Google reportedly plans to ‘shame’ smartphone makers to fix one of the worst problems with Android

, Business Insider, 5/26/16

If you ask developers about their gripes with building apps for Android, the same problem comes up again and again: fragmentation.

On Android, Google’s mobile operating system, it is the smartphone manufacturers — not Google — who are responsible for pushing out software updates to users. This is in stark contrast to Apple’s mobile operating system, iOS, in which Apple can push out an update to every (compatible) iPhone in the world simultaneously. As a result, the Android market is intensely fragmented, with numerous versions of the OS out there.

This complicates development, as the majority of Android phones end up unable to use the latest features that Google introduces for app developers. And more worrying for users, it can leave millions of users open to hacking as security patches never reach them.

In short: Google could build an incredible new feature for the next version of Android, but unless your phone’s manufacturer, let’s say Samsung, for example, gets around to deciding to push the update to you, you will never see it.

Bloomberg reports that Google is renewing its efforts to improve the update procedures of Android manufacturers — and it is prepared to “shame” them if necessary.

The California tech giant has apparently ranked manufacturers according to “how up-to-date their handsets are, based on security patches and operating system versions,” and shared these lists among manufacturers. These lists are currently private, but Bloomberg says Google is considering making the list public to “shame” the manufacturers that are significantly worse than others.

Google did not immediately respond to Business Insider’s request for comment.

Google already publishes data on fragmentation on its Developer Dashboard. Just 7.5% of Android phones are running the most recent version of the OS, Marshmallow, which launched in October. And more than half are running KitKat (which came out in October 2013) or an even older version of the operating system.

android fragmentation may 2016Google

In contrast, 84% of active iOS devices (iPhones and iPads) are running iOS 9, the most recent version, which came out in September.

ios fragmentationApple

Share
Posted on May 26, 2016 at 7:33 am by lesliemanzara · Permalink · Leave a comment
In: Android, iOS, Mobile Technology

PayPal drives another nail into the Windows Phone and BlackBerry coffins

Mashable, 5/26/16

In February, PayPal launched a new version of its mobile app, 6.0, and it wants iPhone and Android users to upgrade to it until the end of June. Windows Phone, BlackBerry and Amazon Fire users, however, are getting a different message: Paypal is discontinuing the app on these platforms on June 30.

In an announcement post Wednesday, PayPal VP Joanna Lambert explained that Windows Phone users will still be able to access PayPal on the mobile web, through both Internet Explorer and Microsoft Edge. Likewise, Amazon Fire and BlackBerry users are advised to switch to the mobile web experience, though the post notes that on BlackBerry, you can continue to use the BBM app to send peer-to-peer PayPal payments.

“It was a difficult decision to no longer support the PayPal app on these mobile platforms, but we believe it’s the right thing to ensure we are investing our resources in creating the very best experiences for our customers. We remain committed to partnering with mobile device providers, and we apologize for any inconvenience this may cause our customers,” Lambert wrote.

It’s no surprise that support for BlackBerry and Windows Phone is dwindling. In a recent Gartner report, the two platforms had less than 1% market share between them (Windows had 0.7% and BlackBerry had 0.2%). In February, WhatsApp announced it would no longer support BlackBerry at the end of 2016. And several companies, including GoPro and Here dropped support for Windows Phone this year.

BlackBerry has been coping with its ever-shrinking market share by pivoting to Android; the company said in April it plans to release two Android phones this year. Microsoft’s mobile future is somewhat muddier, but with the company recently selling its feature phone business, and laying off 1,850 employees from its smartphone division, it looks as if Windows Phone’s fate is sealed.

Share
Posted on May 26, 2016 at 7:27 am by lesliemanzara · Permalink · Leave a comment
In: Android, Blackberry, iOS, Mobile Technology, WinPhone

The new Android N preview is more stable and updates itself

Chris Velazco, Engadget, 5/26/16

At last, a version you can confidently use as a daily driver.

Google surprised us all with an Android N developer preview two months before its annual I/O developer conference, and then it updated the software a few weeks ago with new performance-enhancing features. Now, as revealed during this morning’s keynote, there’s yet another update for you bold souls who crave bleeding-edge software.

While Google refers to Preview 3 on its developer site as an “incremental update,” there’s still plenty here to get excited about. Unlike the last two versions of the preview, Google says this third build is the first “beta-quality” candidate. In other words, you’ll run into fewer headaches if you try to use it as your daily driver on a Nexus 5X, 6P or other compatible devices. We haven’t played with the new N preview ourselves so we can’t confirm how stable it feels, but we’ll follow up with impressions as soon as we can.

Beyond that, this new update also brings with it a new software-updating scheme inspired, surprisingly enough, by Chromebooks. When an update is available, Android N can download the system image in the background and automatically install it the next time you reboot your phone.

“There’s no ‘Android is upgrading’ [pop-up], no delays,” VP of Android Engineering Dave Burke told us. “It’s just a really nice, seamless way to do it.”

It’s a smart move, especially with respect to security updates like the ones Nexus devices get every month. When it comes to those more timely, crucial security patches, Android N lets you know ahead of time that the update is going to happen, and will then just install it upon reboot. Meanwhile, you’ll be alerted to “dessert” updates — the big ones with the delicious new names — in the new suggestions section in settings, where you can choose to install it now or apply it later.

Share
Posted on May 26, 2016 at 7:23 am by lesliemanzara · Permalink · Leave a comment
In: Android, Mobile Technology

Android N gets a high-performance VR mode

Chris Velazco, Engadget, 5/26/16

This improved software could make virtual reality more accessible.

Even the most stubborn fence-sitters have to admit it: Virtual reality is worming its way into the fabric of our culture. Part of VR’s growth is due to the fact that we can now just strap phones to our faces and see other worlds, but that typically requires lots of work from phone makers like Samsung and LG to get their VR hardware and software working just right. The software side of that equation might have just gotten way easier to figure out, though, thanks to Google’s announcement of a VR mode for Android N.

In a nutshell, think of VR mode as a special high-performance setting for your smartphone. As Google’s VP of Engineer for Android Dave Burke puts it, a phone that meets certain criteria and conforms to “lots of rules” can declare itself “VR-ready.” From there, apps like virtual reality games and YouTube can fire up VR mode to unlock extra power and, crucially, drive down the latency between the moment you move the phone mounted to your head and the moment you see your environment change on-screen. Too much latency means you don’t get the sort of persistent, “I’m actually somewhere else” feeling as you spin around with a phone on your face.

We’re told getting to this point required work on just about every level of Android N’s software stack. When VR mode is on, it changes how Android interprets information coming from its sensors. Graphics have to be drawn differently too. Usually, Android uses a double graphics buffer where one set of images is drawn on the screen and then swapped with another set stored in memory. With VR mode on, though, apps can use a buffer where images are replaced much faster, on a scan line by scan line basis.

Oh, and curiously enough, Burke also mentioned in a blog post that developers would have access to an “exclusive CPU core for VR apps” — we’re still trying to figure out how that works. Still, it’s hard to argue with results. To put things in perspective, Samsung and Oculus managed to get motion-to-photon latency to under 20 milliseconds with the Gear VR. Google’s work seems to have matched that ultra-low latency: the Nexus 6P with VR mode inside Google Cardboard also comes in under that 20 millisecond bar, compared to the 100 millisecond latency without VR mode.

 

Share
Posted on May 26, 2016 at 7:17 am by lesliemanzara · Permalink · Leave a comment
In: Android, Mobile Technology · Tagged with: 

Opera’s getting serious about adblocking with its new VPN for iPhone and iPad

, The Next Web, 5/12/16

Opera has announced the launch of a new standalone VPN app for iPhone and iPad that promises to unlock geo-location restrictions and block ads.

The launch follows the introduction of the VPN service baked into its desktop browser last month, but brings it to a far larger potential audience; Opera only holds around two percent of the total desktop browser market.

The company says that more virtual locations will be added soon (it currently offers just five) and that it contains a built-in adblocker to stop ads and improve privacy when using Safari, Chrome and other apps.

Of course, while Opera says it will open up a world of non-georestricted content, the services that provide that content are cracking down on the workaround, so how long it will work for – if it does for you at all – is up for debate.

Share
Posted on May 12, 2016 at 8:37 am by lesliemanzara · Permalink · Leave a comment
In: iOS, Mobile Technology

U.S. Cellular: CDMA carriers face a VoLTE ‘gap’

, Fiercewireless, 5/12/16

While some U.S. carriers are moving ambitiously to deploy VoLTE — T-Mobile is the obvious example here — U.S. Cellular is taking a more measured approach and plans to begin to bring the technology to market next year.

And one major factor is that CDMA operators face a greater challenge with VoLTE than their GSM-centric counterparts, U.S. Cellular CEO Kenneth Meyers said.

“With respect to VoLTE, yes, it is a pretty well documented fact that the coverage on the voice side with CDMA is a wonderful product,” Meyers said this morning during a conference call with analysts. “And when you have built your network with that product and then try to overlay voice over LTE in the same cell sites, there’s an initial coverage — what I’ll call ‘gap,’ almost.”

U.S. Cellular observed that “gap” when it began testing VoLTE last year, Meyers said. Engineers are working to overcome the problems by moving equipment gear to the top of towers, among other strategies.

“And they’re using other tools so that when we do turn on VoLTE we are able to deliver to our customers the same and better services as they’re getting now. So that’s why we haven’t been rushing to it; we’ve been very deliberate with it, proceeding with caution, like we have in every other technology change in the past.”

U.S. Cellular sees VoLTE as a way to increase its roaming revenue from other carriers as the technology is expanded throughout the nation’s mobile networks. The carrier has also said it plans to spread its deployment of VoLTE over time to ease the strain on its quarterly capex budget.

T-Mobile has been the most ambitious U.S. carrier with VoLTE, routing “well over half” of all its calls through the technology, company executives said recently. Verizon and AT&T are also in various stages of VoLTE rollouts, while Sprint has opted to focus on technologies and strategies such as carrier aggregation and beamforming.

Share
Posted on May 12, 2016 at 8:34 am by lesliemanzara · Permalink · Leave a comment
In: Mobile Technology · Tagged with: , , , , , ,

Starting Thursday, the internet will get so much faster

, Business Insider, 4/28/16

CloudFlare, one of the startups that secretly runs the internet by handling as much of 10% of all web traffic, is making a little change to its core service that is opening the way to a vastly faster World Wide Web.

CloudFlare on Thursday becomes the first company to widely activate HTTP/2 Server Push, a technology that lets webpages and apps load as much as 15% faster, potentially shaving precious seconds without your having to do a thing.

That kind of speed will translate into a huge shift in the web and how apps are built.

For people with fast internet connections, it opens the door for all kinds of interactive content and apps that simply would have been too unwieldy in a pre-Server Push era. For those in the developing world, it means making the most out of a slower internet connection. Either way, it’s a big deal that will create some shockwaves.

“I don’t think the internet knows how this will be used,” CloudFlare CEO Matthew Prince says.

To be clear, this massive web speed boost won’t happen all at once, Prince says. It will take at least a year for it to come to full fruition. But HTTP/2 Server Push is a massive step in the right direction, as it encourages the entire tech industry to take notice of this technology.

To understand why the wait for a wider web, you need to know a little history.

A little history

When you visit a website, your browser talks to a server in language we call HTTP, or hypertext transfer protocol. That’s why there’s an “http://” in front of every site you visit — it’s to let your browser know that it’s a webpage you want and not some other kind of web service.

Even if you’re not on a browser, many apps use HTTP to talk to servers on the backend and get their data for display. If you’re online at all, ever, HTTP is unavoidable.

In 1997, HTTP 1.1 became the new — and still dominant — standard for serving webpages. It works fine (clearly, given that you’re reading this in a browser or an app). But it has struggled to keep pace with the modern internet, with photos, videos, and other interactive content presenting a bottleneck as it loads page elements one at a time. That’s why you sometimes have to wait for a video to finish loading before photos appear on a page, or vice versa.

Developers started to rely on their own gumption to get around HTTP’s limitations.

“Everybody had worked around its weaknesses with all kinds of tricks,” CloudFlare programmer (and Silicon Valley legend) John Graham-Cumming says.

http2 server push cloudflare

CloudFlareCloudFlare’s chart demonstrates the benefits of Server Push.

The most important example: Google built its own advanced version of HTTP, called SPDY (pronounced “speedy”), that could do clever tricks like load multiple pieces of a page simultaneously. The technology world liked SPDY so much that it adopted a modified version of it as a new HTTP 2.0 standard in February last year.

SPDY and HTTP/2 on their own are fine enhancements over HTTP 1.1, Prince says. But the really exciting part is that HTTP/2 also enables support for a feature called Server Push, which actually lets a web server “talk” to the browser and “explain” the next thing it should load.

In short, it means the browser doesn’t have to guess what is the most important thing it should display next — the server will tell it. That means a massively more efficient and organized loading process and shrinking wait times, right in our existing browsers and apps.

Google Chrome and Mozilla Firefox fully support Server Push, and Apple’s Safari browser has beta support. Microsoft has promised it’s coming to its Edge browser. More support will probably come with time.

“This is a sea change in the way the web works,” Graham-Cumming says. “This upends the web.”

The problem is that the onus is on developers to actually go through with it and activate HTTP/2 and the Server Push feature. It’s not hard, necessarily, but it requires a little bit of work, and it has been slow to get widespread adoption.

This is where CloudFlare comes back in.

Flare up

CloudFlare exerts a disproportionate level of influence on the internet for a venture-backed startup, even one with investors like Google and Microsoft.

CloudFlare is what you’d call a content delivery network, or CDN. Basically, when you visit a website using CloudFlare, it detects where you’re at and routes you to the closest server.

Not only does it mean a better user experience — it can make the difference between life and death when it comes to trading on Nasdaq, or even a high-stakes game of “League of Legends.” Prince boasts that it handles somewhere between 8% and 10% of all web traffic, including 7% of the million most popular websites.

In other words, CloudFlare already handles much of the hard work when a browser is talking to a website. That means that with CloudFlare doing the legwork to get Server Push up and running, it makes it much easier for its many thousands of users to adopt.

So that should lead to a rush of websites tripping over themselves to activate it and capitalize on a better, faster user experience. And with that rush comes a critical mass, Prince says, as everyone everywhere looks to adopt Server Push. In fact, Prince says, CloudFlare has done it before, in 2014, pushing the adoption of the SSL security technology.

It will take at least a year to play out, Prince says, but the end result is an improved web for everybody and the potential for a new class of web applications. And you might slowly notice your favorite sites loading a little faster.

“This is going to light a fire under everyone’s a–, and it’s going to be awesome,” Prince says.

Share
Posted on April 28, 2016 at 11:34 am by lesliemanzara · Permalink · Leave a comment
In: Mobile Technology

LTE-U coexistence moves along but not fast enough for some stakeholders

, FierceWireless, 4/28/16

Qualcomm, T-Mobile express frustration on length of LTE-U process

While some LTE-U stakeholders are frustrated with how long it’s taking, the Wi-Fi Alliance says it’s on track to meet its goal of releasing this summer a final test plan for ensuring coexistence between LTE-U and Wi-Fi in unlicensed spectrum.

The Wi-Fi Alliance hosted a Coexistence Test Workshop in San Jose on Tuesday, where attendees representing a cross-section of Wi-Fi, cable and wireless industries presented their contributions in order to move the whole process forward.

“There are still disagreements – don’t get me wrong — but attendees are having reasonable discussions and debate to come to consensus” around their areas of disagreement, Kevin Robinson, VP of marketing at the Wi-Fi Alliance, told FierceWirelessTech.

Stakeholders are making good-faith efforts to make positive contributions to move the effort forward, and that commitment was certainly maintained through the workshop, he said. By way of example in how contributions are being made, the Wi-Fi Alliance needs access to LTE-U equipment to conduct the validation process, so it’s reliant on others for that. Toward that end, one LTE-U vendor, Qualcomm, made a commitment to provide equipment to support the process. Robinson characterized that as a positive development, and the Wi-Fi Alliance is looking for similar commitment from others in order to move work items forward.

Qualcomm, however, still isn’t happy with the pace of the process. “Qualcomm believes that the Wi-Fi Alliance’s timeline for finalizing the test plan needs to be greatly accelerated,” Dean Brenner, SVP, Government Affairs at Qualcomm, said in a statement provided to FierceWirelessTech. “For many months, we have bent over backwards to collaborate with the Wi-Fi Alliance and our colleagues in the wireless industry to finalize a joint test plan and to prove — once and for all — that LTE-U will not have any detrimental effect on Wi-Fi — despite the mountain of evidence already proving this.”

He also said that for several months, Qualcomm has offered to provide prototype LTE-U equipment to the Wi-Fi Alliance so the test plan can be finalized, but the Wi-Fi Alliance has not been ready to accept its offer. That apparently changed this week, as the Wi-Fi Alliance confirmed to FierceWirelessTech that it has accepted the offer.

Brenner added: “We also provided feedback on the version of the test plan released on April 1 and engaged in deep technical collaboration to answer questions, provide technical feedback, and ease concerns. While it is positive that a deadline of late June was set to establish the mandatory energy detection levels in the final test plan, prior drafts of the test plan from months ago already included such levels. The time already taken to complete the test plan has been extended for much too long. This process needs to be completed soon. We will continue working to achieve this end result so this innovative technology can be used to deliver improved mobile service to U.S. consumers.”

Qualcomm’s desire to move faster is understandable given it’s been an early supporter of LTE-U and the sooner equipment is out there, the sooner it and its operator partners can reap the benefits. Robinson said the Wi-Fi industry is certainly mindful of the commercial interests of LTE-U vendors and there’s always a balancing act between their desire to get equipment out into the market and the Wi-Fi industry’s goal of making sure there’s no adverse impact on Wi-Fi.

T-Mobile US CTO Neville Ray also expressed a degree of frustration this week when he talked with analysts during the company’s quarterly conference call. “We’re frustrated,” he said. “We’re not seeing the progress that we would like to see. We still have an ambition to push solutions into the marketplace inside 2016 but based on, from where we are from a regulatory perspective at this point in time… the light is dimming there a little. But that said, we are making good progress. We have an STA under review with the FCC which would allow us to advance testing.”

Ray also said the company has a commercial kind of small cell product ready to roll that’s LTE-U capable. The handsets “being the long pole in the tent, we’re still pushing with our OEMs for 2016 capability,” he said. “That piece may slide into early 2017, but that’s kind of the timeframe for us at this point in time. It’s a 4Q kind of 1Q 2017 story and we’re going to keep pushing very, very hard.”

One of the biggest topics of debate has been around what level one would expect an LTE-U device to become aware of nearby Wi-Fi and potentially defer transmission, according to Robinson. Where that threshold is set has important implications for coexistence with Wi-Fi. The guiding principle over all of it is to make sure the introduction of LTE-U has no greater impact on Wi-Fi than if a second Wi-Fi network were to be introduced.

The alliance also is in the process of identifying a third-party test house that can serve as an independent test laboratory. Robinson also noted that the Wi-Fi Alliance is moving faster on the LTE-U test regime process, which started around November, than it has with any of its own certification programs. It’s incumbent on anyone interested in moving the final LTE-U/Wi-Fi test process along to be involved and make contributions, he said.

Share
Posted on April 28, 2016 at 11:27 am by lesliemanzara · Permalink · Leave a comment
In: Mobile Technology · Tagged with: ,

Android N takes baby steps toward becoming a virtual reality platform

, Ars Technica, 4/18/16

Android N gets a “VR helper service” and a “sustained performance mode.”

The second Android N Developer Preview came out this week, and while it brings support for Vulkan, new emojis, and a few UI tweaks, there are also a few references to virtual reality buried inside the new update.


Enlarge / The “VR helper services” screen.

It looks like apps will soon be able to register themselves as something called a “VR Listener” or “VR Helper.” In the latest Android N you can see this by navigating to Settings -> Apps -> Configure apps (the gear button in the top right) -> Special Access -> VR helper services. It looks like this will work similarly to the “Notification Access” screen (used by Android Wear to bring notifications to a smartwatch)—the VR helper services screen will show a list of apps that plug into this API, and users can allow or deny the permission.

In the settings strings there’s a permissions warning related to the VR service that states “[app name] will be able to run when you are using applications in virtual reality mode.” It sounds like when Android kicks over into whatever this VR mode is, the helper app will be able to pop up and do… something. We’re not sure what. We’re also not sure how comprehensive this “Virtual Reality Mode” is.

Right now Google’s VR products consist only of the Google Cardboard app, but we’ve long heard rumors of Google expanding that into a large VR push. Google is rumored to be working on both a smartphone-powered Gear VR-style headset and a standalone VR headset. The Wall Street Journal has also said Google is working on “a version of the Android operating system to power virtual-reality applications.” With this helper service we could be seeing the first of that VR integration work.

The Android N framework also has a new hardware support flag called “config_sustainedPerformanceModeSupported.” “Sustained performance” is something smartphone SoCs are very bad at today. Mobile chips are mainly designed for 2D app usage, so they’re great at spinning up quickly, loading an app or webpage, and quickly going back to sleep to save power. If you push the CPU and GPU for an extended period of time, you’ll quickly hit the chip’s thermal limit, and the SoC will start to throttle. The first Gear VR couldn’t handle extended usage and would actually boot the user out of VR mode when it got too hot. A “Sustained performance mode” sounds like it would change the SoC’s performance mode from a sprinter to a marathon runner, which could benefit gaming and virtual reality.

With Google Cardboard, the Samsung Gear VR, and Project Tango all running virtual reality or augmented reality programs on top of Android, it makes sense for Google to embrace this Android use case with official API support, but rumors from The Wall Street Journal, The Information, and The Financial Times would all put this on the map as part of a bigger push into virtual reality.

Share
Posted on April 18, 2016 at 10:35 am by lesliemanzara · Permalink · Leave a comment
In: Android, Mobile Technology · Tagged with: ,

Google is said to be considering Swift as a ‘first class’ language for Android

, The Next Web, 4/11/16

About the time Swift was going open source, representatives for three major brands — Google, Facebook and Uber — were at a meeting in London discussing the new language. Sources tell The Next Web that Google is considering making Swift a “first class” language for Android, while Facebook and Uber are also looking to make Swift more central to their operations.

Google’s Android operating system currently supports Java as its first-class language, and sources say Swift is not meant to replace Java, at least initially. While the ongoing litigation with Oracle is likely cause for concern, sources say Google considers Swift to have a broader “upside” than Java.

Swift is also open source, which means Google could adopt it for Android without changing its own open source mobile structure.

Could Google do it?

Born at Apple as a replacement to Objective C, Swift quickly found favor with developers as an easy-to-write language that shed much of the verbosity and clumsy parameters other languages have. It was introduced at WWDC 2014, and has major support from IBM as well as a variety of major apps like Lyft, Pixelmator and Vimeo that have all rebuilt iOS apps with Swift.

Swift can’t be copy-pasted for any platform, though. Specifically, Android would need a runtime for Swift — and that’s just for starters.

Google would also have to make its entire standard library Swift-ready, and support the language in APIs and SDKs. Some low-level Android APIs are C++, which Swift can not currently bridge to. Those would have to be re-written.

Swift would also not be useful in bridging higher level APIs in Java; they’d have to be re-written as well.

Using Swift for Android is not impossible, though. Late last year, developer Romain Goyet toyed with Swift for Android — and had some success. While that project was completed well ahead of Swift being open source, it nonetheless proved that it can be done.

That project used the Android NDK, which allows other languages to be loosely implemented into Android. With an open source Swift and support from Google, Android apps wouldn’t require that toolkit.

All told, Google would have to effectively recreate its efforts with Java — for Swift. If the company is motivated enough, it’s very possible to do so without compromising on its open source values or ruffling any developer feathers along the way.

Kotlin

Just reaching its potential, sources also claim Kotlin is being discussed as a first class language for Android.

Like Swift, Kotlin is object oriented with a focus on safety. Unlike Swift, Kotlin works with Android Studio, Google’s IDE for Android development.

Unfortunately, sources tell The Next Web that Google’s current mindset is that Kotlin is a bit too slow when compiling.

But, Kotlin is billed as a language that “works everywhere Java works,” and has “seamless” support for projects that mix it and Java.

It would be much less work on Google’s end to get Kotlin up and running for Android, but could be a tedious transition for developers.

Facebook and Uber

Facebook’s interest in Swift appears to be completely founded in technological advancement.

A benefit of Swift is that it can serve as both a forward-facing language as well as a server-side one. For a product like Facebook, that’s beneficial; apps and servers can speak to one another seamlessly, and it potentially gives the company a wider scope to write APIs for services.

And work may have already begun. A Github pull request in the Swift repository named ‘Port to Android’ was made by a Facebook employee. It’s not clear if his work was official Facebook business or not, though we have confirmed Facebook is already working with Swift internally — it’s just not known how thoroughly.

Uber’s road to Swift is probably a bit cleaner than either Google or Facebook. Though there are many moving parts to Uber’s service (app, server and API), it can use Lyft’s transition to Swift as an example.

When Lyft migrated its iOS app to Swift, it was a ground-up remodel that took a lot of time and effort — but resulted in an app that’s lighter, leaner and easier to maintain. It’s not known how much (if any) of Lyft’s back-end uses Swift, but the company has been highly complimentary of Swift in its existing application.

When could a move to Swift happen?

The short answer: not anytime soon. The reason? Android.

But Swift is quickly finding its way. Several studies suggest it’s one of the fastest growing languages around, and has blown up since going open source (GitHub tells The Next Web the language is currently its 11th most popular). Demand for developers who know Swift is also exploding, which could be all the indication these three companies need to at least explore using Swift more thoroughly.

Google’s onboarding for Swift would be long; it essentially has to rewrite every Android service, app and API. Google would also have to spearhead Swift support for Android — which is still only being poked and prodded at by clever developers in the Swift community.

In a way, Google has already begun moving away from bits of Oracle-flavored Java. It’s now using the Open JDK for Android rather than the proprietary Java API, and may be considering a post-Java life altogether. Talks in London were said to be exploratory; Google is not yet pushing to move on from Java. While it would be a big undertaking, Swift is meant for speed and safety, and Swift’s roadmap suggests it won’t be quite as difficult to use it for other platforms in the future, specifically when it comes to C++.

Though Kotlin is an alternative, it’s a very nascent language without the eager community Swift has.

Facebook and Uber face similarly daunting tasks when it comes to using Swift throughout, but can –and should — wait for Google to shoulder the load with Android. If the use of Swift is going to be as deep as our sources indicate (that is, all companies want to be using it for server side and forward-facing use cases), Android support is integral.

Moving to Swift for any of the companies also makes little sense unless it’s a thorough re-do, but it’s probably not quite as hard as it sounds. Services like Perfect prove that server-side Swift is ready, and it’s worth considering that Facebook’s engineers (perhaps from the Parse team) may already be working on this.

IBM is also working to make Swift ready for server-side functions.

But don’t expect Google, Facebook or Uber to announce Swift-y plans anytime soon. Facebook and Google both have developer conferences on the horizon, and there’s no indication that Swift will play a major part at either.

We reached out to all three companies for comment on the information our sources brought forward. All three declined to comment. Google specifically pointed to its ongoing litigation with Oracle as reason not to participate in this article. Make of that what you will.

Share
Posted on April 11, 2016 at 3:08 pm by lesliemanzara · Permalink · Leave a comment
In: Android, iOS, Mobile Technology · Tagged with: ,