After the “how Google could lose in the 700Mhz network space” discussion a few weeks ago, it was all about “how Google could lose in the mobile OS space” at the WCA Xerox PARC event on November 27th.
Both conferences seem to hint at the need for strong partnerships with experienced companies for Google to achieve their new wireless challenge.
Your phone as the smart remote control for your life
Why the focus on Mobile OS now?
A cell phone today is close to what a PC was three years ago. The capabilities of the software begin to outshine those of the hardware. A consistent, robust and rapid OS platform for rich interfaces is a must to allow third-party software development and compelling applications, said Rich Miner, Wireless Engineering at Google.
Your cell phone will soon become “a remote control for your life,” said Gerardo Dada, Microsoft’s Director of Windows Mobile Marketing.
The smart OS revolution has already started, and their prophets educate other web developers to adapt to mobile devices. “CNN people start thinking of cell phones when putting contents on the web”, Miner said.
The users have more sophistication today, too, Dada said.
This may trigger the tipping point for an explosive growth rate of 300%, according to analysts.
Among 2.7 billion mobile devices, and growing, for now only 115 millions are smart phones, phones with an Operating System and software applications like a PC.
At stake: the mobile OS world market supremacy
Should the existing mobile OS providers, Montavista, Symbian, BlackBerry and Microsoft, be afraid of the new Apple’s iPhone and Google’s Android?
Not at all, Windows is well rooted in the enterprise market, Dada said. “People thought browsers would kill the OS”, he added, smiling.
“We will not let it to Microsoft without a fight,” replied James Ready, the CTO of MontaVista, the leader of the open source mobile alternative, proposing a Linux-based mobile OS since 1999.
“A lot of steps for Google”
Speakers from the incumbent mobile OS providers turn out to provide Google with an impressive “To Do” checklist in order for the Android to compete with the existing mobile OS.
Symbian has 9 years of experience improving its OS to make sure it scales with the application needs in regards to the device limitations and network resources constraints, said Panagrossi, Symbian’s VP of US Operations.
That’s “a lot of steps for Google,” Panagrossi said, and “free does not necessarily means good.
It is “enormously expensive” to put together all the features of an OS on a restricted device to work in a restricted in-network environment, confirmed Ready.
Among the technical difficulties to undertake when developing a mobile OS:
- Power management constraints:
- Device limitations:
Nothing can replace a mouse, and keyboards and navigation are not easy. The interface commands what the user will do, said Victor Brilon, Nokia’s Convergence Product Manager.
- Scalability adapted to the application:
Linux alone was not enough to build a mobile OS, said Ready. Montavista had to develop device power management and real time capabilities that did not exist in Linux.
- Security model:
On Blackberries, the applications are protected from each other and from external access, said Alan Brenner, Senior VP BlackBerry Platform. Similarly, Symbian OS provides a sandbox environment for each application downloaded, that way the application cannot access critical data or system, Panagrossi said.
- OS customization requirements to adapt to in different languages for different countries and different markets, and for a myriad of handsets.
Google’s “YouTube for mobile applications”? “Not realistic.”
The issue with openness is that each country, each operator, each developer, has its own interpretation of the specifications, and the implementation of multiple applications on a single limited device may turn to a nightmare.
What Google needs to deliver is a “YouTube for mobile application,” said Miner, in order to enable long-tail applications and break the system free.
“Optional components, software upgrades are much worse in the mobile world than in the PC world,” said Brilon. Interpretations of the same software will be different for different hardware. To create a single development environment despite the handset and the country diversity is “not realistic, not possible.”
Wait… Is there a market? For phones: yes, thank you Apple
“Apple did a great job with the iPhone and validated the market place”, said Brilon. ”The iPhone proves people are ready to buy a smart phone” and to spend $300 on a phone, without any subsidies from the operator.
Creating an open platform will be the key for the rest of the market: Google helps us by “increasing the size of the pipe,” said Brilon.
In that sense, is the price of the smart phone important? Google’s Rich Miner said that Microsoft mobile OS currently costs about $12, or 20% of the costs of a smart phone. Is that a significant gain for the customer if Google gives away the OS for free?
“What value will people gain from that cost savings?” Microsoft’s Gerardo Dada replied. “Does open means free? Or does open means more people can contribute and make money?”
“Microsoft’s partners make money,” said Dada, “and our mobile software development kit scores millions of downloads.”
Surely the
For applications? Long-tail or not long-tail, that is the question
The user data usage is still “frightening low” due to the lack of availability of services people find useful, said Alan Brenner.
The solution, he said, is a more practical “mobile services portal” to help people find out the applications they’d like to download.
Will that be enough to build a market for mobile data? For Brenner, data usage growth will come from a series of highly used data application, but not from long-tail.
However, one can argue the long-tail market on the Internet gave Google the billions to buy-in and sit at telecom’s poker table today. Can Google apply the same business model to the telecom industry?
“I would not recommend a company to start in the mobile application development business,” said Miner.” Look how many they are, and how many make money.” Miner used to work as an investor for
First you have to fight Product people inside the company. Then the existing OEMs have to validate your application. Once available, how many people will download the application is an issue. The interface should be very easy to use, and the environment is so not integrated the application cannot pass any information. Overall, handsets are “very frustrating platforms to develop for,” Miner said.
“Openness also means more fragmentation,” replies Dada. “At some point there are so many API and implementations that the model is not working anymore. A developer makes more money working for a closed application.”
“The applications are too fragmented in the
A new exciting challenge for the Bay Area and all the developer’s world brains.
