[why do most phone user interfaces look alike? Research Director Andreas Constantinou reviews the technology and vendors that are making it possible for phones to adopt radically different UI Personalities, changing from a Barbie to a BMW at the touch of a button; and putting a new premium content market on the radar]

Grid-based menus, list-based dialers and text-based email applications. Sounds familiar? This is how user interfaces on mobile phones have looked for the past 5 years.

Handset user interfaces today lack differentiation; while industrial design seems to leap ahead each year with fresh layouts and materials, the user interface retains the same monotonous appearance as in the last 5 years with lists, grids, text everywhere and icon menus for a topping. Even in the recent paradigm of touch-screen devices only iPhone has been a true innovator with the likes of Nokia’s 5500 resembling a touch UI welded on top of a traditional Symbian OS.

So why don’t manufacturers produce unique handset UIs ? Why doesn’t each phone have a completely different UI, allowing users to select diverse phone personalities from retro to ultramodern or from macho black to girly pink?

The bottleneck in user interface innovation
The software development process is in fact the bottleneck in UI innovation. From a technology perspective, the user interface is made up of a set of core applications such as the idle screen, dialer main menu, inbox, contacts, calendar, camera, browser shell and settings menu. Core applications are designed on a product line basis, rather than a handset model basis, with UI specifications set in stone 18 months before handset launch. Individual core applications are customised to meet regional and operator requirements 6-9 months before launch.

Core applications are written in native C language, mixing logic and presentation and fused together via a chain of static dependencies. This makes altering application look & feel as risky as a complex surgical operation, which carries a high opportunity cost; delaying a handset launch results in 2-3 million dollars per week in lost profits for a high-end model.

Handset themes, i.e. skin-deep application colouring options today implemented on most handsets, are in effect superficial and do not affect the interaction and graphics design of the phone. All in all, UI innovation has hit a process bottleneck.

UI Personalities
A new class of technologies has been quietly in development since 2004, slowly making it into manufacturers’ product lines to revitalise the UI development process; these are software frameworks like that decouple application engine from the handset look & feel, across the entire core application suite, a type of software solution that we refer to as UI Frameworks.

UIFs are designed for rapid development of new user interfaces, reducing the time to radically change the handset UI from 18 months pre-launch to post-sales; this enables the handset look & feel to change from a Barbie personality to a BMW one, while enabling a new market of premium downloadable content.

Examples of diverse UI Personalities (source: TAT)

The downside is a high upfront integration cost; a complete UIF integration requires an OEM to ‘rip out’ all of their core applications and rewrite the UI-related elements of each application; we understand that it takes an OEM 10 man years to develop the complete software for a key-based device using an existing OS core and a mature UI framework.

Technical challenges must also be noted; UIF performance is an issue for handsets below ARM11, 300 MHz, representing a ballpark 90% of the handsets produced in 2008. Plus OEMs are too cash constrained and wary of competing handset launches to focus on a radical software redesign.

Due to these challenges, very few UIF products exist, namely Acrodea’s Vivid UI, TAT’s Cascades, Digital Airways’ Kaleido and Mentor Graphics’ Inflexion, in order of market penetration. Qualcomm’s uiOne HDK is effectively a UIF solution bundled with a complete application suite, while Adobe’s Flash UI can been stretched to deliver UIF functionality on an application-by-application basis.

Vendor profiles: TAT, Digital Airways and Acrodea
The Astonishing Tribe (TAT) is a software company developing graphics rendering and UI framework technology for mobile phones. Founded in 2002, the company employs 140 people with headquarters in Sweden and offices in Korea and the US. TAT has seen its Kastor 2D/3D graphics framework embedded in over 200 million handsets across 5 out of 6 top OEMs as of mid 2008. The company reports that Motorola, Asus and one more tier-1 OEM have licensed TAT Cascades (a UIF product), and expects Cascades to have been embedded within 30 handset models released to the market in 2008. On legacy platforms, TAT’s customers typically apply Cascades functionality to 3-5 core applications, while in new platforms, the product is used across the complete user interface.

It is also worth noting that the Vodafone Simply series of five handsets launched between 2005 and 2007 featured a UIF product based on Digital Airways’ Kaleido. We understand that Digital Airways is also behind the Porsche P9522 handset introduced by Sagem in late 2008.

Acrodea is a Japanese mobile software vendor offering graphics, messaging and gaming middleware. Founded in 2004, Acrodea employs 127 staff with offices in Japan, Korea, Finland and California and reported revenues of 3.1 billion yen (circa $35 million). Acrodea’s Vivid UI is a UI framework that has been deployed with operators DoCoMo, KDDI and SoftBank in Japan across 70 handset models and 30% share of handset shipments as of mid 2008, according to the company; the Vivid UI engine has been powering the idle screen and main menu across these handsets.

The business case for Vivid UI varies per operator. Since November 2006, DoCoMo has been offering a “Kisekae Tool” service which allows users to download branded UIs for $3-$5 each. As of August 2008, the KDDI Naka-Change service offers branded UIs to buy through its retail stores via a cable connection. SoftBank’s “Onajimi Sosa” service on the other hand uses Vivid UI to help users transition from one handset menu system to another.

Operators provide the Vivid UI SDK to content providers who develop branded content; as of mid 2008 there are 100s of official (on-portal) and unofficial (off-portal) content providers of Vivid UI – based content. Revenue is shared based on existing operator models, while Acrodea monetises under a per-unit royalties models plus NREs. The potential for downloadable UIs cannot be understated; the ‘Kisekae Tool’ service alone is a fast growing one, expected to reach revenues of $100 million-a-year by end 2008, according to Acrodea, particularly as the addressable market expands beyond 30% of handsets sold.

Outlook: UI Personalities as a new form of premium content
UIF technology is poised to break through the bottleneck of user interface innovation; despite the technology adoption moving at a slow pace since 2004, 2008 has been the year where at least two tier-1 OEMs have incorporated UIF products in their product lines.  The main technology inhibitor – the multi-million opportunity costs associated with revamping software development processes – is slowly subsiding, opening up the potential for a new market of UIs as premium downloadable content, following in the footsteps of Japan.

We believe that UI Personalities will eventually emerge as a new premium content market in Europe and US. We see this first happening in 1H 2010, where tier-1 OEMs with an aggressive service agenda will be deploying own premium UI portals in Europe, in partnership with branded content providers. Less aggressive tier-1 and tier-2 OEMs will be approaching the market via tier-1 operators.

Overall, we expect a new premium content market of downloadable UI Personalities to emerge in Europe and US. Following in the footsteps of Japan, this market may also compensate for the declining market of ring-tones, wallpapers, music and other commoditising forms of content.

Moreover, we expect to see downloadable UI Personalities resist price erosion for two reasons; firstly, UIF technology cannot be easily standardized across handset OEMs; secondly UIF technology can permeate the entire look & feel of the handset, resulting in high-value, deep forms of branded content that engage the buyer across the user journey, even adapting to handset usage and context.

Comments welcome as always.

- Andreas