Handset software is a fascinating topic in the mobile industry, albeit one that is often misunderstood. Application environments like Java, Flash Lite and Qtopia open the world of handset software to developers and allow the handset GUI to be customised. Here I introduce a new taxonomy for application enviroments and explain why seemingly distinct technologies like S60, Widsets, Maemo, Qtopia and Openwave MIDAS are essentially part of the same landscape. [Note: I have updated this article in response to much feedback and comments (thanks Nick, Philippe and Malcolm). Thanks also to Judy Breck maintainer of the Carnival of the Mobilists for selecting this article as post of the week.] Today there are 10s of different tools technologies for opening up handset software to developers and generally those who want to customise the handset GUI in some shape or form. These range from skins designers, to games developers, to operators who want to create a ‘signature’ look & feel spanning the entire user interface. I would argue that application environments include not only Java, as well as environments for creating skins, desktop UIs (a.k.a. idle screens) and core applications (dialler, menu system, email, messaging applications, etc which are entirely different to downloadable applications). Contrary to popular understanding, openness is not an exclusive privilege of open operating systems (OSes) as I argued in a previous research paper. A range of technologies exist to ‘open’ access to the handset software internals, and can be divided in two categories: external and internal application environments. The distinction between external and internal environments is down to several variables: level of functionality exposed to an application, who can access that functionality (certified apps only?), depth of integration (e.g. can you replace core apps such as the dialler and contacts application), when an application can be installed (at the point of manufacture, before the point of sale, or after the point of sale) and who can install the application. 1. External application environments, i.e. those which allow application development and handset customisation after the software has been embedded in the handset ROM. These are generally accessible to external developers, scripters and designers. Examples are: – Nokia’s S60, SonyEricsson’s UIQ, Microsoft’s Windows Mobile and Qualcomm’s BREW, which are advanced platforms. These can also be viewed as external application environments, in that they offer a rich set of APIs for application development and deployment post-load, in other words environments designed for developing downloadable but not core applications. For example companies like DreamSpring have created good contacts replacement applications for UIQ and S60. However such applications are not a complete replacement for core apps (such as the contacts app) since the can only hide the built-in app, and they are not able to access sensitive APIs such as the voice-dialling API as this is hidden by the handset manufacturer. See my comment on this post for further analysis on why S60/UIQ etc are designed as external and not internal app environments. – Open C (POSIX APIs on top of Symbian), which are C/C++ APIs aimed for use by application developers – Java 2 ME (in its myriad forms and extensions), Python for S60, .NET Compact Framework and AppForge, a range of application environments for developers or scripters – Adobe’s Flash Lite aimed at scripters and designers developing graphics-intensive apps. – S60 ActiveIdle, Windows Mobile homescreen framework and Nokia’s Widsets, a range of technologies for allowing developers to customise the UI desktop (aka idle screen) – Nokia’s Carbide.UI theme edition tool for deep skinning of the handset GUI 2. Internal application environments, i.e. those which allow application development and handset customisation before the software has been embedded in handset ROM (and in certain cases during the handset lifetime, too). These are accessible to handset manufacturers, network operators and handset distributors. Examples are: – Openwave MIDAS, a ‘deep’ scripting environment for creating operator-customised applications – SVG players from Ikivo and BitFlash for developing graphics-rich, interactive applications such as Vodafone’s Live! Cast. – TAT’s Cascades, Digital Airways’ Kaleido and Mentor Graphics’ Inflexion (previously NextDevice), which are rapid prototyping tools software frameworks for rapidly creating custom end-to-end user interfaces (i.e. the entire suite of core applications) from scratch. – e-SIM (now part of SKY Mobile Media) and Comneon’s APOXI, which are development tools for building suites of core applications (albeit tools which are more aimed towards engineers than designers). – Trolltech’s Qtopia, Maemo’s Hildon port, ALP’s Hiker, OpenMoko’s application framework, TTPCom’s Ajar and Windows CE app framework, which are sets of C/C++ APIs for creating core applications and managing application communication and lifecycle. In the next diagram I attempt to classify the above application environments in terms of the extent of customisation which they permit and the time at which they can be applied. The x-axes show, the time of customisation is directly related to the barriers to customisation; internal application environments are mostly accessible to the manufacturers and partners, whereas external application environments are accessible to all. Note that the chart is quite extensive but it is still work in progress. The y-axis shows that there are four broad types of customisation that can be applied via application environments: – change of themes and skins across the handset (e.g. Carbide UI theme edition) – development and deployment of downloadable applications (using e.g. S60, UIQ, Windows Mobile, BREW) – replacement of a core application (e.g. replacing the idle screen or contacts app) – core application re-design (redesigning the entire user interface from scratch) This landscape map is very revealing, but still it doesn’t do enough justice to highlight some important parameters of application environments: the cost of customisation, the part of the software stack which a technology provides, and the target addressable market in terms of breadth of device models. Another important parameter is why customisation is applied: this can be to re-enforce branding (and retain customers) or to provide easier discovery of, and access to, new services by operators and third parties. A couple of noteworthy remarks on the map: – Flash Lite is moving from an application environment for downloadable, graphics-intensive apps, to an environment for creating other core applications such as the idle screen and the dialler (see the Samsung D900 implementation for example). – Java is also moving towards enabling development of desktop UIs (aka idle screens) and some core apps, thanks to the capabilities introduced in MIDP3. So what are the trends in application environments? On one hand all ecosystem players, from operators to manufacturers and beyond are striving to reduce the high time-to-market impact of pre-launch customisation. At the same time, external application environments do not reach as far as core applications (desktop UI, dialler, menus, email/calendar/contacts applications, camera application, etc), due to the lack of modular design of most operating systems today (including Symbian/S60 and Windows Mobile, the so-called ‘open’ OSes). Therefore, a number of technologies are being introduced to bring much-needed modularity within application environments, both pre- and post-launch. These range from rapid application creation tools from TAT, Digital Airways and Mentor Graphics, to Open Plug’s FlexibleWare modular software development framework and Red Bend’s Embedded Feature Delivery technology. These technologies will be impacting not only handset development practices, but also making an impact to the consumer market. Within the next year we should hopefully be seeing many more examples of unique GUIs such as Vodafone’s Simply range and B&O’s Serene handset. Thoughts ?

If you were at 3GSM a couple of weeks ago, you would have noticed that the term ODP has become quite widely used. Unlike other unfortunate terms (3G or user experience), on-device portals are used to signify something quite specific and meaningful: the use of specialised device software to deliver portal content and store-front capability (hint: it’s not the browser). ODPs excel compared to WAP browsers on several counts: caching content for zero-wait browsing, integration with handset capabilities and functions (such as graphics and messaging), reducing the click-distance to content, and streamlining content purchasing. On-device portals address the need for increased data revenues not through bigger pipes (read: 3G) or bigger content (read: $20M-content deals), but by improving the user experience in content browsing and purchasing. Content may be king, but the user experience is queen as the mobile industry is finding out. However, the ODP space is literally crammed today. I can count 30 providers of ODP client or client-server products today: Access Netfront Dynamic Menu, Action Engine (still around?), Adobe (FlashCast), Airmedia, Cibenix, Communology (mobile catalogue), Comverse, Crisp Wireless (mLogic platform), Everypoint, Geniem (MediaCast and Superstore), Handmark (Pocket Express), InFusio (nMap), ITfinitiny (2Go), mPortal, MobiComp, Mobinex, Nellymoser (ASAO platform), Nokia (Content Discoverer), Opera Platform, Qualcomm (uiOne), Reporo, RefreshMobile (Mobizines), Streamezzo, SurfKitchen (SurfKit series), Tricastmedia (TWUIK), U-Turn, Volantis (BuzzCast), weComm (wave), Yahoo! Go and UIActive. What a crowded market this is ! If you are an operator shopping for an ODP solution, you can go very far for your money (if you know where to look that is, and this list is certainly a good starting point). So how do ODP vendors differentiate ? Every angle in the book of the-hitchhikers-guide-to-routes-to-market has been tried. You have regional plays (MobiComp in Turkey and Onskreen in India), mobile operator propositions (SurfKitchen, Cibenix, and Comverse), content provider D2C propositions (Refresh Mobile and Nellymoser), MVNO plays (mPortal) and real-time video or data (Streamezzo and Everypoint). With maturity comes wisdom. Most ODP players now realise that they are just offering another content distribution channel (albeit an appealing one), and that value lies on how you integrate handset functionality and personalise the content served. Operator Optus in Australia for example has customised the uiActive client to integrate messaging, presence information and location awareness, serving different content based on the user’s location. More interestingly a nearby value area is emerging; not just selling content, but discovering content through the handset home screen. Introducing Desktop UIs There are a few players in this space which I ‘d call Desktop UI, and it’s heating up pretty quickly. Abaxia’s Mobile portal, Celltick’s LiveScreen and Zi’s Qix are the incumbent desktop UI products, helping the user (and the operator) discover handset features and operator services from the idle screen. But as of the last 3GSM, you can add several more products to this list: Abaxia and Tegic separately announced solutions for finding content and features from the home screen through T9-ish predictive matching (much like Qix). Then you have Korean IntroMobile’s IntroPad, Vocel, Aditon (a PA Consulting spinoff that shows adds on your mobile) and of course FlashHome, the reincarnation of FlashCast with home screen replacement features. You can naturally add the usual suspects to this list – ODP players who also provide homescreen replacement features (Cibenix, SurfKitchen, uiOne, MobiComp, et al). What’s important to understand is that Desktop UI products are fundamentally different to ODPs. DUIs are technically complex (hint: handsets are not designed to have their home screen replaced, because OEMs did not think that was a good idea in the first place). Feature-wise, DUIs are about discovering content, not browsing or selling it. DUIs can also be used to discover common handset functionality (e.g. type R-I-N-G-T from a Qix-like application and you have the options for changing the ringtone or buying a new one. Sames goes for bluetooth, camera.. the list goes on). Last and certainly not least, DUIs are about controlling the most valuable real-estate that ever has and ever will exist on mobile handsets. It’s the primary shelf space (advertisers call it inventory) that content providers are keen to get their hands on. Inventory is the reason Google paid such a ridiculously large sum of money for YouTube. Naturally, manufacturers are not staying out of the game; Nokia has Active Idle, Motorola has Screen 3 and Sony Ericsson is rumoured to be baking it’s own Screen 3 variant. Stay tuned.

One of the simplest and most brilliant ideas I ‘ve read in a while comes from Deloitte’s telecommunications predictions for 2007. This market trends paper argues that instead of trying to cram TV content designed for 32 inch TVs into 2 inch mobile phone screens, mobile operators should monetize by enabling their subscribers to upload videos they ‘ve snapped on their handset to the PC (i.e. the web) and their TV. The Deloitte paper references a Business 2.0 article with more insights and research on the subject of video moblogging. As Business 2.0 puts it: “Sending a clip of Junior’s soccer match to Grandma, it seems, is much cooler than watching the latest ‘mobisode’ of ‘Desperate Housewives’ on a two-inch screen.” This is essentially about bring not TV to the mobile, but mobile to the TV. Let’s call this Reverse Mobile TV (-MTV or minus MTV; kind of a nice wordplay, as it also alludes to turning around the power balance and being your own ‘MTV’ or popular channel). I don’t think the term moblogging (or Nokia’s term lifeblogging) does it enough justice anyway, as blogs are associated with a journal of sequential entries around one theme, whereas mobile TV making is about one-off broadcasting. Startups are keen to monetize on this trend. Veeker allows to you upload videos via MMS and share them (although MMS is a fragile technology with impractical limits for file size – e.g. 300KB in the UK – and very expensive for spontaneous video sharing). Vpod.tv is another service that allows users to create their own TV channel (still in beta, with mobile capabilities unclear). Perhaps Shozu is the most promising attempt at video sharing, since it uses a handset application to upload videos to the web. Naturally, YouTube isn’t sitting around dwindling its thumbs and plans to allow mobile users to send clips to other YouTube members within a 2007 according to a Reuters report in late 2006. [updated] UK operator 3 has launched a service called SeeMe TV since October 2005 that let’s users produce and broadcast their own video clips, although the service does not extend to the web or the living room TV. Getting the recipe right It won’t be easy to reverse mobile TV, in other words to do for mobile phones what the Slingbox is doing for TV [the Slingbox comparison is only part of the solution and might confuse. Reverse mobile TV is about uploading your videos from your phone to your living room and your friends’ living room]. So, to implement it you need handset software, distribution channels, brand & communities, 3G pipes, flat data rates and a home entertainment play. Here’s why: – handset software is needed to instantly upload videos to the internet cloud, over cheap reliable and standardised protocols like IP and unlike MMS. Handset software offers a much more compeling experience for sharing videos and silently uploading or downloading videos. Kind of what on-device portals like Yahoo! Go are to WAP portals. Shozu is a model application in this case and a lesson in the success of product vs platform business models in the mobile industry (Shozu is in fact a product created from the ‘ashes’ of the Cognima synchronisation platform). – distribution channels are essential for getting the handset software bundled and made easily accessible from the handset menu. Distribution and bundling power comes with handset OEMs, mobile operators and media brands, in order of decreasing power (market-depending of course). – brand & communities: Communities are an essential social framework of interaction. Think of YouTube’s video sharing community, eBay’s glue that allows buyers to find sellers and Amazon’s collaborative filtering. Today’s brands build communities and today’s communities build brands (Tomi Ahonen is thought-leader here). – 3G channels: At last! -MTV could be a service that justifies (some) of the billions of investments made on building 3G mobile networks. Mobile video making is spontaneous and therefore uploading or sharing has to be instantaneous – you just can’t expect users to connect their handset via cable to their PCs and load up the right software. Or even wait for 10 mins while the video uploads over a GPRS connection. – flat data rates: easily underestimated by industry insiders who rarely care about their mobile phone bills, flat data rates are essential for realising the -MTV paradigm. US, Japan and some European countries (see ‘3’ in the UK) offer flat data rates. If operators were smarter, they would launch pay-per-use tariffs for video uploading (e.g. uploading a video could cost as much as 2 SMSs), based on IP-based or APN-based charging, in association with the -MTV service provider. – a home entertainment play will enable -MTV service providers to channel made-on-your-mobile videos to your living room, and cross-sell sharing, printing or other content services. This should be on the roadmap of companies like i-mate who launched their home-entertainment products at 3GSM last week. Interestingly, the -MTV concept can not only drive data ARPU for the operators through video uploads but can also drive data downloads, as mobile video makers can broadcast themselves to other phones. Furthermore, as the Deloitte paper puts it “a growing body of amateur content, created on mobiles, is ideal for watching on mobiles”. A good charging model would be per-upload charges for mobile TV producers and flat-rate access for mobile TV consumers (a well-understood model for a two-sided market like -MTV). [updated] Operator 3’s SeeMe TV service charges mobile TV producers the price of an MMS for uploading videos, but rewards them with 1p for every ‘3’ customer that watches their clip – a revenue model variant that has spawned more than 30,000 clips from ‘3’ self-styled TV producers, rewarding popular clips with a total of over 100,000 as of March 2006. The reverse mobile TV concept is unique service offering that can be delivered by mobile operators, ideally those who offer both mobile and fixed/home services and can partner with third party brands/communities to deliver multiple, end-to-end offerings. Now, when can I get it on my phone please ?

With voice ARPU steadily declining and data ARPU undercompensating, network operators should re-examine their data services strategies. On 21 November 2006, Informa Telecoms & Media presented their outlook for the mobile industry in front of a packed audience in central London. The data presented by Informa’s six senior analysts at the event was rather unflattering for network operators. You can check here for the full video of the day’s presentations. A gloomy outlook According to Informa, voice ARPU (average revenue per user) fell from a global average of $19.38 in 2005 to $17.65 in 2006 and is expected to further drop by $1.34 (7.6%) in 2007. At the same time, the much-awaited boost in data service revenues has fell rather short of expectations. From a global average of $2.84 in 2005 data service ARPU dropped to $2.81 in 2006 and is forecast to climb by just $0.07 in 2007, according to Informa. In a nutshell, global voice ARPU is forecast to drop nearly 20 times faster than the rate by which data ARPU will increase in 2007! This outlook is set in the backdrop of hundreds of billions of dollars in network operator investments in 3G licenses, network infrastructure upgrades and megadeals with content providers. Source: snapshot from Informa’s Mobile Industry Outlook One might argue that Informa’s global predictions are overly skewed since they include developing mobile markets. Vodafone’s key performance indicators reveal a similar trend. Across Vodafone’s four main markets (Germany, Italy, Spain, UK), average voice ARPU fell by 7.39% between 2005 and 2006, while data ARPU (including messaging) rose by 3.20% in the same period (although performance across individual countries varies widely). Comparatively, in Spain and the UK voice ARPU dropped by over 3.5 times faster than the rate by which ARPU increased between 2005-6. Oddly, in Germany and Italy, voice ARPU is dropping much (40x) faster than data ARPU is rising (!). The sustained decline in voice ARPU is probably down to a couple of market effects: firstly the global mobile subs base expanding to lower-spending customers and secondly the existence of multiple SIM cards per user (1.29 SIM cards per user globally for 2006, according to Informa). Focus on voice, not data The question is, how should mobile network operators react ? My thesis is that operators should focus their strategy not on achieving small increases in data service revenue, but on how to turn around the tumble in voice ARPU. Operators should do so by banking on the capabilities of data services and the improved user experience offered by handsets applications. In other words, network operators should focus their efforts on data services that support voice services by making voice calls easier, more intuitive and more fun. Examples are T-Mobile’s MyFaves, Zi’s Qix, Comverse’s Visual Voicemail and SK Telecom’s avatar-based videotelephony. T-Mobile US leads the way MyFaves makes it easier and more fun to call. Launched in October 2006, T-Mobile’s MyFaves service offers subscribers unlimited calls to five other numbers (to any network) for a flat monthly fee starting from $39.99. In the case of mobile telephony, five is a magic number; Nokia’s 360 study of phone usage patterns concluded that more than 50% of voice calls and 70% of SMSs go to top 5 contacts. (source: study of S60 users, UK/Germany/France as reported in Sep 06). The unique selling point of MyFaves is that the handset idle screen displays a caroussel of images or avatars of the user’s five favourite contacts. These five contacts can be changed once a month by calling customer services, through the MyFaves application or through a web page. T-Mobile’s service is currently available for the Nokia 6030, 6103, 6133, Samsung t209, t509s, t609, t619, t629, Samsung Trace, Motorola v195s, RAZR v3, v3t, v3i and Blackberry Pearl. The uniqueness of MyFaves stems from the combination of a voice service, a data service and a desktop UI application (desktop UI = the type of application that provides service discovery by replacing the idle screen). The data service (i.e. synchronisation between the desktop UI and the network billing engine) as well as the handset application are intended to boost voice usage, through a 1-click access to calling one of the five contacts. The user can send an SMS or MMS to one of five contacts through the same interface. I wonder what sort of ARPU uplift is T-Mobile witnessing thanks to MyFaves. This is a trully innovative, end-to-end service. I wouldn’t be surprised if T-Mobile country operators in Europe follow the US cousin’s lead. Qix: Making calling easier Zi corp’s Qix is another example of how handset applications can be used to boost voice usage. Qix is a desktop UI application that uses T9-style predictive contact search to bring up contacts based on the keys that the user presses. Simply put, instead of going into the contacts application and typing in the name of the contact, Qix allows the user to find contacts just by pressing keys corresponding to the contact’s name from the idle screen. This drastically reduces the number of clicks needed to call up a contact and is claimed to boost voice usage by 3%. Owners of Windows-Mobile handsets will have used this feature widely, which has been built on the operating systems since v1 (with the launch of SPV in 2002). Unfortunately for Qix, Zi has had too many top management changes in the last 4 years, its operating losses have been doubling on a yearly basis (check the reports) and its device support has been surprisingly poor. On the positive side, I ‘m aware of at least two competing products to Qix which are launching next week at 3GSM. In summary, my thesis is that network operators need to heal their wounds first (the sustained decline in voice ARPU) before dashing into new and risky territory (investments into more data services). There are known recipes (and more innovative ones still sitting on the shelves of R&D teams) for boosting voice usage by employing data services and innovative handset applications. Thoughts ?

Conferences are becoming a booming business in the mobile industry. Besides the incumbent Informa Telecoms & Media, there at least 10 conference organisers such as Visiongain, Jacob Fleming and Osney Media. With conference participation fees starting at 1,000GBP and sponsorship packages at 10 times that, it’s no wonder why. Take Mobile Device Management (MDM) for example. MDM has been a hot topic for conferences since 2006, but is now getting somewhat overdone. There are at least five conferences on MDM taking place in the first half of 2006 from Informa, Jacob Fleming and Visiongain. With big business comes intense competition. Informa who takes pride in being the organiser behind the 3GSM World Congress for the past few years, did not secure the contract for organising 3GSM 2007 which is now organised by the GSM Association themselves. A more interesting twist of the conference saga evolves around the topic of SIM cards. Informa (organisers of SIM Summit) seem to have not reached an agreement with the major SIM card suppliers on the format of the SIM Summit 2007 conference. As a result, the SIMalliance (backed by the top SIM card OEMs i.e. Gemalto, Oberthur, Sagem Orga, Giesecke and Devrient) is organising SIMposium, a major SIM conference on exactly the same dates as the SIM Summit, but in a different country; SIM Summit takes place in Prague on 24-26 April while SIMposium is in Berlin on 24-25 April. If you read through the speaker list of the SIM Summit you ‘ll notice that there is not one presenter from the big SIM manufacturers, which is exactly the unique selling point of the SIMposium. It’s rare that we witness such heavy-handed, uninhibited politics in the mobile industry. As it happens, I ‘m chairing two days of the SIM Summit and I ‘m having difficulty securing briefings with some SIM OEMs. With conference organisers keen on securing sponsors, and attendees keen on quality presentations and solid networking opportunities, there is little doubt that too many conferences on the same subject is not a good thing. But with 10+ players in the conference game, it is certain that organisers will be looking to compete rather than collaborate.

If you haven’t been to the Carnival of the Mobilists before, you should give it a spin. Every week the Carnival is hosting a page-long summary of the best of mobile blogging on the web, a one-stop destination for catching up on what’s been happening in the world of mobile phones and wireless services. The Carnival enjoys a faithful community of frequent bloggers who keep the carrousel spinning – mostly industry insiders and analysts – and is compiled each week by a different member of this vibrant community. The Carnival is an exemplary, self-sustained collection of informal but highly informational blogging, thriving on quality reporting and educated opinion that in turn warrants peer recognition and blogosphere visibility. Here at the VisionMobile Forum we have been fortunate to have been featured in the Carnival several times in the last few months, most notably on: A thesis on operator strategy titled Operators: Service Pipes or Bit-Pipes?, which won post-of the-week honours at C. Enrique Ortiz Mobility Weblog and is currently at the top of the shortlisted entries for best post of November 2006 Hampus’ inspirational User interfaces and soft walled gardens of tomorrow mentioned by David Beers on Software Everywhere An opinionated piece on why Motorola’s AJAR is set to follow in Nokia’s S60 footsteps mentioned at Golden Swamp A Linux market analysis titled Mapping mobile Linux: from Lego bricks to castles mentioned at Xen Mendelsohn’s Xelllular Identity The analysis of Customisation vs Personalisation which won post-of-the-week honours over at Tarek’s blog and finally Prioritize: Take over the world or Enjoy good margins, mentioned at All About Symbian Cheers to the Carnival ! Andreas P.S. The Forum is back online after suffering some downtime over the past weekend. Lesson learned: dependable support from web hosts is hard to find. Hopefully our new home will live up to its reputation.

Understanding users can’t buy content if they can’t find it, an increasing number of mobile operators and content providers are scrambling to offer mobile search capabilities as well as an array of tools that will encourage users to explore more of the content at their fingertips. The raft of recent announcements, involving tier-1 mobile operators and market giants such as Google, Yahoo!, Microsoft and a growing number of white-label search providers, including InfoSpace, Fast Search & Transfer, Medio Systems and JumpTap, as well as mobile search platform provider, Mobile Content Networks (MCN), shows carriers and content companies are clearly excited about mobile search. But operators that merely retrofit Web search solutions for the mobile Internet ultimately short change themselves and their users. Popular Web search engines such as Google are fatally flawed. They were designed to treat all searches – and searchers – equal. While the approach consistently delivers the same list of links in response to key words, it fails to recognize the need of individual searchers for relevant results that really matter. Push vs pull business models More importantly, such search schemes patently ignore the shift in the business model from user-pull to content-push. Pull is built on the premise that users know what they want and are prepared to go look for it. That’s quite an assumption when it comes to fast-paced content such as entertainment and multimedia which changes faster than users can keep up. The pull model also ignores the rise of empowered customers who increasingly expect – even demand – content and services consistently tailored to their individual needs and in tune with their lifestyles and life stages. The new paradigm is personalized content-push based on a deep understanding of the individual’s purchases, passions and past click-behavior. It’s even more compelling if the technology can learn users’ likes and dislikes over time to dynamically and consistently deliver the right content mix. The message is getting through, which is why this year will see the usual suspects experimenting with techniques and technologies to deliver personalized mobile search. Perhaps the most vocal on the market is Yahoo!. Personalized search and relevant results are concepts that run like a leitmotiv through the company’s new product offer and its future roadmap. A prime example of this new direction is Yahoo! oneSearch, a Web 2.0-type search engine picks up on users’ intent, intuits the information they want and then presents the relevant content, grouped by subject, in synopsis form. A sports search on oneSearch, for example, will return a relevant bundle of scores from a team’s most recent game, along with game schedules, team rosters, photos, local results, and so on. Adding user choice to the mix While search does assist in delivering a better end-user experience, the much more lucrative business opportunity may be in combining search, personalization and recommendation to provide users personalized and relevant results – as well as the tools to discover other content they might not have otherwise known existed. An increasing number of vendors are clued into recommendations and are lining up to wield the power of the analytics they generate. Some, like Medio Systems, are borrowing from the Amazon.com approach to suggest content on the basis of the individual user’s past preferences or on the basis of what a user’s peers consumed, or both. Others including MyStrands, Gracenote and ChoiceStream, making the transition from online to mobile, have cleverly combined their music recommender capabilities and social networking to help users connect with both content and the like-minded members of their community who share the same tastes. Indeed, the sheer variety of personalization and recommendation solutions enables a multiplicity of business models the mobile industry is only beginning to explore. Moving forward, personalization and recommendation will be must-have features of mobile content services. mPortal, for example, creates personalized recommendations based on users’ billing records, download history and purchase habits. The company is taking this analysis to the next level, drilling down in the customer data to create ‘content referrals’, and so allow users to share content tips with their friends and family. Given the highly personalized nature of the mobile phone and the way people use it, there is a high probability that people will buy content and services if they know a friend – and not an operator – is making the suggestion. It’s clearly in operators’ interests to deliver effective and targeted commerce experiences to their customers. While recommendations based on customer information such as page views and downloads will be an important part of this strategy, it will be the recommendations from the tight-knit communities users know and trust that clinch the sale.

Mobile Linux is a fascinating topic. As an operating system for mobile handsets, Linux commands responses ranging across dismissal, FUD (fear, uncertainty and doubt) and admiration. Mobile Linux is often dismissed as an immature operating system for handsets, being sometimes likened to Lego bricks, i.e. a loose set of components that comes nowhere near a complete product. FUD comes about when companies consider the repercussions of GPL licenses (note: the Linux kernel is licensed under GPL v2) which are sometimes likened to IPR contamination. At the same time, Linux as the basis for handset operating systems must be admired for its market penetration, particularly in Japan and China. For 3Q06 Canalys estimates Linux to be powering close to 40% of the smartphones shipped in China and Japan, second only to Symbian. However, it should be noted that Linux-based mobile operating systems are far more than Lego bricks today. In fact what is emerging is a clear taxonomy of Linux vendor offerings, which I would separate in five types, ranging from support packages, to complete handset products. A) Linux support packages (LSPs): these consist of the Linux kernel, hardware drivers and core OS functionality. LSPs are today provided by the incumbent MontaVista and the newer WindRiver (of VxWorks fame). The vendors provide LSPs with integration and professional services to help manufacturers like Motorola, NEC and Panasonic go to market with Linux-based handsets more quickly. Other manufacturers like Nokia (see 770 and N800) have opted to build their own Linux-based OS distribution from scratch, with the help of external communities. B) Complete software stacks: In 2006 a number of vendors announced complete software stacks based on Linux, combining an LSP with an application framework, graphics subsystem, middleware libraries and core applications. Following the incumbent Korean vendor Mizi Research and Chinese HOpen, Trolltech followed with Greensuite and ALP (Access Linux Platform) announced their offering. C) Productised software stacks: while a complete software stack may include all the software code a handset manufacturer needs, it still lacks interoperability testing, operator acceptance testing and other integration services which today are offered by A la Mobile, Aplix (with their BTO product) and Celunite which is still in stealth mode. D) Moving a step further towards productisation, a range of vendors today offer complete hardware and software design services for Linux-based handsets. PurpleLabs, SysOpenDigia, Elektrobit, Cellon and Flextronics combine a Linux-based software stack with a hardware design optimised for Linux that can be licensed by the handset manufacturer and taken to market. E) Finally, a new type of vendor has emerged that combines white-label hardware and software productisation with marketing and distribution: the example here is FIC, the Taiwan-based vendor who made the headlines in late 2006 with their OpenMoko fully open Linux platform. FIC produces Linux-based handsets both under its own brand, as well as under white-label business model. For reviews of mobile Linux vendors such as MontaVista, A la Mobile, Trolltech and Mizi Research, you can downlad a white paper I wrote recently titled: Mobile Operating Systems: The New Generation. Conclusively the mobile Linux market today consists of the entire gamut of offerings from LSPs (the Lego bricks) to complete Linux-based handset products (the castles). Undoubtedly, this market is still very volatile and competitive. Chipset vendors are increasingly pre-packaging their own LSPs with their hardware reference designs, manufacturers are in some case in-sourcing and in other cases outsouring Linux operating systems for their handsets. Finally operators like Orange and Vodafone are promising commitment to Linux platforms, while industry consortiums such as the X-foundation are hoping to provide a route to market for Linux-handsets. Amidst the turmoil of volatility of vendor offerings and the hype of market demand, one can only refer to the age-old rhetoric: watch this space. Thoughts ?

As an interaction designer working in the mobile space and as someone who follows Apple’s products with interest, the iPhone announcement has been something that I have been eagerly anticipating ever since rumours of its existence started to circulate. Given Apple’s reputation for innovation, I was excited to discover whether Apple could bring some fresh approaches to tired old mobile device UI paradigms and the announcement has certainly not disappointed. Not only was the much-rumoured full touch screen present, but here was a genuinely disruptive UI that was optimised to finger gestures, one that incorporated several sensing technologies and that brought the graphical elegance of Mac OS X to the mobile device. Apple’s introduction of a solid-state scroll wheel to the iPod in July 2002 helped to stimulate a debate on the use of touch technologies for mobile phones. Capacitive sensing technologies brought the promise of adaptable UIs that could dynamically change touch controls to fit different contexts and thereby remove the need for physical buttons, evolving into concepts such as Synaptics’ Onyx phone. It is fitting therefore that the concept has reached its full realisation in the iPhone. For the first time on a mobile device, we have seen multi-touch input put to use to allow users to intuitively zoom the UI without the need for zoom buttons or other controls. Steve Jobs said during his Macworld keynote to much cheering that multi-touch gesturing is part of what provides the magic to the iPhone and that it has been patented, but it is worth noting that Bill Buxton, one of Human-Computer Interaction’s father figures, has been encouraging exploration into multi-touch input for over 20 years, and the technology has been demonstrated to impressive effect already in this video by Jeff Han. Another strikingly intuitive interaction in the iPhone’s UI is in the use of the device’s accelerometer to detect the orientation of the device and to rotate the UI accordingly to allow users, for example, to view a landscape image fullscreen by simply rotating the device to a landscape orientation. This is a classic example of how Apple humanises high technology to cut away needless interaction and provide highly accessible and delightful user experiences. Redefining the rules of operator engagement But, speaking as a designer of mobile devices who has had to trawl through reams of operator requirements, perhaps the most pleasing aspect of the Macworld keynote was to see how the iPhone is pushing operators to meet its design vision rather than the other way around and refuses to be tied to requirements can befuddle the UI and lead to an incoherent user experience. Stan Sigman (CEO of Cingular) revealed that Cingular had entered into a contractual agreement with Apple two years ago without ever seeing the product, based purely on the strength of the design vision. This is unheard of – Apple have truly struck a blow by mobilising the value chain in their and the end user’s favour. It remains to be seen if other manufacturers will be able to follow suit. Since the iPhone’s announcement, reactions to it from people I have spoken to and from across the web have been varied and interesting. Some have been unfazed by the hype and argue that the device contains nothing truly new or innovative either technically or in the UI. However, to have this view is surely missing the point. When it was released back in 2001, the iPod was not particularly revolutionary in terms of its specifications either, as other mass storage MP3 players were already on the market. The innovation came in its commitment to delivering an uncompromised design vision by focussing on the total user experience of the product. The ripple effect caused by the iPod’s success on raising the profile of user experience in the industry has been truly significant. Hopefully, the iPhone’s announcement will also bring about a change in attitude that will allow UI designers to unleash their creativity.

Another idea for mobile phone retail stores is selling hardware features – imagine buying a phone and then adding an FM radio capability to it at the point of sale for a modest surcharge. Although feasible, this concept can be realised mostly in the long term. For more here-and-now ideas on retail stores see previous posts on a visual approach to selling data services and differentiating at retail with a mobile phone health index. Selling hardware features The consumer purchase criteria for handsets today are the style & design of the handset, the price, the features and the bundle offered by the mobile operator (in order of importance) – the importance ranking of these criteria is roughly the same globally. When it comes to features, the consumer will often ask how good the camera is (the more megapixels the better!), whether the handset can play mp3, whether you can use a stereo bluetooth headset with it, whether it has FM radio on, etc. Business users might ask for HSDPA capability or hardware-based security. Often the over-segmentation of handsets and features based on the manufacturer market segmentation plan will mean that one particular handset may appeal to the user in terms of style, but may lack some important functionality. OEMs today are pretty sophisticated when it comes to figuring out their lifestyle-based segmentation plans (with 50+ segments in the case of Nokia), but then again you can’t fit everyone in the same box and give them a label. In addition, consumers in developed markets are pretty demanding and could easily be spoiled for choice. Software-only add-on features like games, ringtones and wallpapers are pretty commonplace today – you can grab them at the point of sale, through WAP, web or even SMS (and they make a several billion dollar industry). But what about selling hardware features at the point of sale ? Why ? Because post-sales activation of hardware features can mean money for the operator or manufacturer who can charge the user for activation. Hardware flexing One technology that can be used to enable selling of hardware features is hardware flexing. This topic was discussed at a meeting organised by the OTA Flash Forum (OTAFF) in Athens in October 2006, where I was invited as an independent analyst. This article is based on the discussions at the OTAFF meeting. So what is hardware flexing? Handset manufacturers today tend to use a single hardware platform for a series of handsets, with some of the features flexed off (i.e. embedded but disabled). The reason is that by using a single hardware platform for multiple market segments with differentiated products, manufacturers can exploit economies of scale, tighten supply chain management and repurpose handsets for different regions. A single hardware platform keeps costs down and allows manufacturers to finalise handset hardware features just-in-time. Naturally, the handset software has to support in-life activation of hardware features. If the necessary software is already ‘baked’ into the handset, this is fine, otherwise firmware reflashing has to be involved (either over-the-air or over-the-cable). To avoid having to manage countless permutations of software images for each combination of hardware features, updating of specific software features can be used. On mass-market (non-modular device software), the only company claiming a solution to this is Red Bend as far as I know. Practicalities Let’s look at some practicalities, in particular, which features to sell, how to make money, and the barriers to adoption. Post-sales activation of hardware features has to deliver a tangible benefit to the end user. Boosting the camera quality is somewhat suspicious (were you trying to sell me a sub-standard product before?), but enabling video playback, mp3 capability, FM radio or bluetooth is ok. In other words, post-sales activation of hardware features must deliver features which are ‘on-off’ or are otherwise clearly perceived by the user as an add-on. How you make money? There are two options: a) The manufacturer makes money from selling the feature, while the operator makes money from airtime traffic, or b) The manufacturer sells the device to the operator/distribution channel with the option to enable a hardware feature. The operator then sells the handset and optionally sells the feature. As for barriers to adoption, there are a few issues although not critical. The cost of equiping a hardware platform with additional features is not great (at least it’s cheaper than having to create a new hardware platform with the extra features). Commodity hardware capabilities are pretty cheap today with bluetooth coming at a cost of sub- $0.5 and FM radio at sub- $0.2. An important challenge I can see is that post-sales features have to be incorporated into the manufacturer’s strategic plan for each platform, which requires long-term planning (12-15 months) and thus carries execution inertia. The retail processes and the mobile operator channel strategies will have to be adapted to handle this capability. If you want to also sell hardware features OTA (via web, WAP and customer care) the process complexity increases. The benefits to mobile operators, handset manufacturers and users alike would be substantial though. For operators it’s an extra source of revenue, as it is for manufacturers who are today redeveloping their strategies for post-sales incremental revenue (for example see Nokia’s Loudeye acquisition and Nokia Content Discoverer). As for users, they would be much less restrained in their choices . Why not pay a bit extra for the handset with perfect style and the perfect feature set ? Thoughts ?

The jury is still out on whether convergence will continue and mobile phones become the ultimate in-pocket-Swiss-army-knife-machine. I believe that the form factor will be what sets the limits and determines whether the functions should be split into separate devices, and nothing else. ..then you break it! But it is time to think differently. These disparate functions crammed inside the phone should of course fulfill the basic rules the converged paradigm, but not all the rules. Look at the ‘take picture’ function of the camera. Does it have anything to do with the mobile phone and all its list menus? No. Then don’t force it to look and feel like one. Sony Ericsson has made an excellent choice in their new CyberShot series! The camera application looks and feels lika a camera, and not a mobile phone. This takes time to adapt to; the first Nokia with camera even had a menu you could fold up and select ‘take picture’. The more there are established paradigms out there the less reason there is to invent a new one. The UI can be the same, but adapted to add the benefits you have from the other functions, for example a context menu with ‘send to blog’ or ‘send as MMS’. These adaptations make the functions add to each other in a 1+1=3 kind of way. I believe we will have devices divided into these functions pretty soon, all looking part mobile phone and part their own: Talk (the most fundamental function and still the king) Organizer (calendar, timer, alarm, etc) Media (music player, picture browser, etc) Share (messaging, blogs, etc) Camera (looking just like any point-and-shoot camera, plus the adaptations) Web (looking just like on the PC – just the web, again with adaptations) Downloaded Applications (which will look like the service on the web they represent, but inherit the basic UI-paradigm of the device, for example a picture browser with Flickr content sown into it.) Comments ?

Mobile phone retail shops are under-utilised today. They could be used to sell not only phones, but mobile data services as I argued previously, as long as these services are presented in a visually familiar and ‘tangible’ form. Another idea for making a difference at retail shops is this: the mobile phone health index. The idea struck me when I was using my new Nokia N73 without a hands-free and was worried about how much I was microwaving my brain. I wasn’t sure what is the SAR index of the phone (SAR = Specific Absorption Rate, the quantity that describes absorption of handset radio waves by the head) and I thought it would be nice if I had known that before buying the phone. Turns out that the N73 emits a maximum SAR value of 0.92 or 1.13 watts/kilogram (depending on the version) according to Nokia’s website, compared to the maximum value of 2.0 permitted in Europe and 1.6 permitted in the US. It’s not that bad, but it could be worse. What if I could find about the SAR values before buying my phone? What if the retail shop where I bought my phone showed me information on the health index of all the phones ? Most people I ‘ve spoken to about their phone use are concerned about the health risk from a micro-microwave next to your head. It’s not that easy to find the information and interpret it. All handset manufacturers certifying their phone through regional authorities (e.g. FCC) are required to compute the SAR index. From a quick search, both Nokia, Motorola have dedicated webpages listing the SAR value of all handsets. CNet has a section of its website dedicated to cell phone radiation levels, including SAR levels for phones from major handset manufacturers selling to the US and a list of the 10 phones with the highest SAR values. The Health Protection Authority in the UK also has an excellent explanation of what SAR is and what it means in layman’s terms. By displaying the mobile phone health index (based on SAR values), handset retail shops could indeed make a difference in the eyes of the consumer. If I ‘m concerned about my health when using my phone, I ‘ll go to the retailer that can educate me about what phone is safest for me to use. Here’s a sketch of what it could look like: an information card next to each phone, displaying a ‘mobile phone health index’ (the higher, the safer the phone is), based on the inverse of SAR values. Naturally, the retailer can never tell you whether your phone is safe or unsafe (they ‘ll need to hire many more lawyers for that), but they could indicate how safe your phone is based on the internationally agreed and readily available SAR index. The health index would make a major difference to retail shops (both operator-owned and independent shops), which currently have little differentiation among them beyond price and brand colour. Over time, displaying a health index for phones may become commonplace, but for the first 6-12 months it would make a major difference in the eyes of consumers, and a major ‘first’ that a retail store chain can brag about. Thoughts ?

Open source software has existed in mobile phones since Motorola’s Linux-based A760 was launched in early 2003. Open source has come a long way, from an early experiment in mobile Linux three years ago, to an accepted strategy today for increasing product value by engaging external communities. Linux-based software stacks for phones are the leading form of open source software (OSS) today. An array of vendors has crowded the marketplace with Linux software offerings, including MontaVista, WindRiver, Trolltech, Purple Labs, Open Plug, A la Mobile, Aplix, OpenMoko, and Mizi Research. However, OSS stretches way beyond Linux. Recently Sun, Microsoft, Nokia and Adobe have announced or renewed their commitment to open source, too. One could say that open source software has been making waves in the last few months. Aside from Informa’s Open Source in Mobile conference that took place in Amsterdam in early November, there has been a recent flurry of announcements on open source software: On November 28 Trolltech announced that it would provide software productisation services to OEM and ODM customers in the form of Greensuite. Specifically, Trolltech’s professional services division will be providing full integration and testing services on a complete stack of software components, whilst offering a choice of components and UI customisation. The stack will comprise of Trolltech’s Qtopia software as well as select third party applications including browser, messaging, multimedia, DRM, VoIP support, Java, office and synchronisation components. Trolltech will be acting as the single point of contact, responsible for software integration, as well as pricing, marketing, legal and commercial issues. The Greensuite partners will be announced over the course of the next few months leading to 3GSM. The proposition is very similar to that of A la Mobile, and Aplix’s BTO offering. ALP is also a complete stack as is OpenMoko, Mizi and PurpleLabs. Nokia‘s Maemo project is continuing to grow and Ari Jaaksi (Nokia’s Director of open source operations) announced at the open source conference the Sardine device which would follow the 770 Internet tablet, both based on the Maemo stack. In July Motorola pledged that more than half of the manufacturer’s mobile phones will use Linux within 18 to 24 months. Motorola has further open sourced their implementation of MIDP3, i.e. next-gen Java for phones. At the open source conference, Christy Wyatt (VP ecosystem development) also said that sales of Motorola’s Ming have reached 1% of total phone sales in China, an impressive feat. In early November Sun announced details of how it is going to open source J2ME and J2SE implementations, marking a major twist in Sun’s Java saga. By open sourcing Java, Sun hopes to leverage community momentum around Java (especially from Vodafone, Nokia and Motorola), sustain the growth of Java and alongside its own engineering services business, and steal the thunder from the competing Flash application execution environment technology. Earlier in November Adobe announced it has open sourced ActionScript 3, a core part of Flash, which is now hosted by Mozilla and will find its place into future Firefox browsers. Adobe’s move may intend to replace Javascript with ActionScript, as its Flash environment moves to a full application execution environment, in the form of project Apollo. Handset manufacturer FIC announced OpenMoko, the first fully open source Linux phone software platform, that competes with Purple Labs, Mizi, Aplix, A la mobile and Trolltech. In late October Access Linux Platform announced it is open sourcing its application framework, a critical part of the phone software stack responsible for managing application lifecycle and communication on the handset. Linux tools vendor Open Plug announced it had secured a $15m Round B funding in early October. A la Mobile was started in June by entrepreneur-in-residence Pauline Alker with $3.5m seed funding and is believed to be looking to secure another $10m to fulfil its promise of ‘the Red Hat of mobile’. In October, private equity firm Garnett & Helfrich Capital announced it had acquired a controlling stake in a privately held U.S. company Celunite, reportedly for $30 million. Celunite is a California-based provider of Linux-based open-source technology, which is still in stealth mode. Microsoft recently expanded its portfolio of software under a Shared Source license, now providing source code access to the full Windows CE 6.0 kernel. More Linux stack vendors are expected to follow Trolltech’s Greenphone example, thus offering developer communities full access to ‘real’ phone hardware, a critical step towards growing community contributions to mobile Linux. This flurry of announcements in the last few months of 2006 can only signal more moves in the forthcoming 3GSM event in February. Beyond the tens of mobile Linux players, the commitment of Sun, Adobe, Microsoft and Nokia to open source is certain to cause disruptions in the predictable future of the mobile industry. Software vendors, mobile operators and handset who haven’t yet taken the time to understand the commercial advantages (and pitfalls) of open source, now need to.

Motorola’s acquisition of TTPCom in June bought the OEM ownership of the AJAR operating system for low-end handsets. In a corporate newsletter published in October, Motorola announced that the AJAR platform would be licensed to other OEMs, following in the footsteps of Nokia’s S60 licensing strategy. I believe Motorola’s AJAR licensing will suffer from the same issues that have caused S60 licensing to fail to date. Quick recap Motorola completed the acquisition of the Cambridge-based company on 28 July 2006. TTPCom is now known as the Motorola TTPCom Product Group and fits within the Technology Office in the OEM’s Mobile Devices Division. As TheRegister points out Motorola paid a huge premium to buy TTPcom. “At first, the price tag of 103m ($192m) looks over-generous for a company that saw a 36 per cent revenue decline in its last fiscal year to 37.2m ($69m) and a slide into the red with a loss of 32.3m ($60m). But this acquisition has nothing to do with revenues and everything to do with gaining valuable technology assets for not one but two of Motorola’s most strategic growth initiatives ultra-low cost handsets and fixed-mobile convergence.” The rationale behind the licensing strategy AJAR is TTPCom’s operating system that includes the applications framework, applications suite and tools for voice phones and low-end feature phones. AJAR (literally means slightly open) was named after the middleground between open OSs and closed, proprietary OSs. As Rob Shaddock, CTO at Motorola Mobile Devices explains in the TTPCom newsletter, Motorola initially licensed AJAR for use by its ODM suppliers. “We decided that it would give us more flexibility and the ability to move a bit faster if we could deliver a software platforms to the ODMs who were all using their own proprietary platforms.”, Shaddock says. The newsletter further explains “most people in the industry thought that we ‘d be swallowed up by Motorola and that they ‘d never hear from us again. That couldn’t be further from the truth”. The newsletter continues saying that “the target is ‘ubiquitous AJAR’ for the mass market and feature phone segments. Targeting the top tier handset manufacturers and their supply chains, we aim to get this tool used right across the industry to reduce the cost of handset production and to enable manufacturers to meet the demands of operators more effectively.” The rationale behind the AJAR licensing strategy is that “operators’ requirements are continuing to increase exponentially” (technically not true – I would say linearly, and are already showing signs of resource fatigue). “To manage the consequences, the industry needs a shared platform for the development of handset user experience – AJAR”. So, Motorola’s plan is to license AJAR to top OEMs and create a community of value-added applications around the platform. Can you trust a competitor as a strategic supplier ? AJAR is targetted to low-end handsets, combining proven modem stacks with a low-memory-footprint OS platform – characteristics which are quite different S60’s target market and design goals. On the other hand, the challenges in Moto’s AJAR licensing strategy are quite similar to Nokia’s S60 challenges. Which OEM will trust a competitor to sell them a platform ? There are four issues with this strategy: 1. Risk of lock-in to the platform and lack of roadmap control – which is why Motorola, Nokia and Sony Ericsson are using internally-developed software platforms. Samsung is also known to be developing its own Linux platform, to replace Mizi. 2. Risk of IP leakage. Who will trust Motorola to build Chinese walls between the AJAR platform group and the OEM accounts groups ? “In sharing a tool like AJAR across the industry’s leading handset manufacturers it is crucial that we recognise the need for: transparency; a declared roadmap; absolute confidentiality of customer information; and a level playing field for pricing” reads the newsletter. Motorola’s words reflect a warm, intimate and trusting welcome, but it will take much more than words to convince customers to walk through the door. 3. The low-end and mid-range segment is very price sensitive and is mostly driven by internal OEM efforts. If a full operating system stack costs $5 for a high-end feature phone, how much can Motorola command for an OS for low-end phones ? The only upside to this is the tens of millions of shipments associated with low-end handsets. 4. The window of opportunity is closing. Most OEMs have already selected their low-cost platforms, although Sony Ericsson and Samsung are still valid customer targets. The answer is no .. and that’s why Moto’s strategy won’t be going far. All in all, acquiring TTPCom was a sane decision for a manufacturer like Motorola with an in-sourcing platform strategy. Most certainly the acquisition will help Motorola strengthen its IP portfolio in 3G+ and convergence products. It comes at the right time, given Motorola’s increase of market share, confidence and R&D spending following the wild success of the lab project that became RAZR – something which reflects on the hiring frenzy at TTPCom’s Cambridge offices. However, putting money into building an ecosystem and selling to competing OEMs seems like money down the drain. #ajar #motorola