The androids are coming
With the release of Android, the battle is on for the mobile operating system market. Can Google conquer the mobile Internet as it did the hard-wired Internet?
Apple may have been catching all the headlines with its revolutionary iPhone. But history could remember Google as the company that really turned the mobile phone sector on its head in 2007-08.
As last year was drawing to a close, and industry observers were counting how many iPhones would be sold for Christmas, speculation was mounting that Google was about to match Apple's masterful move. Many thought Google would use its powerful brand to enter the handset manufacturing market, and make its own mobile phone. The press had even already got a name ready for the gadget: the 'Gphone'.
We were in for a surprise, though. On 5 November 2007, Google finally revealed what all the fuss was about. Google was indeed officially entering the lucrative mobile phone business, but not as a handset manufacturer. In the event, its product was not hardware, but software. And it wasn't called Gphone, but Android.
The sci-fi sounding name comes from Android Inc - a company Google acquired in July 2005 - which had been co-founded by renowned software developer Andy Rubin, now Google's director of mobile platforms. The product Rubin and his team have been working on is a complete software stack for mobile devices. In essence it was a new operating system (OS), associated middleware, user interface and a collection of applications.
The mobile OS market is very different from the computer OS market dominated by Windows. Microsoft is a player in the mobile OS business, but the leading vendor here is - rather than Microsoft - Symbian, a company owned by a group of telecom equipment manufacturers headed by Nokia.
Out of every 100 smartphones currently shipped worldwide, 75 are powered by Symbian OS. Unlike the PC market, though, this doesn't mean we already have a clear winner. Since it is only high-end smartphones that can support these types of advanced operating systems - and since smartphones account for only 10 per cent of total mobiles sold - it means that the battle to develop the core software that will drive 90 per cent of the world's next-generation phones is still very much on.
The first thing to say about Android is that it is built on a Linux kernel (version 2.6 for those familiar with the open-source code). Google's grand plan is based on promising something that sounds almost too good to be true. Google is offering a free-to-use yet versatile OS that could help handset makers reduce time-to-market for new devices.
Android is also a fully open, customisable OS that could help cellular operators differentiate their service propositions while lowering the cost of innovative devices. Android is also an open invitation to mobile software developers to write exciting third-party applications. Finally, Android may give mobile phone users an Internet experience on their handsets to rival that which they've come to expect from their PCs. Too good to be true?
Well, nobody has yet seen a fully operational handset built with Android inside it. The closest any hardware vendor has come to showing what the new OS will look like and what it will be capable of doing was a group of semiconductor companies (including Texas Instruments, ARM, NEC and Qualcomm), which earlier this year at the Mobile World Congress in Barcelona had some prototypes with basic functionality.
However, the first commercially available Android-based phones are expected to launch before the end of this year. LG and Samsung are rumoured to be two of the original equipment manufacturers (OEMs) already working on integrating the software into a device.
No surprise there, considering both Korean OEMs are members of the Open Handset Alliance (OHA - see box, above).
But don't expect every other phone maker to come up with an Android product straight away simply because they joined OHA, warns David Wood, Symbian's executive vice president of research. "My experience of alliances is that, often, companies join just to keep an eye on things and avoid the risk of being left out," he says.
"It's very difficult to work in these kinds of committee situations. I also notice that many of the companies that are in the OHA are also in the LiMo Foundation, which is a different, competing Linux consortium, and it's unclear how serious the commitments from many of these companies are. No doubt some are committed, and no doubt some will produce successful phones, but I don't think that people should be misled by just a list of names to think these companies are then definitely going to be bringing out great phones based on Android."
Carolina Milanesi, research director for mobile devices at Gartner, is equally cautious regarding the extent of support Android will initially receive from supposedly friendly OEMs: "We are waiting to see how many vendors are going to be participating in this. You have Motorola, which at the moment has bigger problems than trying to work on a new platform; it already has plenty of those." Motorola's bigger problem being mainly a dramatic slump in phone sales which is dangerously eroding its global market share.
One of the ways in which the PC and mobile OS environments differ considerably is in the number of competing platforms they have. Although competition between platforms is usually accepted as beneficial as it increases choice, quality and price, having more than 30 incompatible operating systems (as the cellular industry currently does) doesn't help developers who are writing applications that are intended to be used by as many users as possible.
Apart from Symbian OS and Windows Mobile, BlackBerry OS, iPhone OS and Palm OS are some of the other proprietary operating systems powering smartphones from a variety of OEMs.
Then there is the growing list of open source, Linux-based offerings such as Openmoko, Access Linux Platform and LiMo Platform, to which we now need to add Android. So, isn't there a risk of a new Linux-based platform actually compounding the fragmentation that already characterises the mobile OS market?
Not necessarily, according to Rich Miner, vice president of wireless strategy at Google and co-founder of the former Android Inc: "Much of the fragmentation we see today occurs because OEMs are forced to re-implement the same features multiple times due to the lack of 'off-the-shelf' solutions. This is even the case where detailed industry specifications exist.
"By working closely with key industry leaders to build a rich and complete mobile platform with as much off-the-shelf software as possible, OEMs will no longer need to deviate from the open-source code base. They can instead differentiate their offerings with no fragmentation. The OHA is committed to reducing the threat of fragmentation by building a platform that can be responsive to industry requirements," he says.
Gartner's Milanesi agrees that Google's involvement in the open source movement could help bring some much-needed order: "The industry was expecting Linux to become more of a player in the market. What Google is doing is trying to really move this forward and rationalise the Linux offering, trying to put an end to this very fragmented offering that we have so far with a lot of different Linux initiatives and proprietary solutions.
"We knew Linux was coming; it's just that now it's going to be coordinated if you like behind Google. If anything, Google is going to help drive this force," she expects.
'I'm feeling lucky'
Ask Microsoft how much of an impact it reckons Google is going to have in the mobile OS sector, and the company will invariably point to the search engine specialist's lack of experience in this territory.
"Microsoft has spent five years focused on building software and services for mobile phones that invite broad industry support, including third-party developers, mobile operators and OEMs," says a spokesperson from the Redmond giant. "We've made great progress and expect that to continue, but it's taken us time and investment. Android will have to tread a similar path, so we're unlikely to see any impact in the short term."
Ask Symbian, and you'll hear a similar argument: "We have already shipped 222 different phone models with the eight leading phone manufacturers of the world," says David Wood. "We have got 12 years' experience in doing this, and that knowledge is very precious.
"I'm not saying that people in Google aren't clever as well, but this is a different kind of rocket science from for example the rocket science that's involved in search [engines]. That experience can't just be duplicated overnight, no matter how clever people are."
Can new search providers offer a superior alternative to Google? And who needs Google anyway, when you can build your own search engine? See 'Search engine futures' special feature in the 19 June issue of E&T.