
Why you should care: Disruption can sometimes require figuring out some pretty specific and occasionally mind-numbing details. In this case, that includes things like worrying about whether the customer left their gas flap open for the delivery driver. The story also offers an in-your-face reminder of the hubris that often accompanies the “disrupt everything” mindset. A spokesperson for the Los Angeles County Fire Department told Bloomberg that this form of gasoline delivery wasn’t currently allowable under the fire code. Bloomberg relayed that to Bruno Uzzan, CEO of the Los Angeles-based fuel-delivery startup Purple, and asked if the company would stop delivering in the city. Uzzan’s response: “No, why should we?” Leaders of other companies in the field had similarly tone-deaf responses to their legal and regulatory challenges, taking the position that “If it was illegal, they would have and should have told us many months ago.” Hey, that shoot-first-and-ask-questions-later approach has worked pretty well for successful disruptors like Uber.
- Gas Delivery Apps Are New On-Demand Fad—USA Today
- Killing an American Icon: Fuel Delivery Start-Ups Could Downsize U.S. Gas Stations—Silicon Angle
IBM Makes Quantum Computing Available in the Cloud—Computerworld
What it’s about: How do you make an incredibly complex and prohibitively expensive next-generation technology—in this case, the still-experimental world of quantum computing—accessible to the masses? Put in the cloud, of course. That’s what IBM did this week, announcing that it would make its five-qubit quantum processor available via the IBM Cloud on any desktop or mobile device. (An IBM exec tells Computerworld that it will give the cloud service away for free.) “With Moore’s Law running out of steam,” IBM said in its announcement, “quantum computing will be among the technologies that could usher in a new era of innovation across industries.”
Why you should care: As multiple observers noted, quantum computing has so far been the purview of a select group of sophisticated researchers and deep-pocketed organizations—like IBM—able to experiment with the technology. “The birth of quantum cloud computing,” as IBM is calling it, promises to open up the technology to a much larger range of researchers, developers, and others who could potentially achieve new quantum breakthroughs. Critically, making its quantum processor available to the public could help solve another key obstacle. As IDC analyst Earl Joseph tells Computerworld, actually programming a quantum computer is a major challenge: “This experiment provides the opportunity for a large group of people to start to learn how to program quantum computers, which will help to develop ways to use this new type of technology.”
- IBM Wants Everyone to Try a Quantum Computer—The New York Times
- IBM Is Now Letting Anyone Play With Its Quantum Computer—WIRED
- Moore’s Law Running Out of Room, Tech Looks for a Successor—The New York Times
What it’s about: You can make a lot of money in tech, but you already knew that, right? But where can you make the most money? If compensation is at or near the top of your priority list when job hunting, CIO’s roundup of the best-paying firms in tech may be worth a peek. The slideshow lists the top 10 technology employers by median total compensation, based on data collected by job search site Glassdoor. While lists like this shouldn’t be treated as scripture, they can be illuminating: some of the company names might make you say, “Wait … who?”
Why you should care: Yeah, Google and Facebook make the list, but the top slot went to Juniper Networks, with an average base salary of $147,000 and median total compensation of $157,000. On the other hand, you won’t find some of the biggest or trendiest names in tech here—no Apple, Uber, or Dropbox. Instead, you’ll find firms like Guidewire, which makes software for the property and casualty insurance industry, and ranks fifth with a median total compensation of $150,020 per year. Ditto electronics design firm Cadence, which came in at #6—one spot ahead of Facebook—with a median total compensation of $150,101 annually. Perhaps the unlikeliest “startup” rounds out the list at #10: Walmart. More specifically, Walmart eCommerce, the retail giant’s technology division based in San Bruno, Calif., which includes its @WalmartLabs unit. Take heed: The trendiest names might not always be the best-paying employers.
What it’s about: What once was widely considered a lark is now all but official: Donald Trump will be the Republican nominee for president of the United States. We’re not here to talk politics, however, but data. Even in our increasingly data-driven world, with more powerful analytics tools than ever, pretty much no one outside of Trump’s camp predicted his success. Nate Silver, whose claim to data fame stems in large part from the accuracy of his analytics-based predictions for the 2008 and 2012 elections, was particularly off target: Silver and his site FiveThirtyEight wrote off Trump’s candidacy multiple times, and he conceded this week that “we basically got the Republican race wrong.” At the same time, as WIRED and others note, Trump bested candidates like Ted Cruz, whose campaigns invested much more heavily in data analytics.
Why you should care: Data can be enormously valuable, but that doesn’t mean it’s infallible. Trump’s prediction-defying triumph is a reminder that being data-driven doesn’t automatically let you predict the future, and analyzing complex human behavior is far from easy. It can help to incorporate other signals, including human thought, perception, and good old-fashioned gut instinct. On the other hand, it would be a mistake to believe that Trump’s triumph means that data analytics don’t matter. Alex Lundry, who ran data science operations for Jeb Bush and Mitt Romney, tells WIRED: “[Trump] is a singularly unique phenomenon in United States politics, and future candidates that think they can run a Trump-like campaign will be sadly mistaken. That includes the use of data and analytics.”
- The Republican Horse Race Is Over, and Journalism Lost—The New York Times
- Why Republican Voters Decided on Trump—FiveThirtyEight
- Nate Silver: ‘We Basically Got the Republican Race Wrong’—Business Insider
- Data Journalism Didn’t Fail: Nate Silver Pushes Back—Salon
- 4 Reasons Data Analytics Endeavors Don’t Succeed—Government Technology
- This One Number Foretold Donald Trump’s Rise—Yahoo Finance
What Happens When Your Tech Predictions Tank—The Atlantic
What it’s about: Political pundits and data analysts weren’t the only folks revisiting predictions gone wrong this week. But The Atlantic looked a lot farther back with a retrospective on the 1980s forecasts of former New York Times columnist Erik Sandberg-Diment. As writer Adrienne LaFrance notes, Sandberg-Diment is something of a forefather of the current tech press and analysis scene, from back when floppy disks and dot-matrix printers still ruled. And he’s getting something of an encore online today, thanks to sites like Reddit, where people can guffaw over predictions that turned out to be way, way wrong. Consider Sandberg-Diment’s 1985 declaration that laptops would never achieve widespread use: “People don’t want to lug a computer with them to the beach or on a train to while away hours they would rather spend reading the sports or business section of the newspaper.”
Why you should care: The tech industry loves to obsess over predictions, and you’ll be hard-pressed not to chuckle at more than a few swing-and-miss moments, such as when Sandberg-Diment dismissed email as no more efficient than snail mail, also in 1985. But Sandberg-Diment was also quite often right. Among his bulls-eyes: He saw the ubiquitous computer mouse coming as well as the networked office, and he gave a particularly early and prescient forecast of the future bane of the internet, the pop-up ad. LaFrance’s interview also takes a philosophical turn: Sandberg-Diment tells her that he’s largely disappointed by the digital age, noting isolation and a decrease in human interaction as negative consequences of people constantly tethered to their phones and other devices. He does cop to being wrong about the laptop, too—though he later in the interview insists that his prediction is now finally coming true, thanks to the rise of mobile and declining PC sales. That might be the best part about technology predictions: If you wait long enough, just about anything can happen. Recommended reading.
Want to suggest something that we should cover in the next edition of TWiMS? Email us at email@example.com.
Tune In to the Future
Can’t get enough modern software news and commentary? Be sure to check out our 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 episode 12 or subscribe on iTunes.