Survivor's Guide to 2005: Mobile and Wireless

Mobile Devices

On the surface, wireless-capable mobile devices, from notebook computers to smartphones, sport a certain elegant simplicity. But we know that designing and managing a wireless data infrastructure is complex--under slick designs and vendor vows of seamless interoperability lurk a polar bear or two waiting to swallow you alive.

In 2005, watch for the most significant platform shift since the emergence of the notebook computer more than 10 years ago. Don't get the wrong idea: Notebooks aren't going away, nor will their refinement stall as they make the transition from desktop supplement to desktop replacement. But providing access to business information on a handheld computing device is compelling.

The mobile platform of choice for the information elite in 2005 will be the smartphone, a handheld mobile communication device that integrates cellular voice services with packet-based data applications, including messaging and Web services. Smartphones are rife with trade-offs--most notably speed versus battery life, and screen size versus portability--and the range of alternative embedded wireless capabilities makes finding the perfect fit for your users a bit of a shot in the dark.

Our advice: Look for devices that strike a compromise between the key trade-offs while providing a single platform that integrates Bluetooth (for connection to headsets and integrated vehicle systems), 3G (GSM/EDGE or CDMA for wide-area voice and data) and Wi-Fi (for high-speed data). Many end users will eschew these devices in favor of miniature cell phones, but others, especially those accustomed to using PDAs, will happily accept a phone-PDA that's a little large for the luxury of having to carry only one device. This isn't just a matter of physical convenience. Integrating messaging and personal information management on a single device is a big win in terms of personal productivity.

As to choosing a vendor partner, unlike the desktop and notebook market, where Microsoft reigns supreme, there's intense competition in the smartphone platform space among Microsoft, PalmOne, RIM (Research in Motion), Symbian and others. Competition is good, but it also leads to application challenges. Although standards-based browsers and e-mail clients provide a common foundation, there's no skating around the need for multiplatform client support, not only for applications that run on the devices, but also for the middleware that integrates these devices into enterprise information systems. We expect to see significant maturation in all these markets in 2005, but the fun is just beginning. We'll be fighting these battles for the rest of the decade.

Last year, more than 500 million cell phones were sold worldwide. Were Dick Vitale a mobile communications pundit rather than a college basketball analyst, he'd probably have this to say about the market: "It's all about voice, baby!"Cellular voice protocols will continue to dominate in 2005 and beyond, but wireless VoIP is beginning to make inroads into what could become a huge market. In fact, it's not much of a stretch to predict that by the end of the decade, wireless VoIP will eclipse more conventional architectures. Eventually, voice becomes just another data application.

Initial deployments of wireless VoIP will focus on vertical market segments, such as health care and supply chain, where internal business processes can be improved through converged mobile devices. Beyond the numbers game of conventional PBX versus IP PBX for new installs, considerable ROI comes through the delivery of compelling new applications, including integrated messaging and the capability to provide users with a single phone number that follows them everywhere. As the VoIP market matures, so too do the applications for this technology: We expect 2005 to be a breakout year for wireless VoIP--if not in market volume, then in the creation of next-generation devices, standards and back-office systems that will enable future deployments.

Beyond voice, e-mail is the next most compelling application for mobile devices. RIM has carved out a nice lead in this area by providing reliable and functional purpose-built devices with very usable keyboards, and many IT folks have a good comfort level for integrating the BlackBerry Messaging Server into their Exchange environments.

Still, even as RIM continues to enhance its offerings, its competitors are catching up. All the cellular carriers have mobile e-mail service plans, with many offering choices ranging from fully managed services to middleware-oriented, behind-the-firewall options. Closely related to e-mail is the trend toward instant messaging, a technology that has caught on with younger professionals as well as in organizations that rely on physically distributed work teams. Look for significant improvements in mobile messaging in 2005.Beyond wireless voice and messaging, it's a wide-open field. Most enterprise application developers, including purveyors of ERP (enterprise resource planning), CRM (customer relationship management) and SFA (sales force automation) tools, are working to enhance application delivery to mobile devices. In addition, there's a wide range of specialized mobile applications that don't yet have a well-defined market niche. That's the landscape for future innovation.

Wi-Fi

Wi-Fi products, mostly conforming to the 2.4-GHz 802.11g specification, continue to sell well at leading retailers. Twenty percent of SOHO broadband users now use WLAN technology, up from 13 percent last year, according to Dell'Oro Group estimates. That growth is putting more pressure on enterprise IT managers to deploy their own infrastructure--once people have wireless at home, they want it in the office as well and get frustrated when IT tells them it can't be done. IT managers who resist the avalanche not only will lose favor with internal customers, they may also spur rogue deployments, which are never fun to eradicate.

While debate continues about the ROI associated with enterprise Wi-Fi, most IT managers are realistic about the need to deliver wireless service, not only for internal employees, but often for visitors as well. Although the market is maturing and component costs are decreasing, the cost of deploying a secure and scalable enterprise-class Wi-Fi system is still quite high. For organizations that have not yet committed, the choice in 2005 will be whether to take a tactical or a strategic approach.

Tactical deployments focus on public spaces, especially conference rooms, where laptop-toting professionals often demand network service. In essence, this is a private hotspot service with the goal of serving the needs of nomadic, rather than truly mobile, users. And because the primary applications for today's Wi-Fi hotspots are e-mail and Web access, performance is not as great a challenge. You can often accommodate 20 or more concurrent users on a single access point that covers lots of square feet.Denser deployment of APs may be necessary to meet scalability requirements, but such designs also present challenges. When you've deployed more than a thousand APs, you gain membership to the mega-Wi-Fi club, and you've probably developed a keen appreciation for the challenges associated with making such a system work. You'll worry about an array of management issues ranging from configuration and monitoring to security and roaming. All the leading enterprise Wi-Fi vendors address these concerns, with varying degrees of sophistication and polish. Perhaps most important, you'll need to consider the myriad challenges associated with integrating wired and wireless network services.

This may lead you to favor your incumbent wired vendor for wireless services, but you shouldn't assume that just because you're dealing with one vendor, your integration challenges are solved. It's much more complicated than that. You need to have a clear understanding of which services require integration. Wired and wireless products sold by a single vendor don't always include seamless integration. And even when they do, the cost of sticking with a single vendor's gear may be higher than that of competitive offerings.

Wi-Fi has a virtual lock on both the home and enterprise markets as a foundation for wireless data service, but once you venture outside their rooms and corridors, the picture gets fuzzier. It's understandable that the industry would try to leverage the ubiquity of Wi-Fi-equipped mobile computing devices, but "Wi-Fi Everywhere" might not be the best solution.

Clearly, there's demand for Wi-Fi outside the home and enterprise, evidence of which can be found in the hotspot market. As of Dec. 11, 2004, there were 71,076 hotspots at 51,214 locations in 91 countries, according to JiWire, a Web-based hotspot-tracking service. Those numbers sound impressive, and in many respects they are, but the physical coverage associated with all these geographic areas probably represents less than 1 percent of all possible metropolitan locations. Providing a coverage footprint equivalent to that of cellular voice would require millions of hotspots.

That's only a small part of the problem. The bigger issue is coordination, not only of access-control and accounting systems, but also of the unlicensed spectrum over which these systems operate. Voluntary self-regulation efforts are under way in some areas, but it's almost impossible to imagine a future broad-coverage environment where interference between overlapping systems doesn't become a huge problem.This issue and others have led a number of local governments, from Philadelphia to San Francisco, to propose plans for public metro-area Wi-Fi deployments. However, the technical and political obstacles are immense. How will such networks be funded over their life? (Think about potholes before you answer that question.) How will governments provide widespread coverage while avoiding interference with private residential Wi-Fi deployments? What effect will these publicly funded networks have on the motivation of private enterprise to expand conventional hotspot deployments? The idea of broadband wireless data as a public good is admirable, but it hasn't been thought out very well so far. This public-policy debate should escalate dramatically in 2005.

And of course, Wi-Fi isn't the only game in town. Commercial rollouts of advanced cellular data services continue to gain momentum. Spurred by industry consolidation and carriers' desire to increase their per-subscriber revenue stream during a period of decreasing cellular voice costs, these services enjoy the luxury of operating in licensed spectrum.

Although this affords service providers the advantage of not having to deal with thousands of unlicensed devices interfering with their spectrum, it also introduces challenges. First, because the market is nascent and fragmented across two competing technologies (CDMA versus GSM), equipment costs are high. For example, while Wi-Fi NICs are readily available for less than $50, an equivalent 3G modem costs more than five times as much. Second, while more than 90 percent of notebook computers sold in 2005 will likely come equipped with internal Wi-Fi, the market for embedded 3G wireless modems in notebook computers is nonexistent. Third, monthly service charges currently run about $80 for unlimited service, which is too high for individuals and most enterprises to justify. Finally, while licensed spectrum offers carriers protection from interference, the relatively limited amount controlled by each carrier leads to serious questions about service scalability, should costs be reduced.

Beyond Wi-Fi and 3G, there are other technologies competing for attention. These include both proprietary and standards-based offerings built around the highly robust OFDM (Orthogonal Frequency-Division Multiplexing) signaling technology used in high-speed WLANs. Successful mobile broadband wireless trials have been running for a couple of years, and the imminent release of products based on the IEEE 802.16 (WiMAX) standard has many dreaming that WiMAX may have the same effect on public wireless that Wi-Fi had on private wireless. However, despite technical and marketing support from Intel, 802.16 is still immature, and its market impact during 2005 will be minimal.Dave Molta is a Network Computing senior technology editor. He is also assistant dean for technology at the School of Information Studies and director of the Center for Emerging Network Technologies at Syracuse University. Write to him at [email protected].