Did you hear that Facebook, arguably the biggest champion of HTML 5 (a set of web technologies) decided to throw its weight behind native mobile apps instead? So did Steve Newcomb, the co-founder of Berkeley, Calif-based Famo.us, a 17-month-old startup that announced itself at TechCrunch Disrupt in September 2012. That didn’t deter him a wee bit, because his company was working on an entirely new approach to developing HTML5 apps and having them work on all sorts of devices including tablets, cars, televisions and smartphones. Since the September announcement, the company has been in quiet mode, but now Famo.us is breaking its silence and today is ready to show off not only elements of its technology but also talk about how much it is going to cost and how developers can access it.
So what’s Famo.us
Famo.us is the brainchild of CEO Newcomb, a co-founder of Powerset, a search-engine company that was acquired by Microsoft and now is part of the Bing. His co-founder is Mark Lu, who is also the head of engineering for the company. The company has only ten employees and has raised about $ 1.1 million in seed funding from the likes of Greylock Partners, CrunchFund, Javelin Venture Partners, Matt Ocko and Barney Pell.
Famo.us has developed technology that makes it easy to write HTML5-based apps and perform like native apps — as long as there is a modern browser and a modern OS, mobile or otherwise. In addition, these HTML5 apps get the added benefits of their web roots; for instance, they can be indexed by Google. Newcomb believes that as more “screens” become prevalent in our lives – PC, tablets, phones, television and the car – developers and companies will need ways to reach their audience through those screens. Today’s model of developing for one platform at a time doesn’t quite scale.
There is no doubt that Famo.us is entering crowded waters and will be jostling for developer attention from other mobile-oriented frameworks such as Sencha Touch, jQuery Mobile and Appcelerator Titanium. Newcomb, however, believes their focus on performance gives them an edge. In order for HTML5 apps to meet the native app experience, the performance needs to be between 40-to-60 frames per second, Newcomb says. They need to be jitter-less and flicker-less. ”We built an engine and then a car around it,” he explains, when asked to compare his product with others.
Anthony Wing Kosner, a web-content guru and a blogger for Forbes.com, in an illuminating article about Famo.us put it best when he wrote:
If the original HTML specifications were designed for navigating hierarchies of documents, HTML 5 was designed to enable web pages to be more like applications. But this “more like” has turned out to be a sticky point. Web browsers are still optimized for the display of documents, and applications—apps—are not collections of documents. More properly, apps are dynamically generated views of data that a user can directly manipulate. In a Venn diagram, there is an overlap between web pages and apps—many websites dynamically generate views of document based content—but it is the ways that they differ that have given developers problems. The difference in the amount of interaction between an app, its data and its user is of a high enough order that it is like the difference between different dimensions.
Game on: Apps not documents
Later, when I talked to Newcomb, the Famo.us approach became even more clear. He argues that in today’s world, most mobile browsers are using the WebKit rendering engine and that was developed to display document — it is essentially made up of WebCore and Javascript Core. WebCore renders documents. Famo.us renders apps.
As Kosner says, today’s apps aren’t documents — so Newcomb and Lu decided they were better off writing their own rendering engine, and in doing so, they were taking a cue from the video gaming business. While some video game developers write their own game rendering engine that powers their games, many more try and use game rendering technology developed by other companies to build games. In that sense, Canvas for 2D games and WebGL for 3D games are analogous to Famo.us.
For a developer, nothing much changes: each app developer includes tiny bits of Famo.us specific code in their app. The first time an app is accessed, the Famo.us javascript libraries are downloaded to the client and that’s that. The data comes from a datastore and via JSON goes through the Famo.us rendering engine which in turn taps into the graphic processor unit (GPU) and displays the app on the browser in a manner befitting the screen it is being displayed on.
And now for the news
Newcomb told us that he was surprised by the number of people interested in his technology. A few thousand app developers have already signed up for the service, though it will be a few months before Famo.us is available in beta. The company has introduced a new sandbox so that developers can play around with the technology and some of the key features.
At present the demo will work on any modern computer running Chrome or Safari or Dolphin with keyboard and mouse controls, iPad and iPhone iOS 4.2 to iOS 6, Android phones and tablets with Jelly Bean or Ice Cream Sandwich with Chrome or Dolphin with JetPack. It also works on Samsung SmartTVs 2010 and later.
Many of the capabilities on show should be enough to lure developers, says Newcomb. The plan for the business is to offer Famo.us under a GPL license to non-commercial users — but anyone who has commercial aspirations will need to pay $ 99 fee for a non-GPS license and will be able to get access to between 10-and-20 templates that can be modified to build apps. Famo.us is currently looking for developers who’d like to build famo.us app templates – which are core design patterns archetypes found in the iOS app store.
Why Famo.us matters
I am of the belief that apps of tomorrow have to be data informed and as such need to be able to take in a lot of information, and render it in a simple and easy to use, touch-oriented manner. So, when Newcomb announced the company and showed his demo at Disrupt, I got quite excited because I believed that they had developed technology that would allow us to solve the vexing problem of a data soaked future.
At Disrupt, Newcomb showed off a periodic table that could be manipulated to call up data behind each element. Now imagine a similar immersive interface that sits in front of Pinterest, eBay or Mr. Porter for a new kind of online store front. Or imagine a brand new way of interacting with Facebook photos and data. You get the drift!
The real opportunity for the company could be working with larger players who desperately need a way to stay relevant in the bi-polar world of Apple and Google. In a time when we are swimming in a sea of mediocre and meh startups, Famo.us clearly has a big, hairy audacious idea. Today it is almost famous. And if things go right – that is, developers sign up – then Famo.us will truly be worthy of its name.
 |
 |