How HTML5 will kill the native app
Over the past two decades, the mobile industry has become increasingly stunted by fragmented protocols, standards, and regional differences. But a hot new technology called HTML5 promises to remedy this by delivering an unprecedented open, democratic and wonderfully fertile mobile web.
Evangelists say the HTML5 movement has so much momentum that it could defeat the native app — an application that is designed to run on a single platform — in as little as two years.
Sundar Pichai, who leads Google’s HTML5-happy Chrome OS initiative, agrees that the “incredible advantages of the Web will prevail” over the dominant native app model. Another mobile developer expert Mike Rowehl adds: “We’ll forget that we even passed through another era of native apps on the way to the mobile web.”
The transition comes at a time when the mobile revolution is driving economic growth in the US and abroad. Phones are quickly become our second brain, and users are snapping up the smartest phones they can find. Companies, large and small, are investing billions of dollars to create a smartphone presence.
HTML5 heralds huge efficiencies for web publishers, because it lets companies develop once and distribute across any device via an Internet browser. An HTML5 triumph will not only save billions in development costs, but it will also allow publishers to direct those savings towards more innovative, productive projects.
HTML5 apps are searchable by crawlers such as Google’s search engine, ensuring that the apps can be discovered by billions of consumers. They can mash content with data or apps from third parties, and access analytical services such as traffic measurement tools, and ad server targeting technologies. You don’t need to get anyone’s permission to distribute an HTML5 app. And to top it off, at least one study says consumers prefer the convenience of them (though the research was commissioned by Adobe, which is partial to web apps).
HTML5 is so-called because it is the fifth generation of HyperText Markup Language, which is the coding language used to create web pages. By distributing over a web browser via fast, new mobile networks, HTML5 gets to bypass much of a phone’s underlying “iron,” or the chips, graphical cards and other components — all things that native apps rely on. Most phones being sold today have modern browsers that will operate on super fast 4g or LTE networks — the sort of thing that the HTML5 technology needs to thrive. Thus, as HTML5 advances (developers are working hard to improve it), companies will no longer need to build native apps.
So there’s tremendous logic behind HTML5’s onslaught. Opponents, of course, say it’s not an assured victory. HTML5 has some limitations on things like speed, and access to certain phone features such as bluetooth. What happens over the next 12 months, however, will say a lot about its chances. Its destiny primarily depends on the next steps by Apple, the biggest proponent of native apps, and thus the antagonist of this story. And precisely because no one knows how this will play out that makes this drama so riveting: Apple is like the Joker in the Dark Knight, a fiend with flair and with a knack for eternal comebacks, while Batman (Google) works to keep the mobile metropolis safe.
(This debate about the emergence of HTML5, and its promise of a future beyond fragmented native app platforms is the focus of one of the sessions at the VentureBeat Mobile Summit April 25/26, a conference for the 180 executives active in transforming the mobile industry. Folks like Google’s Pichai, will be in attendance, as will the major carriers and CEOs of the most disruptive private companies.)
The story started in 2007, with the release of the first iPhone. Led by its enigmatic leader Steve Jobs, Apple gave developers their first real taste of independence from the carrier oligarchy. The iPhone’s beauty was manifold, but first and foremost, it allowed developers to build applications and sell them for a fee — to users who could conveniently tap their iTunes account to buy things through the iPhone’s App Store. This bypassed the control of the carriers, which had long dictated what phones featured on their “decks.”
By the time the dust began to settle, Apple had stolen a two-year lead. Not only that, millions of developers have invested in learning Objective-C, Apple’s programming language for the iPhone, and other developer tools; these developers become specialists with vested interest to stay loyal to Apple. Now, well into 2011, Apple keeps pushing efforts to make native apps more attractive than HTML5 web ones, in an effort to keep those millions of developers — and thus users — hostage.
And so paradoxically, Apple has turned out to be controlling, closed and manipulative. It has no incentive to push to full democracy on the web front. It is enjoys huge profits from its position, not only because it gets a 30 percent cut of the revenue from downloaded apps, but because its phones, and now iPads, are selling like crazy. It is now one of the most valuable companies in the world. At every turn, it seems, Apple finds a way to hamper or limit the features that allow HTML5 to work efficiently on its devices. It remains to be seen what tricks it has up its sleeve going forward, but it’s true that many people think Apple will be be able to stay ahead. There are so many areas where Apple and other companies have hived off their own platforms from the Web that Wired last year declared that the “Web is dead.”
But if you look closely, despite the Apple/Joker’s continued pranks to keep native app alive, you’ll see how much Google/Batman keeps closing the gap on him. The following are the areas where native app gained a quick advantage over HTML5. Note that in almost all areas HTMl5 has caught up. In several areas, HTML is about to catch up. In a few areas, HTML5 has a plan to catch up, but is admittedly at least a year or two away from doing so:
- Touch/gestural interfaces — Gestural technology has been implemented by HTML5 framework vendors, such as Sencha. UI components that are controlled by touch and swipe, such as carousels, scrolling lists, disclosure panels and related widgets are all supported on the HTML5 web. Vendors like Sencha are also helping get rid of things like back buttons, refresh buttons, passed links, bookmarks and other “anachronistic” features of the desktop web that don’t translate well onto the mobile web. Thus coding time has been cut down too.
- Visual Scale — There’s nothing here that HTML5 can’t address. The web page now has sufficient ways to ask what size screen its on, and size images and resolutions accordingly.
- Video/Audio — Now addressed by HTML5 for sustained playback. Audio synchronization for short sound effects still needs work in the browsers.
- Graphics & FX — Native apps are faster for some operations – particularly anything very graphics-intensive. Graphic-intensive games won’t render as effectively in HTML5 anytime soon. However, increasingly, vendors like Sencha are working around many of the speed issues by doing things like embedding a map component that can be primed for loading maps — addressing the slowness you’ve seen in things like Google maps or other sites.
- Camera/Video access — HTML5 can handle photo capture from a web page on Android devices (at least on the latest versions, run by the Honeycomb OS; but it can’t handle it on iPhones yet).
- Contacts access — Here, HTML5 addresses file access, but most apps are beginning to draw from the cloud anyway, and not from the device client.
- Accelerometer access – HTML5 can handle this.
- Bluetooth access — This is one device access feature HTML5 has not addressed yet. That said, even for native apps, bluetooth access is fairly limited
- Disconnected Operation — Web apps through HTML5 can now work in disconnected mode; you can get up to 50MB of database space if you ask user permission, in order to keep operating without an internet connection.
- App Store Services (discovery, updates, payments & trust) — Not only can HTML5 apps be sold through HTML5 or Chrome app stores, they can be sold directly through Apple’s App Store, Android Marketplace or Blackberry App World, after being placed in a simple “native” app shell such as Nimblekit or Webworks.
- Running in the background and sending notifications – There are HTML5 specs for these capabilities, but they haven’t been implemented in the leading browsers yet. When placed in a native wrapper, HTML5 can do this, but it still means it can’t do this without extra help.
- Business model – Ad revenue works well on HTML5, since the mobile web already has ad networks. But ads aren’t doing as well on mobile as many expected, so other monetization methods are necessary, such as payment technologies for subscriptions or virtual goods. For HTML5, there are PayPal and Google APIs, but the experience isn’t very good. Lately, however, companies like Zong and Boku are making payments dead simple for the mobile web.
To conclude, native apps are still extremely popular for many developers, because HTML5 is still working to close the performance gap. Take Trulia, the company that offers real estate information online. It’s not a game company, and so theoretically doesn’t need the blazing speed offered by a phone’s underlying chip iron. Still, mobile is a significant portion of the company’s traffic (20 percent and growing) and it’s map-heavy– and HTML5 can’t handle the intensity of map graphics as well as native can.
Chief executive Pete Flint told me he hired ten developers to make native apps, and those apps have shown far superior engagement and page views, he says. “As a brand publisher, I’m loathe to create native apps,” he told me, “it just adds massive overhead.” Indeed, those developers need to learn specific skills to building native mobile apps, arguably having nothing to do with his core business. They have to learn the different programming code, simulators and tech capabilities of each platform, and of each version of the platform. By diverting so much money into this, he’s having to forgo investment in other core innovation. (do the back-of-the-envelope math: at least $100,000 per developer, or a total $1 million investment). In an ideal world, Flint says, he’d have embraced the evolution to HTML5 Web apps — but HTML5 is just not there yet.
But HTML5 will emerge competitive on just about every level within two years, says Michael Mullany, VP of marketing and products at Sencha, adding that already 95 percent of the functionality of native apps is being delivered by HTML5. And if you have any doubts about this, he points to the story already played out on the PC web. For at least 15 years, developers have been able to create “better apps” on the Windows PC desktop, compared to what they can do on a web browser. “But when was the last killer Windows native app developed?,” Mullany asks. It was probably Microsoft Outlook, which came out in 1998, he says. “Native has always had a performance advantage on the desktop,” he says, “but it hasn’t mattered because of the other benefits of being on the web.”
HTML5 graphics performance for fast-moving games that have a lot of animation can’t match native’s performance, and probably won’t for some time to come. But for pretty much anything else, HTML5 is good enough, an increasing number of developers are saying (see this great review by Redfin’s Sasha Aickin). The benefits gained from a slightly faster native experience will be so marginal for the vast majority of apps that it just wont’t matter enough to forgo the considerable benefits of the open web.
Things are moving very quickly. In just the past month or so, HTML5 has shown momentum in other areas: The main browsers, from Chrome to Firefox to Explorer have bolstered their support of the web framework. Facebook, one of the fastest growing companies, and most popular companies in mobile, has largely embraced HTML5, but is expected to say more soon at its f8 conference. Enterprise players are realizing its advantages, too. Slow to embrace the smartphone native app, they’re now balking at the cost of developing those apps, especially now that Microsoft-Nokia is offering yet another compelling alternative to iOS, Android (we’ll talk another time about how Google is like Two-Face, sometimes), Palm’s WebOS and RIM — why even deal with the splintered distribution each each app would have? After all, many companies have already spent decades developing web apps for the PC, and so they don’t want to start over with native mobile apps. So some enterprise companies are embracing HTML5 apps instead. Sencha says it saw its business double last year, largely because of this trend. Finally, publishers are getting pissed off at Apple’s insistence to retain 30 percent of the revenue from apps sold through it’s store.
This story is still in suspense stage: We just don’t know when it will end, but we do know it will end. The logic behind HTML5 is just too compelling for native to win, but at the same time we just don’t now how many more tricks the Joker has up his sleeves to stave off this inevitable tidal shift to an open, democratic metropolis.
