Google strengthens Android

Intel may have cut ties with Android on smartphones and tablets, but the company’s partnership with Google on Android for the internet of things is growing stronger. Google’s Android Things, a slimmed down version of Android for smart devices, will be coming to Intel’s Joule 570x computer board. The combination will allow makers to cook up Android-based gadgets or smart devices for use in home, retail, or industrial settings. The Intel board adds a lot of processing and graphics muscle to projects. With 4K graphics capabilities, the 570x is good for devices with screens or computer vision, like robots and drones. Intel demonstrated a bartending robot that used the board at its annual trade show last year. A standout feature in Joule 570x is a RealSense 3D depth camera, which can recognize objects and measure distances. The board has an Atom T5700 processor, 4GB of LPDDR4 RAM, 16GB of storage, and 802.11ac Wi-Fi. Right now, only three boards — Raspberry Pi, Intel’s Edison, and NXP’s Pico i.MX6UL — support Android Things. On paper, the Joule 570x has better specifications than the Raspberry Pi 3. But it could also be overkill for Android Things, which can also work on work on sensor devices that require only basic processors like Quark on Intel’s Edison. Putting Android Things in more devices will help Google effectively compete with Amazon’s Alexa, the voice-assistant technology that is being used in more gadgets and home appliances. Last week, Google hinted that makers will be able to build devices with the company’s machine-learning technologies like voice and speech recognition, which are mainly based in the cloud. Google’s will bring its TensorFlow APIs (application programming interfaces) to makers later this year. Android Things is still in preview, and a final version of the OS hasn’t been released. The OS previously went by the name Project Brillo, and a release date for a final version of the OS isn’t available. Android Things is also one way for Google battle Microsoft’s Windows 10 IoT Core, Ubuntu’s Snappy Core, and other Linux-based embedded OSes. Billions of IoT devices will ship in the coming years, and there’s an OS battle raging in the area, much like the OS competition in the PC and server markets. Source:

Microsoft’s mysterious ‘Windows Cloud’ could be the second coming of Windows RT

Windows Cloud: That name has appeared in system files deep within some of the most recent Windows 10 Insider builds. While a few experts guess it could be a new version of Windows, what it actually is remains a mystery. As far as evidence goes, it’s pretty slim pickings. Two names, “Windows Cloud” and “Windows Cloud N,” appear in a list of Windows versions as early as the recent Windows 10 Insider Build 15002, as originally reported by the Walking Cat Twitter account. (The “N” designation likely refers to a version specifically designed for European countries.) ZDNet’s Mary Jo Foley, however,unearthed one other key bit of information: Instead of an operating system that lives in the cloud, as the name suggests, “Windows Cloud” is actually an operating system that can only run Microsoft’s own UWP apps, downloaded from the Windows Store. Foley draws the obvious conclusion: Windows Cloud is essentially the second coming of Microsoft’s unpopular Windows RT. Windows RT was the operating system that powered the original Surface tablet as well as the Surface 2. Users criticized it for its inability to run anything but a limited number of apps directly from the Store. (At the time, Microsoft’s “universal” apps were in their infancy, while the vast majority of Windows applications were coded for the Win32 environment.) Though Windows RT had its fans, most customers quickly turned to the more advanced, Windows 8-powered Surface Pro tablets, and the Surface 2 quietly died  in 2015. Windows Cloud, though, may actually prove to be useful in specific applications, including schools. Foley reasons that Microsoft developed Windows Cloud to fend off the wave of Chromebooks sweeping across schools. Over half of American classrooms use Chromebooks, according to a Futuresource study released last fall, and their simplicity has made them attractive to school administrators. That hasn’t gone down well with Microsoft, which is working to recapture the classroom for Windows. Microsoft believes that its recent Intune for Education device and app management software is an important part of that. Further locking down those PCs with a dedicated OS would make a Windows 10 PC even more attractive, while offering digital inking and other features that Chromebooks lack. So is Windows Cloud actually Microsoft’s bid to take over the classroom? Microsoft declined to comment, so we’ll have to wait and see. Why this matters: Because consumers soundly rejected Windows RT, it stands to reason that Windows Cloud will be a pretty niche offering—assuming all the reporting about Windows Cloud is accurate, of course. One thing to keep in mind: While we all value Windows’ complexity for general-purpose computing, the capability to lock it down to a single app or focus can be extremely useful. That’s why several versions of Windows offer “kiosk mode,” where Windows 10 can be locked down to a single app. The new Intune locks down Internet access during test mode. It sounds like Windows Cloud might offer just a bit more flexibility while still maintaining control. Source:

Samsung security camera hacks

The saga of hacks made on Samsung’s popular SmartCam security cameras are a perfect illustration of why your network defense must start with a well-managed firewall. It has become impossible to rely on IoT device makers to create completely secure devices, and unworkable to have to keep them all up to date even when patches are provided. So while the history of these hacks isn’t unique, it does provide a good case study. Devices start out problematic and usually get worse Most IoT devices use a customized version of an off-the-shelf OS distribution, often Android or Linux. So they immediately come with whatever problems those OSes have. But once the software has been modified to fit the needs of each particular device, the benefit of any standardized platform updates is lost (although even those would assume that the device maker supported an automatic update mechanism). So each manufacturer needs to stay on top of all the exploits of the platform, as well as of its own software, and address them in a timely fashion. Some vendors manage to do that. But it only takes one hacked device to start to cause problems on your network. In the case of Samsung SmartCam cameras, the original hack was revealed in August 2014, but from what I can tell Samsung didn’t distribute a patch until 2016 (which is when it was pushed in a firmware update to one of my SmartCams, among many others). It’d be one thing if Samsung was unique in this lag time, but it isn’t. Plenty of other IoT makers aren’t any faster. Worse yet, in the case of some of the commercial cameras exploited as part of recent DDoS attacks, they don’t have a simple way to distribute patches. When I did a roundup of security cameras recently, none of them impressed me as being rock-solid on security. The problem with patches Samsung’s patch for the original exploit illustrated two typical problems. First, it broke LAN access to the cameras, so those relying on access for local recording and streaming suddenly found their cameras unusable after the firmware update. It was not lost on users that Samsung rolled out a fee-based cloud DVR service on exactly the same day. Many users elected not to install the new firmware — and live with the exploit — rather than cripple their systems. Second, the patch wasn’t very good. It left plenty of stubs of the services there. On the bright side, this meant that enterprising developers figured out how to restore streaming functionality. On the darker side, it meant that it was only a matter of time before the original hackers hacked the patched cameras. This week showed how a hacker could alter the camera’s passwords and have it run arbitrary code. Clickbait headline writers need to take a chill pill All you need to do is search the web for articles on webcam hacks and you’d think zombies are about to invade your house and kidnap your pets and children. Only some of the articles bother to point out that almost all of these hacks (including the ones on the Samsung SmartCams) require the ability to get to the device directly using its IP address. In almost all residential and commercial networks, that address is local, probably dynamic, and sits behind a firewall whose job it is not to let hackers in. So yes, if someone was on your LAN, or hacked your Wi-Fi, they could potentially hack your security camera or your future toaster. That would probably be the least of your problems, though. How many of your other computers, tablets and phones would they be likely to go after first? As to the sensational headlines about zillions of security cameras being exploited for DDoS attacks, those are almost entirely ones found in industrial installations (think remote locations, for example) where they are directly Internet-addressable. They also are primarily low-end versions sold in developing countries. I’m the last person to downplay the importance of patching security flaws in IoT devices, but at the same time we need to make sure that users realize they need to be responsible for protecting their entire home network. By the time hackers get to where they can start poking around your local devices, you are already in trouble. One obvious place to start is to make sure whatever router you use has firewall software that is kept up to date. And think carefully any time you open up a port on it to the outside or enable port forwarding. Source:

Mozilla’s new logo is 20 years out of date

Mozilla, the nonprofit organization behind Firefox, announced its new “brand experience” and visual guidelines. I’m not going to argue that branding doesn’t matter — after all, people love to argue that marketing doesn’t matter, yet decades of research into consumer buying habits and how people respond to advertising have shown that it absolutely does. Mozilla’s share of the browser market has slumped badly since 2010, and the organization’s efforts to fix that problem should touch on every aspect of the Firefox experience. It is therefore absolutely proper to consider how branding impacts people’s perceptions of Mozilla and Firefox, and to make design changes and updates where necessary. That said, this is not the design I’d have personally gone with. Mozilla’s new logo is written as Moz://a, with the “ill” replaced with a colon and two slashes, just as displayed in a conventional URL. Mozilla’s creative director Tim Murray explained the decision to Wired: “Because it has a portion of URL embedded in the middle of the logo, you know this must be some kind of internet company.” I genuinely don’t want to sound insulting; I have no doubt that Tim Murray and his team worked hard for months to sort out various logo options and designs. Instead of just unveiling a logo to the world once it had been finalized, Mozilla sought public commentary at every stage of the process. Users weren’t allowed to vote on options, but they were encouraged to comment and leave their thoughts on why the various proposed logos did or did not work. You can read more about this process here, or watch the video below for Mozilla’s own discussion of why it went with this design. I respect that Mozilla wanted a logo that was easy for people to mash up and use. I appreciate that it’s actually a logo people can type without resorting to non-standard fonts or trying to embed images mid-story. But at the same time, it’s the kind of logo that screams “1990s Internet company,” and hearkens back to an era when people thought AOL was the Internet, rather than a walled garden of content and information that itself relied on the Internet to function. It reminds me of nothing so much as the era when people thought you needed a .com in your brand name to be cool. And the worst excesses of the dot-com era, and its emphasis on brands and clicks at the expense of income and profit, aren’t really something I’d want to associate my brand with, even 20 years after the fact. What makes the situation slightly more awkward is the fact that Firefox seems to be the only browser left standing that actually displays the http:// in the first place. Edge and Chrome both hide it (at least by default). So does Vivaldi. Given Chrome’s market share, it’s not clear how many people will even associate the :// with a URL. If Mozilla wanted to cater to the faithful, that’s its prerogative, but the new logo may not do much to build recognition of what the foundation is all about. What Firefox means to me I use Firefox on a daily basis, along with multiple other browsers, because it plays fairly nice with ExtremeTech’s content management system, or CMS. Unfortunately, my daily experience with Firefox goes something like this: Start Firefox Open tabs, browse web *Time passes* Observe that my entire system is lagging Open Task Manager Note that Firefox is using a constant 5-8% CPU (roughly one core on a six-core / 12 thread system) and 1600 – 2200MB of RAM Kill Firefox, restore previous session Observe Firefox now using 1-3% CPU and 400-800MB of RAM Rinse, wash, and repeat, often multiple times a day. I’ve shut off every extension and add-on, including all ad blockers. I’ve used the browser in safe mode. I even tried manually enabling the new multi-threaded engine. Closing all open tabs does not solve the problem; it’s only resolved by a full restart of the browser. I’m aware that the Firefox team has done a great deal of work to firm up the nuts and bolts of their browser, improve its performance, cut its RAM usage, and build a better product. Despite this, it’s the only browser I use that needs multiple full restarts on a daily basis. Heck, I’m willing to believe that some of the reason I run into these issues may be poor coding practices at the sites themselves, but that doesn’t change the fact that Chrome and Chromium-derived browsers like Vivaldi don’t run into problems nearly as often, for me, as Firefox does. (Edge has its own issues and problems, but Edge is also less than two years old). If Firefox wants to win back users, a new brand may help a little — but what would really help is fixing some of the problems that drove those users away in the first place. Give people like me reasons to write reviews praising Firefox for new features and improvements. I’d take that over a new logo, any day. Source:  

All Chromebooks released in 2017 and later will support Android apps

Google began adding support for the Play Store to Chromebooks last spring, but the rollout has been very, very slow. At first, only a few ARM-powered Chromebooks had support, then Google’s own Pixel 2 was added. Now, many recent Chromebooks have gotten the Play Store, at least in the developer channel. Google is going a step further today, announcing that all Chromebooks that debut in 2017 and beyond will include support for running Android apps. Chromebooks are designed to rely almost exclusively on web apps and various online tools. Almost everything is accessed via Chrome, which exists inside a stripped down Linux environment. This is great in a lot of ways — it keeps the machine lightweight, secure, and fast; plus all your data is backed up online. If you lose that particular device or need to use another Chromebook for a short time, you can simply log in and have all your stuff ready to go. This approach to computing is usually convenient, but it’s limiting. Something as simple as editing documents offline or playing a game can become a challenge. Adding support for Android apps instantly gives Chromebooks a plethora of software to choose from. Granted, not all of it is optimized for a computer interface, but much of it works surprisingly well. The key is finding apps that have been optimized for Chromebooks or just don’t rely on touchscreen-only interaction. Some useful apps that run well include the free Microsoft Office suite and VLC media player. If you’ve already used Android with your account, you’ll have access to a ton of content as all your past purchases will be available on the Chromebook. Several new Chromebooks were announced at CES this year, including two models from Acer and a few from Samsung. All those machines will ship with the Play Store pre-installed and ready to go. That means it will be part of the stable channel of Chrome OS. You still need to be on beta or dev to get Android apps on many existing Chromebooks. You can check the Chromium Project page for details on which devices have support. Google guarantees Chrome OS updates for five years from a device’s launch, but that doesn’t necessarily mean all the Android apps that come out during that time with work. Google is also rumored to be working on a different project called Andromeda that will merge parts of Chrome OS into Android for a more compelling large-screen experience. This is still unofficial, so it might never come to be. Chromebooks are a hit, though, so Google will continue with rolling out Android apps in 2017. Source:  

A spy in Windows 10

When I first saw Mr. Spock talking to the Enterprise’s computer, I thought it was so cool. I still do. But the more I look at Cortana, Windows 10’s inherent virtual assistant, the more creeped out I get. Let’s start with Cortana’s fundamental lust for your data. When it’s working as your virtual assistant it’s collecting your every keystroke and spoken syllable. It does this so it can be more helpful to you. If you don’t like that, well, you’ve got more problems than just Cortana. Google Now and Apple Siri do the same things. And it’s not just virtual assistants; every cloud-based software as a service (SaaS) does this to one degree or another — Google Docs, Office 365, whatever. But Cortana doesn’t stop there. With the recently released Windows 10 Anniversary Update, hereafter Windows 10 SP1, you can’t shut Cortana off. Maybe you don’t mind Microsoft listening to your every word so it can catch when you say, “Hey, Cortana.” I do. Yes, I want the coolness factor of being able to talk to my computer. But I want the reassurance that it’s not listening when I don’t need it to be. I want a simple on/off switch. Windows 10 SP1 doesn’t have one. This is interesting, though: Windows 10 Education does. Microsoft apparently is willing to respect the privacy of students. The rest of us? Not so much. What you can do in Windows 10 SP1 is cripple Cortana when you install the operating system. But Cortana then becomes no more than a front end to Microsoft’s Bing search engine. You lose the ability to talk to your computer. You’ll no longer be able to tell Windows 10 to get you an Uber or tell you how the Chicago Cubs did today. If you’re anti-Cortana, don’t install Windows 10 SP1 with “Express settings.” Instead, follow the steps described by Jared Newman in PC World. You will make Windows 10 less useful but a lot more private. If you’re not comfortable with Cortana collecting your contacts, location, calendar data, and text and email content and communication history, you’ll want to do this. Don’t, though, if you want the full Cortana experience and you don’t mind Microsoft collecting everything except your car keys. And maybe you don’t. Many of us are reconciled to the mantra of the internet economy: “If you’re not paying for it, you are the product.” Companies such as Facebook and Google give all their free social and search goodies in return for our web history, which they then transform into cash with targeted advertising. And as for Microsoft, it makes a point of saying Cortana doesn’t do that. Why do I not feel reassured? Now that I think of it, though, you can’t (easily) get Windows 10 for free anymore. So you get to pay Microsoft with both cash ($199.99 for Windows 10 Pro) and your data. Oh boy! Microsoft also claims that Windows 10 SP1 is safer than ever, which I find even less assuring than the promise not to exploit all that Cortana data. Think about this: You can use Cortana from the lock screen. That’s right; Cortana is active and listening to when your PC is locked. Well, it’s supposed to be locked, but if it’s able to listen, how locked down is it, really? Not very! Microsoft calls this a feature that gives you the ability to ask your PC simple questions without logging in. But I call anything that lets me input data into a PC without being logged into it a bug. It’s a security hole begging to be exploited. Windows, which God knows has had more than enough security problems, now has a new attack surface. Fortunately, you can fix this one easily. Just open Cortana’s Settings and turn off the “Use Cortana even when my device is locked.” By the way, Microsoft always claims that Windows is new and improved and more secure than ever. And yet, if you look at any significant Windows patch report, you will notice that every major bug affects every supported version of Windows. Shouldn’t the new and improved Windows 10 be immune from the bugs that affect Windows 7, 8 and 8.1? It’s funny how they seem to slug every version of Windows.I like Microsoft a lot more than I used to, but I’m not ready to trust it with everything and the virtual kitchen sink. So I followed Newman’sadvice when installing the OS. I’m afraid I will never be as cool as Spock. I should note that, if your distrust of Microsoft exceeds mine, you can rip into your operating system’s guts and totally disable Cortana. You need to beware, though, because it involves going in deep, to places where it’s really way too easy to foul up Windows. In killing Cortana, you could end up seeing a lot more Windows crashes. In Windows 10 Pro, you type gpedit.msc into the Start menu. Head down to Computer Configuration > Administrative Templates > Windows Components > Search. Once there, double-click on Allow Cortana to toggle it to Disable Cortana. Log off and back on, and you’re done. In Windows 10 Home, open the registry with regedit and head to HKEY_LOCAL_MACHINE\SOFTWARE\Policies\Microsoft\Windows\Windows Search Next, right-click the Windows Search folder and choose New > DWORD (32-bit) Value. Name this new DWORD AllowCortana and set it at 0. Now log off and reboot your computer. Let me reiterate: If any of that sounds mysterious, don’t do it. And, you know, why should you have to? Why can’t Microsoft just make it easy to turn off Cortana? I’d appreciate it. Source:

Yahoo reveals newly-discovered hack of 1 billion accounts

Yahoo today revealed it had recently identified a new system breach that occurred in August 2013 and involved data associated with more than one billion user accounts. The company said it believed the incident was separate from the breach it disclosed in September, when information associated with at least 500 million user accounts was stolen from its network in 2014. Yahoo, which is being acquired by Verizon, said an unauthorized third party had stolen the data in the latest breach, and that it was working closely with law enforcement. “We have not been able to identify the intrusion associated with this theft,” CISO Bob Lord said in a statement. The company said the stolen user account information may have included names, email addresses, telephone numbers, dates of birth, hashed passwords and, in some cases, encrypted or unencrypted security questions and answers. “The investigation indicates that the stolen information did not include passwords in clear text, payment card data, or bank account information,” Lord said. “Payment card data and bank account information are not stored in the system the company believes was affected.” It is urging Yahoo users to change their passwords, and has invalidated unencrypted security questions and answers so they can’t be used to access an account. Source:

‘The Glass Room’ In SoHo Will Make You Want To Burn Your iPhone

When we walk around this city, we bleed data—invisible pieces of information about us that companies can buy and sell; government officials can track; and your employers, friends, and enemies can manipulate at will. We know all this, but it’s hard to visualize just how much of our information is leaking out into the world. Enter The Glass Room, a stark art exhibition on Mulberry Street that serves as a chilling “investigation and intervention into our online lives,” now on view through December 18th. Set up like a tech store, The Glass Room—spearheaded by the Berlin-based Tactical Technology Collective and Mozilla—features a variety of art pieces dedicated to visualizing how data is sourced from the unsuspecting public, and how sinister its use can be. One piece, “Online Shopping Center” by Sam Levigne, mimics Amazon’s ‘predictive shopping’ algorithm—which pre-purchases items for you based on patterns from your account and from others like you—by using a brain-scanner to predict what you’d like to purchase when you think about death. Another piece, “Unfitbit,” by Surya Mattu and Tega Brain, finds ways to trick the FitBit device into thinking you’re active when you’re not—this was dreamed up after the artists realized health insurers could use FitBit data to determine how much to insure you for, and that employers who give their employees FitBits can see when their workers are at their desks, and for how long. Then there’s “Smell Dating,” by Tega Brain and Sam Levigne, a dating service of sorts that drops swiping in favor of pheromones. Participants wear a white T-shirt for three days without washing it, then send it to the service—in return, they receive a selection of pre-worn T-shirts and pick dates based on which scents appeal to them (this is a real thing!) And “The Library of Missing Data Sets,” by Mimi Onuoha, visualizes “a list of data sets that are missing – datasets that do not exist, but should.” Offerings include Hillary Clinton’s 31,000 deleted personal emails, Donald Trump’s tax returns, and reasons for the existence of dark matter. The art pieces are fun to play with, but things get scarier a little deeper into the exhibition. One section of The Glass Room features a series of real apps available on the market, like Sickweather, in which people who have colds or feel unwell can mark their locations to warn others walking by. There’s a tracker people can attach to their aging parents’ medication bottles to make sure they’re taking their pills each day. And there’s a section on the very controversial predictive policing, where visitors can play with algorithms that police departments all over the country use to proactively determine where crimes might take place, though some justice advocates believe these algorithms only amplify existing police bias. The information you can peruse in The Glass Room isn’t necessarily new, but it’s presented in a way that’s more digestible for those of us who don’t usually deal in data. “When we talk about data, we don’t really know what we’re talking about,” Heinrik Chulu, a researcher with the Tactical Technology Collective, told Gothamist. “The goal here is to take the intangible nature of the Internet, and figure out how to understand the nuances and complexities of everything going on.” And the visualizations are helpful. For instance, though we know the LinkNYC public wifi system makes it easier for the NYPD and/or other officials to tap into users’ web activity, Glass Room artists show just how much of (and how easily) that data is made available—they put up four antennae outside the show’s space on Mulberry Street, and visitors can watch data belonging to unsuspecting passersby ping on a big screen. “You can see the make of each phone, what wifi they’re connecting to,” Chulu pointed out, noting someone could track people with a laptop using the information available. “If you walk by a kiosk, they can see your phone and your network. They could track you around the city. If you know someone’s identifiers, you can ping them anytime they enter the room.” But The Glass Room doesn’t just frighten you into trashing your cell phone and throwing your laptop out the window. It also attempts to offer solutions to help you cut back on some of that data you’re dribbling out all the time. Free workshops like “What The Facebook?” show you how much information you’re sharing to the public on social media, and how to cut back on it. Another, “De-Googlize Your Life,” will point out how much data you feed to Google on Gmail, Google Docs, Google search and other services, and how to minimize your trace (you can also take a look at your data shadow online). The Glass Room also hands out free 8-Day Data Detox Kits, which will help you determine which apps bleed less of your information, though it might cost you Twitter, Uber, and Google. These tips might not be enough to get you off the grid, but it’s useful to know exactly how much of you is out there on the Interwebs, especially when facing whatever dystopian future awaits in the coming years. The Glass Room is at 201 Mulberry Street through December 18th. The gallery is open from 12 to 8 p.m. daily, and is free to visit. You can visit their website at Source:

PHP 7 user migrations are well underway

PHP users are slowly but surely migrating over to PHP 7, the upgrade to the server-side web development language that came out a year ago. But issues like incompatibility and IT user policy restrictions are stalling their progress. A recent survey of nearly 1,300 PHP users taken by PHP tools producer Zend found that nearly 20 percent had already made the move, 21 percent were in progress, 15 percent planned to move in the next six months, and 17 percent planned to move in less than one year. But 14 percent were putting off the migration for more than a year, and another 14 percent were not planning to move at all. Barriers to migration include incompatibility with custom code (nearly 31 percent), lack of support for third-party frameworks and libraries (17), lack of available extensions (11), IT policy restrictions (11), and customer requirements (10). When respondents were asked about their security concerns related to PHP, they mentioned vulnerabilities in custom code (48 percent), open source or third-party components (30), infrastructure (12.2), and the PHP stack (6.5). The survey also found, unsurprisingly, that about 93 percent of users were deploying PHP for web applications. It’s also being used for services or APIs (63 percent), CMS software (48), internal business applications (46 percent), e-commerce (35 percent), and as a back end for mobile apps (34.35). Respondents encountered few issues when deploying PHP applications into production, with 25 percent saying they almost never had problems and about 42 percent saying they had problems fewer than one-quarter of the time. About 9 percent said they had problems with almost every deployment or with more than half of deployments. The survey asked about the problem resolution and maintenance vs. developing new functionalities equation, and it found imbalances in both directions. Slightly less than 35 percent of respondents said they spent three-quarters of their time on new functionality and the rest fixing problems, and almost 26 percent spent the vast majority of their time on new functionality. For about 25 percent, the breakdown was half and half. Zend also inquired about the frequency of code deployments. Nearly 32 percent deploy several times a week, while 14 percent deploy several times per day. Slightly more than 21 percent deploy weekly, and 27 percent opt for one to three times a month for deployments. Just 6.26 percent deploy one to three times a year. Source:

Say goodbye to the MS-DOS command prompt

My very first technology article, back in 1987, was about MS-DOS 3.30. Almost 30 years later, I’m still writing, but the last bit of MS-DOS, cmd.exe — the command prompt — is on its way out the door. It’s quite possible that you have been using Microsoft Windows for years — decades, even — without realizing that there’s a direct line to Microsoft’s earliest operating system or that an MS-DOS underpinning has carried over from one Windows version to another — less extensive with every revision, but still there nonetheless. Now we’re about to say goodbye to all of that. Interestingly, though, there was not always an MS-DOS from Microsoft, and it wasn’t even dubbed that at birth. The history is worth reviewing now that the end is nigh. Back in 1980, the ruling PC operating system was Digital Research’s CP/M for the z80 processor. At the same time, Tim Patterson created Quick and Dirty Operating System (QDOS). This was a CP/M clone with a better file system for the hot new processor of the day, the 8086. At the time, no one much cared. Until, that is, IBM decided to build an 8086-based PC. For this new gadget, IBM needed to settle on programming languages and an operating system. It could get the languages from a small independent software vendor called Microsoft, but where could it get an operating system? The obvious answer, which a 25-year-old Bill Gates seconded, was to go straight to the source: CP/M’s creator and Digital Research founder, Gary Kildall. What happened next depends on whom you believe. But whether Kildall was really out flying for fun when IBM came by to strike a deal for CP/M for the x86 or not, he didn’t meet with IBM, and they didn’t strike a deal. So IBM went back to Microsoft and asked it for help in finding an operating system. It just so happened that Paul Allen, Microsoft’s other co-founder, knew about QDOS. Microsoft subsequently bought QDOS for approximately $50,000 in 1981. Then, in short order, IBM made it one of the PC’s operating systems, Microsoft renamed QDOS to MS-DOS, and, crucially, it got IBM to agree that Microsoft could sell MS-DOS to other PC makers. That concession was the foundation on which Microsoft would build its empire. Late last month, in Windows 10 Preview Build 14791, the command prompt was put out to pasture. Dona Sarkar, head of the Windows Insider Program, wrote, “PowerShell is now the defacto command shell from File Explorer. It replaces Command Prompt (aka, cmd.exe).” That “defacto” suggests that it’s not all over for the command prompt. And it’s true that you can still opt out of the default by opening Settings > Personalization > Taskbar, and turning “Replace Command Prompt with Windows PowerShell in the menu when I right-click the Start button or press Windows key+X” to “Off.” But you might as well wave bye-bye to the old command prompt. Build 14791 isn’t just any beta. It’s the foundation for the Redstone 2 upgrade, a.k.a. Windows 10 SP2. This is the future of Windows 10, and it won’t include this oldest of Microsoft software relics. PowerShell, which just turned 10, was always going to be DOS’s replacement. It consists of a command-line shell and a .Net Framework-based scripting language. PowerShell was added to give server administrators fine control over Windows Server. Over time, it has become a powerful system management tool for both individual Windows workstations and servers. and its NT-twin brother, cmd.exe, were on their way out. They had a good run. A good way to understand how they held out for so long is to look at DOS as a house under constant renovation. First, all there was was the basic structure, the log cabin, if you will, of Microsoft operating systems. That log cabin was given a coat of paint, which is what Windows 1.0 amounted to — MS-DOS all the way, with a thin veneer of a GUI. Over time, Microsoft completely changed the façade in ways that made the old log cabin completely unrecognizable. With Windows NT in 1993, Windows started replacing the studs and joists as well. Over the years, Microsoft replaced more and more of MS-DOS’s braces and joints with more modern and reliable materials using improved construction methods. Today, after decades, the last pieces of the antique structure are finally being removed. All good things must come to an end. It’s way past time. Many security problems in Windows trace back to its reliance on long-antiquated software supports. Still, it’s been fun knowing you, MS-DOS. While you certainly annoyed the heck out of me at times, you were also very useful back in your day. I know many programmers and system administrators who got their start with you on IBM PCs and clones. So, goodbye and farewell. While few users even bothered to look at you these days, you helped launch the PC revolution. You won’t be forgotten. Author: Steven J. Vaughan-Nichols,