A new report on mobile internet devices certainly refuses to share the habit of being over-impressed by the Apple iPhone. In its generally rosy forecasts for the MID sector (see separate item), Forward Concepts dismisses the iPhone’s claims to class as an MID, despite its undoubted impact on raising levels of confidence in the mobile internet. The research firm believes a true MID must have certain features even the new iPhone lacks, notably a 4-6-inch screen, VGA, TV-out and optional mobile TV support. This just highlights the fact that the iPhone has always lacked many features that are common in rival smartphones (like high end camera and GPS) but despite that, has helped push forward the boundaries of the mobile web user experience, and stimulate usage of this among its aficionados because of its friendly interface. And every time Apple sneezes, the ripples are felt throughout the emerging mobile web ecosystem, forcing the key players to adjust their positions slightly as they gear up to try to mold the shape of this new industry.
In the wake of the unveiling of the 3G iPhone, we have already seen an intensified stand-off between Apple and Adobe over the iconic handset’s refusal to support Flash; and rising tension between Apple and Google, which despite scoring a home run in mobile search ratings this year, is under increasing threat from other search offerings that are more optimized for mobile users. Google has shown off the most convincing prototype of Android so far, but still has a long way to go to prove itself as the mobile internet software engine of choice, especially with Nokia making new moves to dampen the belief that open source platforms will be commercially ready for the mobile web any time in the near future.
Opinions on the multimedia capabilities of the iPhone vary widely. On one side, some users desperately lament the absence of support for Adobe Flash. On the other, there is the view that Apple’s Steve Jobs is right, and that Flash is not up to the job of supporting such a sophisticated handset, and therefore subscribers get a better all-round experience from the iPhone, even if some content is off-limits. Now it seems Apple is developing its own platform to bypass Flash, using a technology called SproutCore, while Adobe is working on Flash emulation for the new iPhone.
Whichever line you take, this is becoming an epic stand-off that is as much to do with industry politics as user functionality ‘ both Apple and Adobe are in the front row of vendors that aim to drive the agenda for mobile internet standards, along with Microsoft, Nokia and Yahoo. Adobe, which has recently upgraded Flash Lite and announced a plan for a unified platform for developers that bridges the PC and phone, has also founded the Open Screen Initiative to increase backing for Flash to become a standard on all advanced handsets.
It has powerful allies, not least Nokia, and Sony Ericsson, which recently set up Project Capuchin to work on techniques to bring together Java and Flash on mobile platforms. But Microsoft remains ambivalent ‘ supporting Flash reluctantly on its mobile systems but also pushing the rival claims of its Silverlight system. And of course, Apple remains stubbornly aloof.
This proliferation of rich media and video streaming capabilities is certainly driving innovation and creating new choices for end users at a rapid rate. But it is also creating confusion among developers, who would rather build to a single platform and deliver to all smartphones, and risks denying some users important functions.
Adobe, of course, is desperate to come to a truce with Apple. This week, CEO Shantanu Narayen was asked whether Flash would make an appearance in the imminent 3G iPhone, though Apple has already said not, at least in the first generation. Narayen said on an analyst call: ‘With respect to the iPhone, we are working on it. We have a version that’s working on the emulation. This is still on the computer and you know, we have to continue to move it from a test environment onto the device and continue to make it work. So we are pleased with the internal progress that we’ve made to date.’
In other words, Adobe wants Flash to be included in the iPhone, but is not full of hope, and in the meantime will just have to offer these capabilities via emulation, a far more clumsy option for developers and users. Adobe has the difficult task of making Flash iPhone-friendly enough to lure Apple, without compromising its promise of being a single product, consistent across all handsets, and eventually the PC too. And it faces the even more difficult task of coming to a political agreement with Apple that will enable both companies to pursue a pole position in the mobile internet market, without treading on one another’s powerful but sensitive toes.
In the mean time, Apple itself is working with SproutCore, an open source Javascript framework that should be powering the new suite of MobileMe applications for the new iPhone. This suggests that it is trying to build its own rich internet applications system, bypassing Adobe, based on its Safari browser, SproutCore multimedia, QuickTime and H.264. Apple says it does not want to be locked into one standard for rich media and that Flash does not fulfil all the requirements of the iPhone’s more demanding users.
SproutCore is a framework for building applications in JavaScript ‘ the main language used for Ajax, widgets and rich internet applications (RIAs) – with only tiny amounts of code. It supports ‘thick’ client applications in the browser that can create and modify data, often independently of the web server, communicating with the server via Ajax only when they need to save or load data. Relying just on JavaScript and HTML frees Apple from dependence on any third party technology, such as Adobe’s, that it regards as over-powerful and a threat to its own desire to run its value chain with an iron hand ‘ and helps it continue to lock in its own users and software houses. However, this remains a gamble, since SproutCore is an unknown and unproven product and many developers, despite the clout of Apple in its own programmer base, will prefer the safe havens of Flash or Silverlight.
Of course, there is a serious question over whether Apple really wants to create a universal platform like Android, based on its own favorite technologies, even if these are open source. Would it rather accept a ringfenced influence over the mobile web ecosystem as whole, but instead keep its own developers and environment tightly controlled and highly profitable, and then take that semi-closed platform and present it as a vastly superior option for the high end, high margin end of the mobile web market ‘ using the old Windows argument, that a framework driven by a single vendor may not be open, but at least is not fragmented and constantly delayed by vendor in-fighting. (See analysis section for more).
If SproutCore is to be a Flash killer, Apple will deliver a platform on which a broad base of developers could theoretically deliver RIAs. But the manufacturer has so far seemed indifferent to massive developer adoption (only 1.5% acceptance rate for the iPhone developer program) and more focused on using tools like SproutCore to create a strong platform to deploy online applications that specifically support the Apple brand.
Apple, Google and Microsoft are the companies that are most reluctant to be tied into Adobe Flash, as RIAs centered on JavaScript and HTML take center stage ‘ and each of them is taking a different approach to reducing the power of Flash, for so long the only viable technology for RIAs, especially on handsets. Google’s attempt is bound up with Android, also a platform made up of a collection of open source elements and driven by the search giant itself plus its supporters, the Open Handset Alliance (OHA). This week Google showed a working prototype of Android at the CommunicAsia show in Singapore, and previewed a series of applications (most still under active development), which it boldly claimed would ‘revolutionize’ mobile communications by driving down handset prices and sparking an
“eruption of innovation”. A commercial offering should be on the market by the end of this year.
Whether Apple chooses to venture out into the broad mobile internet fray or remain within its high end, high margin walled gardens, Google will certainly aim to take Android right into iPhone territory. This was highlighted last week when Google CEO Eric Schmidt was reported admitting that he must sometimes be excused from Apple board meetings due to Google’s involvement in Android and competitive conflict with iPhone’s software framework. Schmidt was asked by journalists whether the growing competition between the two companies might prompt his resignation from Apple’s board. “It has not so far,” was all he would say, and tried to insist that Android-based devices and iPhones would be in very different categories.
The working demonstrations of the Android prototypes this week begged to differ. Google’s Andy Rubin, speaking in Singapore, showed Android on a handset ‘ with trackball and a touchscreen similar to the iPhone’s ‘ from an unidentified manufacturer. He showed screen touches to ‘flick’ through three desktops, to add icons and widgets, and to look around a location using Google Maps in the same way as using a camera viewfinder.
Rubin said the mobile internet will only boom when software development costs on mobile devices are reduced, and claimed that, once Android appears on large numbers of handsets, there could be cost savings of one-fifth overall. ‘We think about 20% of a mobile’s cost is software, and because Android is free and open, that means close to zero cost on the software side,’ he said. He added that Google expects Android to be extensively customized, with operators, developers and end users creating new iterations and applications, and with mash-ups being the central concept underpinning the service.
One forthcoming application will come from Enkin and uses a device’s camera, GPS, motion sensors and numerous web services to lay and display location services over live imagery, delivering a kind of ‘X-ray vision’ ‘ for instance, viewing buildings ahead even through vehicles. This type of product shows what Google needs in order to gain differentiation from those competitors that boast stronger and deeper understanding of mobile specific features and of services optimized for mobility. Nokia leads this camp, of course, and rather than emphasizing the goal of ‘just one internet’ that is the same across all platforms, tends to focus on the additional functionality and usability that a specific mobile implementation can deliver. The most important example in real commercial markets is the integration of location-based services into web services at the sort of sophisticated level that the Enkin innovation promises ‘ and which reflects many developments taking place among Nokia’s and Yahoo’s developer communities.
Such mobile-oriented enhancements will be essential if Google is to win the battle for the dominant mobile search engine, one of the key foundation stones that it requires in order to transfer its online advertising business model, and its market share, from the PC to the handset.
So far it is making good progress. According to Neilsen Mobile Research, Google claimed 61% of mobile searches during the first quarter of 2008, with Yahoo in second on 18%. However, usage rates remain low, which many observers attribute to a clumsy user experience. Google searchers used the facility, on average, just nine times a month, and Yahoo users 6.7 times. This shows that the opportunity is wide open for a company that can provide a search engine that is more friendly and more functional on the handset, something Nokia, among others, is promising, focusing heavily on integration of search with location awareness, and on the different norms of the mobile internet, compared to the PC. Google knows that mobile search has different requirements – for instance, searches need to be more focused on the answer to a question, rather than a whole list of relevant sites. So far, Nokia has made more progress in this direction than the search leader, but Google and, to an even greater extent, Yahoo, are not sitting back. For instance, Google has addressed the fact that a mobile search takes more than three times as long as a PC-based one, by introducing
application-based or idle screen search.
As Yahoo and Google agree on an ad-sharing deal – which will see Google ads displayed alongside Yahoo search results and other Yahoo online assets ‘ in the mobile world the smaller company still has a real opportunity to leap ahead of the leader. While we might still hold out for a ‘dream team’ of Yahoo working with, rather than competing with, Nokia, in the short term its strategy is mainly focused on getting closer to operators, some of which are suspicious of Google because of its direct-to-consumer and alternative network activities. Building on its expanding reach via cellco alliances, Yahoo announced a series of new mobile initiatives targeting the Asia-Pacific region, with five new oneSearch partnerships that extend the product’s reach to 600m consumers. Yahoo signed these deals with India’s MTNL, Hong Kong CSL, two Philippines operators (Smart Communications and Digital Mobile), and Taiwan’s Vibo Telecom. In addition, Yahoo announced mobile advertising partnerships with India’s Idea Cellular and Malaysia’s Maxis as well as localized versions of its flagship mobile offering Yahoo Go 3.0 for India, Australia and Southeast Asia, including a local language version for Indonesia and an English version of oneSearch with voice that recognizes regional accents. Yahoo has now signed over 60 international oneSearch partnerships in the last 18 months, with 23 mobile search deals across the Asia-Pacific market.
Amidst all this jockeying for position, and attempts to woo developers and end users to a particular mobile web services platform, all the candidates remain largely unproven in wide scale commercial deployments. This is a point that Nokia frequently drives home, insisting that, with its long experience in real world mobile business models and developer programs, it has a significant headstart in bringing internet services from slideware to real, profit-making applications. Although Nokia is stepping up its activities in Linux and open source, it is still inclined to position its Symbian/Series 60 framework as a superior platform to create web smartphones, compared to the still fragmented and immature Mobile Linux and surrounding initiatives like LiMo and Android. While apparently accepting that Linux and open source will be dominant in mobile internet one day (parts of Series 60 are already in open source, and the software could be ported to Linux in future), in the short term Nokia wishes to portray Linux as commercially too risky. Open source developers targeting the mobile space need to learn business rules, and fully to understand issues like digital rights management, Nokia’s software chief and head of open source activities, Dr Ari Jaaksi, told the Handsets World conference in Berlin, Germany.
His theme was that the open source community needs to be ‘educated’ ‘ in the way the mobile industry currently works, because the industry has not yet moved beyond old business models. “There are certain business rules developers need to obey, such as DRM, IPR [intellectual property rights], SIM locks and subsidized business models,’ he said, all of which ‘go against the open source philosophy”, but remain necessary components of the current mobile industry. “Why do we need closed vehicles? We do,” he said. “Some of these things harm the industry but they’re here. These are touchy, emotional issues, but this dialog is very much needed. As an industry, we plan to use open source technologies, but we are not yet ready to play by the rules; but this needs to work the other way round too.”
Nokia and Linux:
Nokia’s main open source activities are its acquisition of Trolltech last year, and its involvement in Maemo, the Linux-based OS that runs on its N800 internet tablets.
In his presentation, Jaaksi said Nokia had learned key lessons from its work with the Maemo developer community. The key lesson: “Don’t make your own version. The original mistake we made was to take the code to our labs, change it and then release it at the last minute. The community had already gone in a different direction ‘ Everybody wants to make their own version and keep it too close to their chest, but that leads to fragmentation.”
Meanwhile, Trolltech makes Qt, a graphical toolkit that is used in the KDE Linux desktop environment and in much commercial software. It is a non-participatory member (so far at least) in LiMo, but LiMo and Maemo are both using the GTK+ toolkit and Gnome desktop environment, not Qt or KDE. This could lead to Nokia pulling Trolltech out of LiMo and setting up its own framework initiative as yet another alternative to LiMo, Android and the single vendor platform initiatives (Adobe, Microsoft, Apple). Jaaksi did not enlarge very far, except to say he wanted to see KDE and Gnome come closer (no doubt under Nokia’s auspices). While blowing Symbian’s trumpet as the most functional mobile internet OS for the short to medium term, he also told ZDnet that his company was “only now” able to start thinking about what to do with LiMo, and that Nokia had “a huge responsibility from a desktop and user interface point of view to see how we play our cards”.
The browser wars:
Microsoft’s Internet Explorer browser has never achieve anything like the position it enjoys on the PC, in the mobile world, and limitations of the traditional PC browser have often been blamed for slow uptake of mobile web services. In the early years it was faced by WAP interfaces based around the Openwave browser, and then as devices became bigger and more powerful, full browsers were enabled, but Opera was there to deny IE a dominant place. Now Opera has embraced Ajax and other Web 2.0 technologies, and several contenders from the open source world are becoming important, including Apple’s Safari and now the new version of Mozilla Firefox.
Mozilla launched Firefox 3 on Tuesday, and crippled its own network in its attempt to break the world record for the number of downloads in a day (target 5m). Despite teething troubles, though, Firefox is gaining ground on IE in the PC environment (it now has 18% share), and later this year its mobile version will debut, promising major impact on the still nascent mobile internet sector. With the industry moving towards open standards and open source, Firefox will be well positioned as it is designed to work smoothly with other open source software, and so will be a natural choice to be supported on the new breed of handsets and frameworks.
The number of high performance browsers on new mobile phones should increase from 76m last year to 694m in 2013, according to predictions by ABI Research. The handset makers will have a major influence on which ones gain market share, unlike the PC manufacturers, because even in the era of open internet access from wireless networks, the browser will remain optimized for certain platforms ‘ as with iPhone/Safari. Nokia will, of course, be the main kingmaker. It uses a range of browsers on different devices, including Opera, its own text-based products, and emerging graphics-oriented developments, some emanating from its open source joint development with Apple, based on Safari.
As well as Nokia, the Linux-based industry bodies, vying to create a standard platform, will also be influential. Last month, Mozilla became the first browser specialist to join the LiMo Foundation, but it is not included in Android, while Apple’s nascent full web platform, of course, incorporates the company’s Safari.