Openness, Linux and Mobile Innovation

As the most dynamic and commercial industry on Earth — which has grown its annual revenues from zero to US$1 trillion in less than 25 years and today manages 4 billion consumer relationships — mobile has propelled itself forward with a highly discriminating attitude toward openness within device technology.

From the beginning, the need for interoperability has sustained widespread pragmatism about “opening” individual technologies through closely crafted proprietary licenses and establishing supporting industry standards. However, there has also been longstanding caution among carriers and device vendors about open device platforms such as Microsoft Windows Mobile and Symbian OS, based upon doubts about the real intentions of their ownership structures, as well as an unwillingness to cede value and control to a dominant proprietary platform in the manner of the PC industry.

Thus, a patchwork quilt of closed vendor-specific device platforms was perpetuated, to the chronic detriment of efficient innovation, until it became clear that the “business as usual” approach was incapable of bringing the mobile Internet to life as an accessible, relevant and, of course, fully monetizable mass mobile consumer experience.

However, a new class of well-backed industry platform initiatives has been launched in the past two years to redress this situation: First came the LiMo Foundation consortium, followed by Google’s Open Handset Alliance and, more recently, the Nokia-driven Symbian Foundation.

The aim of these initiatives is to assuage objections by adopting technologies, licensing practices and governance approaches from the more egalitarian open source world, and thereby unlock the innovation needed to ignite the constantly connected world.

These initiatives are undoubtedly now taking the industry in a better direction but — as is ever the case within mobile — critical nuances can still be found beneath the new surface level uniformity.

Defining the Terms

To recap the key definitions: A closed system limits access to the source code and APIs to internal developers, along with selected external developers under private licenses, and is used when significant internal functionality lies above the competitive threshold; an open system makes the APIs (and sometimes the source code) generally available under a proprietary license, which ensures that the flow of innovations supports the competitive aims and objectives of the system owner; an open source system makes its source code and APIs available on an unrestricted royalty-free basis under an OSI-approved open source license and, depending on the exact nature of the license, may also require modifications and derivative innovations to be distributed under the same license (although so-called permissive licenses do not have this requirement).

A device software platform today consists of a set of discrete but highly interdependent commercial systems — typically considered as the operating system kernel, the middleware, the application environment, the SDK and the application distribution system (or the “app store”) — which should be looked at individually from an openness perspective when evaluating the core aims and objectives of the platform provider.

Success for LiMo

LiMo Foundation concentrates fully on producing a Linux-based device platform, making extensive use of established open source technologies and confining its scope to the Linux kernel, associated middleware technologies, and standardized native and Web application environments — while agnostically enabling companies using its platform to produce their own SDKs and app stores.

LiMo is setting out its stall to attract the vast existing open source and Web-developer camps into mobile with standardized and well-understood programming models. Its success will therefore depend in good measure on the desire of major operators and device vendors to create associated routes to market by putting in place their own self-branded SDKs and app stores to counter platform-specific app stores from Google and Nokia, as well as device-specific app stores from Apple and Research In Motion.

Two recent industry developments suggest that this will be case: the last minute nonconsummation of the expected marriage between China Mobile and Apple centered on the iPhone and Apple’s App Store due to unsolvable branding conflicts; and the announcement of an ber app store with global reach for standardized Web applications by the Joint Innovation Lab (JIL) consortium of four large and regionally complementary operators: of China Mobile, Softbank Mobile, Verizon and Vodafone.

Incidentally, the JIL group has also stated that it is making certain network APIs — powerful assets uniquely held by carriers that have previously been underexploited — available to developers in a consistent cross-network manner. The aim is to enrich application functionality and enable applications and services to be billed seamlessly through subscribers’ existing payment arrangements.

Success for Android

Google’s Android also consists of the Linux kernel, associated middleware, and an Android-specific application environment (Dalvik) provided under the permissive Apache v2 open source license, plus a proprietary SDK (which is open but not open source) the license for which ties developers’ innovations to the Android platform.

It seems that the success of Android will depend primarily on a) whether the Open Handset Alliance (or another body) can be positioned to discharge any governance over Android that is plausibly independent of Google; b) the degree to which Google’s own app store — the Android Market — manifests business model conflict, brand conflict or risk of consumer disintermediation in the minds of carriers; and c) whether Google preserves meaningful platform integrity while using the permissive Apache v2 license or, like Java and other well-intentioned platform efforts, falls victim to fragmentation — the Achilles’ heel of the mobile industry.

The foregoing points are wholly interdependent: If there really is good alignment of commercial and strategic interests between Google and the prospective implementers of Android, then the requirements placed on the system should be compatible enough to permit independent governance.

Success for Symbian

Nokia’s Symbian Foundation has also begun the process of transferring Symbian OS — which, by virtue of having a head start of nearly a decade on the other two platforms, has attained far greater unit volumes (albeit nearly all on Nokia’s own devices) — into open source.

Perhaps the critical success factors for Symbian OS within its new setting will be a) the degree to which the industry accepts that Symbian Foundation can exercise any meaningfully independent governance over the platform — or whether it is adjudged to be quasi-independent at best, by virtue of Nokia’s recent full acquisition of the platform and employer relationship with the platform engineers; b) the extent to which the owners of the spectrum of proprietary technologies embedded within Symbian OS will come to the view that it is in their best interests to move their IP into open source, and c) whether the wider open source developer communities, which are already deeply invested in Linux and other well-established collaborative technologies, will warm to and invest in the unfamiliar and, by some accounts, challenging qualities of Symbian OS.

The recent launch of oFono, the new Linux-based open source device platform project jointly lead by Nokia and Intel, suggests that Nokia may also be reflecting on these issues.

Success for Mobile

Whatever the respective market shares that are eventually held by these major new device platform initiatives, it is now clear that meaningful openness, boosted by Linux, is finally poised to deliver unprecedented benefits for mobile consumers, developers and manufacturers of mobile service providers and device manufacturers.

Device vendors and wireless operators will be able to bring to market innovative new products faster and at a much lower cost. Developers will gain access to key device and network capabilities and tools that will enable them to build more compelling and usable services, bringing the Internet developer model to the mobile space. And consumers worldwide will have access to a less expensive and more enchanting mobile experience.

Morgan Gillis is executive director of LiMo Foundation, the nonprofit industry consortium dedicated to creating the first truly open, hardware-independent, Linux-based operating system for mobile devices.

Leave a Comment

Please sign in to post or reply to a comment. New users create a free account.

LinuxInsider Channels