Splitting apart browsing tools and the actual browser

Mozilla's new iPhone app isn't just smart; it might herald a real change in the way we think about what a browser is and does.
Written by Simon Bisson, Contributor and  Mary Branscombe, Contributor

Mozilla's new iPhone app isn't just smart; it might herald a real change in the way we think about what a browser is and does.

It's smart as a business model because holding your breath and waiting for Apple to allow full alternative browsers onto the iPhone will leave you blue in the face (Opera Mini is more of a viewer for pages re-rendered in the cloud than a true browser and the version of Skyfire that's almost ready for submission to Apple is "not a browser as much as it is a 'cloud service' that boosts the performance and features of the native browser on any device," according to CEO Jeff Glueck). It may be a smart move for the BlackBerry and Windows Phone 7 platforms as well. The Silverlight platform on WP7 isn't powerful enough to build an alternative browser in and RIM's investment in its own browser has made coping with the complexity of the BlackBerry programming model to deliver an alternative browser less attractive; Glueck says "the Blackberry developer environment is not as favourable for cutting-edge application development. The APIs are fragmented and inconsistent, and the Java virtual machine Blackberry requires is not efficient." Skyfire for BlackBerry is on hold at least until RIM comes out with its new browser, which Skyfire can improve on by transcoding Flash so it works on the handset.

Mozilla's Firefox Home takes a different approach, supplementing not the rendering power of the browser but the navigation and the usability. You get your Firefox desktop history, bookmarks and open tabs on the iPhone, rendered in Safari. If the page looks good enough and the speed of loading the page is determined as much by your network connection as by the power of the browser on your phone, then choosing an alternative browser becomes about convenience and usability. You have to be online to browse - so why should something as useful as your bookmarks and your browsing history be locked on any one machine? Firefox Sync (which used to be called Weave) will let you have the same bookmarks and history on the desktop with Firefox and on mobile with Firefox Home and Fennec.

But how about giving me that same history in any browser on any device? The Awesome bar doesn't have to be coupled to the rendering engine; why can't it be a cloud service that I can use with IE9's impressive GPU acceleration or whatever it is you like best about Chrome?

No one browser will ever implement all of the Web standards - there are just so many of them, and some are philosophically opposed to others; Robert O'Callahan (who by the wonders of open source 'works for Mozilla full-time as an employee of Novell in New Zealand') put this nicely in his blog post explaining why SVG Fonts, H.264 and WebSQLDatabase won’t be in Firefox. "No browser plans to implement every "open standard", not even every W3C Recommendation - nor should they. Some standards were junk from the start, others are obsolete and are being replaced by better alternatives, others are simply not good for the Web." When Chris Wilson was on the IE team, he put it a bit more diplomatically; "you differentiate the browser by what standards you choose to implement".

If a site I want to use needs a browser that supports a specific standard Google Wave and canvas, say), I might be able to get one browser housed inside another (like Chrome Frame) or I might be able to open a page into another browser - or worst case, I copy and paste the URL (or use a different device entirely). You might also like the security, the addins (and there's always an interesting tension between those) or just the interface of one browser over another. But if my history and bookmarks and passwords (with Mozilla's Account Manager project) could all just follow me from browser to browser and device to device, life would be a lot easier. Now that really would free the Web.


Editorial standards