Internet content and services have to date been targeted at a de facto standard presentation device, a desktop computer. Regardless of differences within operating systems, browser technology, and version, the base-level content has been created in a uniform fashion and to a known and constant set of base requirements.
When looking to extend the Internet to the wireless world one is faced with a discontinuity and a range of mark-up languages and protocols for rendering Internet content on the range of form factors for wireless devices. As a result, the majority of the content that exists, aimed at the PCs form factor, is unsuitable for, and unable to be displayed upon, the full range and form factor of devices available.
WAP-The “First-to-Market” Standard
The Wireless Application Protocol (WAP) is becoming the de facto world standard for wireless information and telephony services on digital mobile phones and other wireless terminals. It is optimized for small devices and is based on the Internet client/server architecture. WAP introduces a wireless-optimized mark-up language, WML, which is a subset of HTML, but introduces new syntax. As a result, all content and applications must be supplemented with (or even rewritten to include) WML notation before the content can be delivered to the consumer. Thus, WAP only provides “Internet-like” access to information.
By contrast, in Japan, NTT DoCoMo’s i-mode (see box) technology actually is the Internet delivered over the IP protocol stack to the handset. HTML content intended for display on the handset requires only the addition of a few new commands which are specific to the navigation mechanism used. i-mode does not require special servers or gateways-or major rework of the existing web content-to be available to customers. The growth of the i-mode web sites is due to the ease for individuals to generate the markup language (Compact HTML) from standard web authoring packages.
I-Mode
In February 1999 NTT DoCoMo, Japan’s biggest mobile phone company, released i-mode, a mobile phone that allows subscribers to surf the Internet as well as make calls. People are already using the phone to check the news headlines, follow the stock market, and download the latest jokes. Since launch, independent service providers have been developing (and making money from) mobile e-commerce sites specifically to allow i-mode users to buy cinema tickets and manage their bank accounts.
The service launched with 67 application alliance partners which had links from the i-mode portal page, 7 search engines had developed interfaces for i-mode Internet web sites and thousands of voluntary i-mode web sites have been created by Internet subscribers and other organiations-all of which can be accessed directly by entering a URL. These partnerships have enabled the company to release services focused on the entertainment market. These include mobile Karaoke and “Chara-Pa” (virtual friends)-graphical representations of cartoon characters which the user can download for ¥100 per year. As at August 99 there were already 130,000 subscribers to this seemingly frivolous service netting its supplier, Bandai Corporation, ¥13m per year, of which NTT DoCoMo nets a 9% commission.
Six months after launch i-mode had 1,081,841 subscribers., and enjoys a 12,000 to 14,000 daily increase. The company achieved its target to acquire 4 million by the end of March 2000. The average monthly user spend for connection and data-packet transmission charges is US$12-15 which means when they hit their target NTT DoCoMo will be reporting base subscription revenues in the order of US$48m per month.
The commercial success of i-mode will lead to smooth introduction of IMT-2000 which will enable 64k-384kbps, high-speed access from mobile devices unleashing the possibility of delivering more services.
WAP has made significant progress since its inception in moving toward being an open standard, but certain obstacles still exist for an enterprise customer seeking to gain interoperability across various device types and across various implementations of WAP servers. Although initial work by the WAP Forum did not address certain issues (including security and interoperability), in sufficient depth for the broader user community to be confident of the outcome. The Open Group is already working closely with the WAP Forum working groups to address these shortcomings.
Despite the best efforts of the WAP Forum, traditional PC web content providers are faced with the daunting task of re-engineering their content so as to be able to meet the specifications of the WAP standard. Even if the human resources were available (at any price) the overhead this represents-particularly for the presentation of realtime data feeds-is quite costly, and feedback from one content provider suggested an unwillingness to accept the cost involved. This has created an opportunity for, and led to the creation of, wireless-specific portals and content providers who are scrambling to grab their share of the market before the giants can re-orient themselves.
In addition to the efforts within the WAP Forum, content translation and transformation engines are being developed to interactively overcome the quirks and inadequacies of various wireless devices. An example of this is U.K.-based Argo Interactive Group which is developing “web-to-WAP” software called ActiGate which, when used in conjunction with a custom-defined template, will perform the necessary translation in realtime.
Industry Convergence Forces Open Standards
The WAP Forum has further recognized that the WAP standard cannot exist in isolation from the “rest” of the Internet, and there is a visible agenda within WAP toward removing some of the proprietary approaches which were implemented as “necessary evil” in early releases to overcome limitations in the network and client technology. Indeed the ETSI-backed Mobile Execution Environment (MExE) definition has adopted WAP specifications as Classmark 1, with Classmark 2 clearly signalled as Open Standards including Java, IP, and use of common mark-up language. Recognizing the need to closely coordinate with the Internet community, the WAP Forum has signed a mutual cooperation agreement with W3C to ensure that the work between the two organizations is complementary in nature and enables seamless extension of Internet developments to the wireless environment.
The challenge posed by WAP is merely one example of a much more general problem which faces information providers. It may be acceptable to have specific point solutions to translate from standard HTML to WML for WAP, but this solution is not scaleable as the number of different device types and form factors proliferates. It is simply not possible for information providers to be aware of the specific characteristics of every kind of access device.
The holy grail for content providers is to achieve independence of platform, language, and media. Two complementary approaches are currently evolving.
A standards-based approach from W3C, referred to as XHTML, includes facilities to profile both the original content “page” and the characteristics of the device. A single version of the original information is translated either by the server or browser for display on the specific device.
XHTML is an evolution of HTML which exploits the strengths of XML. XML developed within W3C is effectively a language which allows the rigorous definition of languages for information interchange.
HTML and WML can both be viewed as “first generation” mark-up languages, primarily addressing the rendition of information. XML-based languages go further and allow the semantics of the information to be defined. XHTML is designed to be both human-readable and computer-processable.
There is a clear convergence path from both HTML and WML towards XHTML.
The second approach involves the introduction of policy-based intermediation systems, such as the recently announced WebSphere Transcoding Publisher from IBM. Such systems can and do utilize XHTML to profile content and devices, but are capable of much more, including protocol conversion and content filtering.
In rounding off this section, as often happens in our industry, Microsoft launched an alternate initiative with similar aims called BizTalk. It did this with the support of many enterprise resource planning (ERP) and e-commerce software vendors and service providers.
The Open Group should continue its close cooperation and support for the WAP Forum by delivering interoperability test suites and building a world-class certification program. As appropriate, The Open Group should try to bring to the attention of WAP members other initiatives which impact on the WAP standard, and will strongly encourage convergence with XHTML.
The Open Group should use its existing relationships with the suppliers of intermediation technology to ensure full integration with XHTML and common definition of extended transcoding rules.
The use of XML is evolving at web speed, driven in part by the growth of the wireless and mobile market. While the definition of XML itself is centralized and carried out within W3C, the definition of schemas, which define specific XML-based languages, is a distributed activity, being carried out by groups that understand the semantics of the information involved.
The Open Group is already a forum where industry groups are working on XML schemas for information exchange; this activity should be extended to handle the specific requirements and characteristics of wireless devices and mobility, working with other groups such as those above and OAG, OBI, OTP, OFX, and so on.