Share

Linux has been mentioned as a potentially leading platform for mobile devices for as long as there have been mobile devices. However, mobile Linux is still largely missing in action. The new crop of high-visibility smart phones such as the Samsung BlackJack, the Nokia E62 and the Treo 680 are based on Microsoft's Windows Mobile, the Symbian and the aging Palm OS platforms.

So why are some in the mobile industry saying, once again, that Linux is on the brink of becoming a significant platform for advanced mobile devices such as smart phones? And why should anybody but industry insiders and geeks care?

"Linux has had peaks and troughs in the mobile industry, but it's looking quite positive at the moment," said Matt Lewis, research director at ARCchart. The market research company recently issued a report entitled "Linux: The New OS Celebrity."

One reason for that optimism, according to mobile Linux proponents, is that this is an area in which Microsoft doesn't dominate, despite its best efforts. The reputed lower cost of Linux and its success in the corporate server market also encourage its supporters.

Perhaps most important, however, is that Motorola, the second-largest mobile-phone vendor in the world, has thrown its support behind the operating system, saying recently that it expects more than half of its phone models to be based on Linux by the end of 2008.

According to its proponents, the success of Linux will affect both the price of mobile devices and the development of mobile applications for consumers and enterprises. In other words, they say, the future of Linux is far more than just a matter of industry infighting.

The sceptical point of view

Linux's success as a mobile platform has been predicted before. In 2002, for example, a firm helping electronics manufacturer Sharp Corp. market its Zaurus Linux handheld device predicted that Linux-based handhelds would capture 12 per cent of the market by 2004. Linux never succeeded as a handheld operating system, and except for a few Asian markets, the Zaurus didn't succeed either.

Skeptics also claim that the competitive landscape is too harsh for mobile Linux to succeed, even without Microsoft's dominance. Multiple market studies have shown that, worldwide, Symbian is by far the leading platform provider for advanced mobile communications devices. Symbian is co-owned by Nokia and several other major phone manufacturers, ensuring that it will continue to succeed.

In those studies, Microsoft's Windows Mobile and Linux trail far behind Symbian in second and third place. However, skeptics point out that Microsoft is working hard -- and spending a lot of money -- to ensure the long-term success of Windows Mobile. The fact that several high-visibility smart phones such as Palm Inc.'s Treo 700w, Samsung Corp.'s BlackJack and Motorola's Q are based on Windows Mobile attests to Microsoft's progress, the skeptics point out.

Even sceptics acknowledge that Linux is, in fact, currently being used in mobile devices. However, with only a few exceptions, its primary use is in low-end devices aimed at consumers such as so-called feature phones. These devices provide some mobile computing capabilities, like the ability to maintain personal information, but aren't nearly as powerful as smart phones.

"It's best suited for consumer-level devices right now," ARCchart's Lewis acknowledged. "It's very much a feature phone platform at the moment."

The question is whether mobile Linux can succeed as an operating system for more advanced devices such as smart phones, which are rapidly gaining in popularity. The answer to that question may well hinge on Motorola.

The Motorola gambit

Motorola has long built feature phones and basic cell phones on the Linux platform. But it is having success with Linux on smart phones, reportedly selling 1 million of its Linux-based Ming smart phones per quarter in China. Overall, Motorola has released nine Linux-based phones in recent years. The best-known is the ROKR E2, which has a built-in MP3 player and SecureDisk storage, according to a company spokesperson.

Even though Motorola was once a co-owner of Symbian and has also created phones based on that system, the company understood the value of mobile Linux long ago, according to an executive.

"In August of 2001, we had our first kick-off meeting to look at Linux on phones," said Mark VandenBrink, chief architect for Motorola's systems software and mobile device business. "We shipped our first Linux handset in 2003."

He noted that one reason for that early interest was that Linux is indeed a less expensive platform.

"We found that all chip set manufacturers were using Linux to test chips and to do functionality and coverage tests," VandenBrink said. "So when we started talking with our chip set manufacturers, they were already doing it, and we found their source code was already available. The same was true when we moved ahead to the next level of functionality like audio and video codecs at the application level -- everybody was already using Linux. So when we looked at pulling everything together, a lot of the pieces were already available.

"Finally, there is a huge ecosystem of people all around the world doing all sorts of fascinating things with Linux," VandenBrink said. "In little universities, little mom-and-pop shops and big companies, people around the world are using Linux for innovations."

Motorola has long since sold its stake in Symbian and, earlier this year, stepped up its commitment to mobile Linux. It announced a partnership with several major players in the wireless space including NEC, Panasonic Corporation of North America, Samsung and cellular carriers NTT DoCoMo and Vodafone to develop common application programming interfaces (API) for mobile Linux. That is essential to the success of Linux as a mobile platform, many observers agree.

"The problem with Linux has been that it doesn't have any one owner," said Ken Dulaney, an analyst at Gartner. "What you need are specific profiles that define the target device. Microsoft has that with smart phones and Pocket PCs. Developers can develop [applications] and be sure they'll work. Symbian has it, too. But since Linux developers have been so focused on open source, they haven't created these profiles."

These profiles will enable applications developed for mobile Linux to run on phones from multiple vendors, which it can't do currently. The operating system's current lack of cross-vendor support is one reason it has been relegated largely to relatively unsophisticated devices, the analysts agreed.

If the Motorola-led effort succeeds, it could mean wider acceptance of mobile Linux for advanced devices like smart phones, the analysts said. However, the effort has a long way to go, Dulaney noted.

"I don't see it going anywhere," he said. "Every time somebody tries this, a vendor says, 'If you do it my way, we'll do it.'"

VandenBrink disagreed that the effort would fail, although he acknowledged that the Motorola-led effort will take time.

Mobile Linux for the enterprise

If Motorola's effort to create standard APIs succeeds, mobile Linux could be successful in the enterprise, even though that's where Microsoft's Windows Mobile is particularly strong, many people believe.

"From an enterprise standpoint, Linux integrates better with a lot of [business] applications," said Neil Strother, research director at mobile device content and service for market research firm The NPD Group. "That gives Linux an opening. If you go to a company with a strong Windows platform presence, it also could have segments of its back-end operations on Linux. Plus, if there's a core group of IT guys who are Linux geeks, and the company is cross-platform aware, they might be more open."

That success would come, however, despite Microsoft's efforts to convince IT personnel that Windows Mobile is easier to manage than its competitors in shops that use Windows desktop and server products.

"If you compare Linux to the Windows Mobile environment, Windows could be more attractive because that's where most enterprise developers have the most experience," ARCchart's Lewis said. "But there's no reason enterprises can't develop apps in the same way with Linux as they do with Windows Mobile or Symbian."

Motorola's VandenBrink, not surprisingly, agreed.

"What the enterprise market really boils down to at present is e-mail and personal information management," VandenBrink said. "That's an important part of the enterprise, but if you look at a broader aspect of what enterprise [mobility] really means, it's where sales people [on the road] can get portals back into their home system. That's the real enterprise market. That's what an IT person or an IBM looks at. We're at the beginning of trying to figure out what type of device will play into that."

A Palm Linux resurgence?

There's one other wild card for mobile Linux: The Linux version of the Palm OS. There are a number of mobile Linux platform developers, but Access, which now owns the once-popular Palm OS, is the most visible. It previously has said it will release the first Linux version of the venerable and familiar Palm OS to device manufacturers by the end of 2006, which would have resulted in devices being on the market by the third quarter.

News reports, however, indicate the release to manufacturers could be as late as the middle of next year. A spokesperson for Access did not return calls about this issue or about participating in this story.

Motorola's VandenBrink said the effect of that development is hard to know, but that Access' success isn't required for mobile Linux to succeed.

"It's hard to speculate about something I haven't seen or have no understanding about," he said. "But there are a lot of people in this market. I talk to four to six companies a week, and a lot of them are coming out with Linux platforms."

Dulaney was even more dubious, saying that there's no evidence that any vendors, except perhaps Palm, would embrace the Linux version of the Palm OS. "[Access] just don't have enough people supporting it," he said.

Whether Access succeeds, however, the analysts agreed that in the end, some level of standardisation is needed for mobile Linux to succeed.

Motorola could succeed in an effort to standardize some portions of Linux, although history is not on its side, Dulaney said.

"There are a lot of good intentions but, when it gets down to it, [Linux vendors] want to fight among themselves," Dulaney said. "In the consumer market, sure, there will be Linux. But the prospect of continued fragmentation in the developer community is very high."