With already over three times the number of mobile phones on the planet than desktop or portable computers, this article The Mobile Web Explained addresses why the Web was destined to go mobile.
Table of Contents
There are challenges, reminiscent of the early days of the Internet where lack of coherent standards and the melee of big business encouraged divergence instead of a coherent, logical and structured level of technological development. In other words, access to the Internet has been available for mobile users but the huge number of different platforms – basically phones or handheld devices – meant actual Web content delivery has been extremely patchy, often with poor page layout, cumbersome navigation and, at worse, terse error scripts and no content.
WAP (Wireless Application Protocol) was created as an open protocol suite designed to facilitate a richer and more stable user experience for users of mobile devices wishing access to the Web and other services such as email. Unwired Planet was founded in 1995 and tasked to develop handset-based access to the Internet and by June 1997 had launched the first WAP draft, during which time – and after much hard negotiation – the WAP Forum was created by manufacturers and vendors from across the wireless sector to develop, monitor and push forward WAP devices. By September of that year, the mobile communications industry was committed to the specification and began ramping up production of WAP-enabled devices to meet a 100-million subscriber base.
Unfortunately, access to the Web proved to be a disappointment. Great expectations were dashed by patchy connectivity across the cellular network, premature hardware and software releases resulted in system (cell phone) crashes and lock-ups, after which users were frequently forced to remove SIM chips to resuscitate their phones… the user experience was poor. Moreover, the user paradigm shift necessary to view the colour desktop web on a tiny monochrome monitor proved too great for many. The mobile Web was a novelty. In the rush to capture market share the real needs of the user were passed over in the marketing and technical delivery exercise.
What was needed was a concerted effort to unify both delivery and development practices and a clear understanding of user expectations; give mobile phone users a solid experience with valuable applications delivered under an umbrella of consistency.
Lessons were learned and the industry moved on.
WAP 2.0 built upon the earlier standard for mobile and cell phone mobile internet access and, as well as offering more robustness and dispensing with the earlier gateway protocols, opened up the opportunity for more sophisticated standards-based markup. While WAP 1.x used WML (Wireless Markup Language) and was XML compliant, WAP 2.0 went further with an enhanced subset of XHTML (eXtensible Hypertext Markup Language), XHTML-MP (Mobile Phone). To complement the move to a more feature-rich language set and in line with the modern practice of estranging content from the format, a subset of CSS (Cascading Style Sheets) was introduced as a part of the specification, CSS Mobile Profile.
The adoption of XHTML and CSS was a partial endorsement of the W3C (World Wide Web Consortium) who proposed the Mobile Web Initiative guidelines in an effort to promote methodologies of best practice for device manufacturers and software developers. The W3C further set out a guide to Mobile Web Best Practices which aims to address current and future issues and limitations with cellular and WiFi device connectivity and constraints as well as recognising transitional challenges from desktop web access to mobile web device accessibility.
To further encourage standardisation across different platforms W3C released in May 2006 a working draft of DIAL (Device Independent Authoring Language). While embracing existing development standards like XHTML and CSS, DIAL proposes a methodology to present content dependant on device capabilities (limitations, perhaps). It offers developers the opportunity to restrict content where, for instance, no graphical capability exists and importantly, does this at the server level thus reducing unnecessary bandwidth overheads.
DIAL works by utilising Media Queries, an earlier set of W3C recommendation designed to more specifically target device capabilities but builds upon the concept by using conditional device-specific branching to omit sections of markup which may cause problems for certain devices.
Rhys Lewis, Chair of the Device Independence Working Group – who oversees DIAL’s development – noted that “The DIAL suite makes the most of existing, established web technologies and principles, and extends their reach with new features that better serve the needs of diverse devices.”
The advent of DIAL and the concept it promotes – that of device identification – is not new. Developers of the Mobile Web have long sought a technique – especially an endorsed, standardised approach – to serve content based on device capabilities. WALL uses Java conditional tagging to discriminate devices (or at least their functionality) against a device database. The markup code tag, for instance, ‘ ‘ is shuffled off to the WALL database library which in turn returns the format best understood by the target device – which could BR, dependant on the flavour of (X)HTML the device understands.
Now while this may sound an arduous path with inherent processing and thus time lag (latency), the end result is a far more selective and thus accurate delivery of content to any device supported in the database. It also offers a pretty good shot at reliable, serviceable web pages.
The advocacy of development standards resides not only with W3C but the industry itself made up of service providers and manufacturers have their own standards body, the Open Mobile Alliance (OMA). Formed in 2002, the OMA is a consolidation of initiatives to provide a framework of standards and drivers to promote interoperability of services across countries and service providers to meet the requirements of mobile users. It comprises wireless vendors, IT companies, mobile operators and application and content providers from across the globe. Businesses supporting the OMA work towards generating new technologies and services within a forum where interoperability issues can be resolved speedily.
The OMA supports several technical specifications including DTDs (Document Type Definitions) which describe the proper format of markup for the selected target device.
Now that acknowledged recommendations and standards are in place one would imagine a developer writing for the mobile Web would be confident in producing a website viewable across a range of mobile devices. This is true; to an extent. But a number of decisions must still be made.
XHTML and CSS
For developers versed in standards-compliant markup the most immediate and obvious opportunity to render an existing site for the mobile web is via the addition of an alternate stylesheet. W3C mobile phone CSS offers mobile-specific presentation, a subset of existing CSS. Handheld stylesheets are added easily as an alternative media presentation layer in the Meta markup of a web page but must be coded to observe the specifics of the limited selector (presentation tags) feature set and in some instances, as with the Opera browser, are not always given appropriate precedence… they’re ignored.
In a purist world the addition of alternate mobile stylesheets represents the better and more device- and platform-independent method of delivering an existing website to mobile devices. Unfortunately, there are as yet few devices whose operating system and/or installed browser recognise them, or which do so faithfully.
A far safer method is to deliver content bereft of stylesheets in an XML format that WAP 2.0 will understand, with a marginal fallback such that those users of archaic cell phones employing WAP 1 will not choke on.
Plain old HTML sites, stripped of any presentational layer (CSS or inline styling) will render well provided they are properly marked up. Immediate advantages are the broad device base and likely unambiguous rendering this offers since the mini-browsers (basic browsers offered as a default by manufacturers) will unlikely baulk at the content. However, pure HTML will never take advantage of the compression and delivery optimisation speeds inherent in Symbian XHTML devices and carrier technology and may incur a download overhead deemed unacceptable to those users whose contracts demand payment by the kilobyte for web access.
In May 2006 the .mobi TLD (Top Level Domain) was released and trademarked companies began registering their websites. By August of the same year, the domain suffix would be thrown open to the public for general domain registration.
Approved by the Internet regulatory body, ICANN in 2005, DotMobi was originated specifically for the Mobile Web in an effort to satisfy the perceived urgent needs of companies to capture the anticipated burgeoning traffic as individuals and businesses worldwide access the Web from a multitude of mobile devices.
However, the release of the new TLD has not gone without criticism since many in the industry consider the opportunity a threat to the Web as a division of content, encouraging developers to build device-specific content thus splintering the Web. Tim Berners-Lee, head of W3C publicly voiced his dissent:
“I thought the .mobi domain was a bad idea. I still do.” (Sir Time Berners-Lee)
Nevertheless, W3C is actively working with promoters of DotMobi in an effort to standardise the development of websites intended specifically for mobile devices.
The pros and cons of building to specific devices
With the .mobi domain, name developers have been offered a devil’s advocate: build websites specifically for mobile phones or revamp existing sites in the hope they will pass muster. While this may present a beguiling chance to dispense with dilemmas inherent in delivering existing web content to cell phones and PDAs, it also represents content aberration: the Web will be forced to observe device dependence, anathema to its founding precepts, that of content available to all peoples on any device.
Building to a specific device – as DotMobi encourages – has immediate attractions. Websites can be tailored to the mobile environment. Content optimised for small screens. Delivery and interaction will be far swifter without extraneous markup. And, importantly, the user experience will likely be far more enjoyable because the type of content will be geared to user expectations.
Of course, this could all be achieved without building a specific DotMobi site by purposing the site for mobile devices under a .com or other TLD but it is likely mobile browsers will afford precedence to the domain suffix simply through belief in functionality.
The final alternative for developers is to build for the Web and leave the rendering to the mobile device. While this may sound attractive it does present problems, if nothing else with the verbosity level of information delivered.
Both Nokia and Opera offer small screen browsers which offer viewing of websites not specifically written for mobile devices. While this may be innovative (certainly with respect to navigation) and in line with the vision of a single device-independent web, it does not address the issue of device nature: delivering content appropriate for use.
Apart from the wireless networks proliferating for mobile phones, other forms of wireless provision were under development and these were mostly concerned with the transmission of information to and from PCs. Contemporary WiFi (Wireless Fidelity) networks using the 802.11 standards run at far higher frequencies (think speeds) than do mobile or cell phone connections, anywhere from 2.4 GHz to 5GHz.
WiFi networks are currently used to connect PCs to either LANs (Local Area Networks) or directly to the Internet. WiFi hotspots – located at airports, hotels, internet cafes, shopping malls, business locations – offer reception coverage from an office area to a block or, as overlapping hotspots, for many miles. However, unlike cellular connectivity, WiFi hotspots cannot handover transparently and so do not offer the roaming capabilities available to the majority of current mobile phones.
Read about the world’s Internet speeds
But what WiFi does offer, especially for localized business communities, is VoIP (Voice over Internet Protocol), the ability to conduct voice conversations over the Internet or, in fact, any IP-based network, significantly reducing the costs of human telephony, often permitting free phone calls.
VoIP allows users to make and receive calls anywhere in the world provided an internet connection exists – but it does have challenges: ensuring data packets are received in the right order whilst minimising jitter and latency problems causing unnatural pauses and bursts of dialogue. Nevertheless, VoIP will no doubt impact classical telephony as the technical challenges are ironed out, especially once data compression and the new breed of dual-mode handsets penetrate the market allowing both WiFi and cellular connections to be dropped and swapped transparently. Of course, traditional telephony providers will respond to the inevitable revenue losses as VoIP takes hold.
Bluetooth developed as a 2.4 GHz wireless transmission and arbitration protocol by Ericsson enables a range of devices to communicate with each other over distances of up to about 10 meters. Though of little direct use in mobile web, Bluetooth enabled devices can act as a hopper service to WAP-enabled devices.
Nokia, in the same month, released the source code for their S60 mobile browser in an effort to encourage take-up by developers but also to capture the existing web market. The browser enables mobile phone users to view existing websites in much the same way as they might on a PC.
Quite apart from operating systems and web development, delivering content across the airwaves requires
much thought. A number of technologies are being developed to surmount problems inherent in the limited bandwidth – the amount of space available to broadcast information – available to services providers.
Live, streaming video broadcasts like television or commercial music video require considerable bandwidth. Current thinking is a realistic rate of 15 frames per second is satisfactory, given the size of most screens but even here it is proving difficult to deliver stutter-free content, especially where users may – by their very nature – not be stationary when receiving content and will hop from one transmission cell to another when, for instance, travelling by train or car.
Other fears concern privilege levels or ‘network neutrality’ of content. Service providers are in hot competition and, with broadcast space at a premium and demand high, must make tactical decisions about what to deliver, where and to whom and, significantly, at what price.
Ever since the birth of the World Wide Web, it has spawned a number of web browsers to serves the needs of users. Some, like Netscape Navigator, have fallen by the wayside or been enhanced and superseded to become mainstream competitors to the ubiquitous Internet Explorer, like Firefox, and Google’s Chrome, now commanding a healthy market share. Other browsers exist – Opera, Mozilla, Safari, Konqueror, Lynx – serving different PCs and operating systems.
The adoption of standards developed by W3C has meant the web development community now enjoys a relatively clear reference set offering both software vendors and developers guidelines to produce websites and services which run across platforms, operating systems and browsers with little need for target-specific tweaks. Conformance and the adoption of W3C standards means a well-designed website will run across browsers.
Mobile web browsers
The mobile web or, more specifically, the mobile browsers and devices used to access the mobile internet and other media content is undergoing the state of flux once experienced by the early Web.
Much of the problems arise from the number and diversity of mobile devices available, probably some 50 or more. The platforms of computing quickly evolved into two camps: IBM-type PCs and the Apple Macintosh range of personal computers. Equally speedily two operating systems established themselves, Windows and Mac OS for the respective architectures. There is no denying the different flavours of each presented challenges to developers as the OS browsers matured and were replaced leading to quirks requiring a little fudging for earlier versions but the trend was towards consistency of operation based upon W3C standards.
At the time of writing this article, many new mobile devices are known as smartphones and come equipped with operating systems enabling considerable functionality beyond that of basic phone services. The power of the newer mobile chipsets is fast approaching that of early desktop computers with storage now on par with mobile computers such as PDAs like the Blackberry or Hewlett Packard’s range of mobile PCs.
Irrespective of the type and power of circuitry, the diminutive size of handsets presents usability challenges in terms of ergonomic design and functionality.
Openwave, a mobile phone and messaging services company, were one of the first operating system browser providers and today over a billion handsets have been supplied with their software. With the addition of their latest browser version, Mercury Edition, Openwave remain at the forefront of mobile browsing, with over 50 handset manufacturers using their software.
Nokia offers the S60 operating system and attendant web browser for many of its smartphone range models. Nokia partnered with Apple computers and adapted its Safari OS browser, WebKit, to meet the needs of the smartphone their OS. Interestingly, they took the step in May 2006 of making the S60 platform source code available to developers in the open source community.
Much of the appeal of the S60 browser is its ability to view websites not specifically designed for mobile devices. Intelligent navigation and site scaling mean users can move about a web page in miniature then zoom in on areas of interest.
Whether the developer community will take up the opportunity remains to be seen but with a predicted, 3 billion phone subscribers in the next couple of years and the commensurate enormous revenue streams, the opportunity for a level of conformance in smartphone development is not something likely to be ignored.
Like PCs, mobile phones and PDAs (personal digital assistants ) use operating systems through which to interact with the user. The newer systems built into today’s smartphones can accommodate a variety of services and applications.
Today’s range of smartphones is often powered by the Symbian Operating System (OS) who devices account for 85% of annual global mobile phone sales.
Regardless of which operating system in use, it must provide the functionality to the mobile user.
As discussed earlier, the Web will go mobile, if for no other reason than the number of mobile handsets in worldwide circulation. This presents challenges for the classical desktop web because it represents a threat of division of content by a demarcation of mobile and non-mobile websites.
The mobile device and service provider industry – cell phone, PDA and wireless deliverers, in other words, handheld devices and providers – is in a state of both flux and consolidation.
Emergent technologies are compromising traditional telephony providers. VoIP means a reappraisal of service contracts. Traditional web multimedia content delivery from the streaming providers now has overlap with mobile carriers who maintain expectations of live media across their network. Existing web producers have a vested interest in maintaining their subscribers; the mobile web and the service providers feeding content would have it otherwise. There exists a convergence of content which can be satisfied only by the broadband connections on which they are delivered. Ultimately this resides with the suppliers of web traffic, those responsible for the pipelines and transit media to which users subscribe.
A media bill passed on June 8, 2006, by the US House of Representatives called the Communications Opportunity, Promotion, and Enhancement Act (COPE) has effectively offered internet and telecommunications companies the opportunity to degrade of impair user services as they see fit by abandoning a regulatory amendment. In other words, they can discriminate web and content access to users as they see fit.
Network (cable and phone) companies are at liberty to decide what type of content and from whom can be carried across their networks.
And because of the limited bandwidths currently available profitability decisions will determine the type and quality of traffic available to users.
To further complicate matters, users will likely enter into a contract in the same way they do for standard mobile phone usage to receive web services or indeed connect to the Internet.
And because of the limited bandwidths currently available profitability decisions will determine the type and quality of traffic available to users.
A mobsite is… a mobile website. Nokia developed the idea in 2004 with their Mobile Web Server project. From a traditionalist perspective, the concept is interesting, to say the least. A fully fledged server on which resides a website ready to deliver location-specific content.
Understanding social networking on mobile phones
Imagine a roving reporter. They are location-specific, they make a report on their experiences in a particular venue, incident or holiday resort. Their content is then tagged and when an individual enters the report province they are presented with an automatic update alert on their mobile device. Locale-specific delivery is a prime candidate for mobile devices, notably when those individuals have immediate access to content; in other words, their mobile device is equipped with a server.
The mobile web is an inexorable trend which embraces the Web but, inherently, represents grave challenges and opportunities to content neutrality. Sir Tim Berners-Lee (founder and chairperson of W3C) constantly emphasizes neutrality of delivery but whether this remains a viable vision is questionable.
The mobile web represents big money to service and content providers, many of whom will unlikely care for constraints of delivery based upon standards when they might garner substantial revenue for any method which delivers content.
The Mobile Web Explained, Published 2008 by Sonet Digital.
Vincent Zegna is a consultant at Sonet Digital, a digital marketing agency that provides online marketing strategies that work across desktop and mobile devices. For further information: https://sonet.digital/strategy/