Technology is pushing out the boundaries of the possible. Its potential is intoxicating-except if you're the CIO, who's responsible for managing this technological event horizon as its boundaries become indistinct. But that's technology: The more it advances, the more tenuous its connections, kind of like the ripple that fades after a pebble is dropped into a pond.
To illustrate this concept, CIO looked at the past, present and future of four technologies-handheld computing, customer relationship management, outsourcing and peripheral connectivity. Unlike the ripple, however, these technologies are not dissipating. They're gaining momentum, even though they're a long way from the pebble that initially propelled them.
HANDHELD COMPUTING About eight years ago, pen computing was all the rage. Every company out there, from NCR Corp. to Grid Systems Corp., marketed tablet-and-stylus computers for the gamut of users, including mobile professionals and factory floor personnel-anyone who found keyboards inconvenient. Other companies, such as Apple Computer Inc. and Casio Computer Co. Ltd., sold handheld computers or personal data assistants because they housed calendars and to-do lists and contacts. But they all failed, either because they were slow or complicated or because Doonesbury lampoons made them laughingstocks.
Into this handheld graveyard walked a ghostbuster named Jeff Hawkins, a Grid Systems alumnus who reconsidered what handheld computers should be and founded Palm Computing Inc. "Before Hawkins integrated software and hardware together in the original Palm Pilot, systems were self-contained products that didn't connect to the desktop," says J. Gerry Purdy, president and CEO of Mobile Insights, a market researcher in Mountain View, Calif.
"People were misfocused on pen-based computing," adds Donna Dubinsky, who worked with Hawkins at Palm Computing and cofounded with him their new company, Handspring Inc., also based in Mountain View. "It was like saying 'keyboard-based computing' or 'mouse-based computing.' Jeff realized it wasn't about the pen-it was about making the device small enough that you could walk around with it."
People wanted devices that were small and ready to use, and would be complementary to PCs, not supplant them. As a result, the Palm computer had no printing capabilities. "When you put printer drivers in a small device, it slows it down," says Dubinsky, remembering that, at the time, the idea was so counterintuitive that she and Hawkins couldn't even find investors to fund the business.
Today, those are some very unhappy investors, since International Data Corp. (a sister company to CIO Communications Inc.) estimates that by the end of this year at least 5.5 million handhelds based on the Palm operating system will have shipped since 1996. In comparison, IDC estimates that in that same time frame some 3 million Microsoft Windows CE devices will have shipped since the operating system debuted in 1996. Like laptops and PCs before them, Palm computers have infiltrated the enterprise, and knowledgeable CIOs are taking advantage of their connectivity.
But we're still in the tin lizzie days of handhelds, maintains Purdy. "Wireless computing is tolerated more than it's embraced. When we get to the point that it's so easy to use that it disappears from our thinking process, then it will be pervasive." And we're slowly moving in that direction: The Palm V has a lithium-ion battery that automatically recharges in its cradle, whereas earlier versions use AA batteries that have to be replaced.
As for the future, Handspring is focusing on expandability, says Dubinsky.
Although when she and Hawkins first founded the company-whose products use the PalmOS-they expected to create new consumer devices. But they soon realized that CIOs and vertical markets were crying out for Palm-like computers that incorporate GPS systems, expanded storage, bar code scanning and other capabilities, and they promptly changed their focus.
And will the new cellular telephones from Motorola Inc. and Nokia Corp. that allow Web access do to Palm computers what Palm did to pens? Purdy says no.
"There are voice-centric and data-centric differences, and the form-factor differences of tall versus wide." Eventually, he says, the two will rely on technology from the Bluetooth consortium (www.bluetooth.com), which is working on connectivity between wireless devices. "You won't have to put all the features of a phone in a computer or vice versa; you'll just communicate between the two."
CUSTOMER RELATIONSHIP MANAGEMENT Technology has taken CRM far beyond just tracking frequent flyer miles. Companies can not only compile information on what customers buy from the Web, but CRM software lets them interact with customers (through e-mail or the Web), determine who's buying what and figure out how to create loyalty in customers' hearts. The key word here is ideally, since on the Web your customer is just a click away from your competitor.
Those who sleep and eat customer relationship management disagree on whether its heritage lies in databases, contact management systems or both. "It's contact management on steroids," insists Gayle Crowell, president of the E-Commerce Division of E-piphany Inc.
On the other hand, Martha Rogers, partner at the Peppers and Rogers Group in Stamford, Conn., and coauthor of The One to One Future: Building Relationships One Customer at a Time, believes that CRM descended from mass-marketing databases. Rogers cites four steps to CRM: identifying people, differentiating them, asking them questions and treating them differently based on their answers. "The first three [steps] are database marketing. It's when you take that fourth step, using the information to do something for that customer that [he or she] can't get without starting all over, then you've gotten the customer to add the value to the relationship."
Federal Express Corp.'s PowerShip system is a perfect example of snagging the customer, says Rogers. The shipping service lets customers label their packages and track their progress from their own desktops. "It's gotten you to do all the work, and you're thrilled about the whole thing." Also, FedEx now knows exactly how much you ship and when you ship it, and it could conceivably adjust its rates so that you are more likely to ship packages when it's easier for FedEx to handle the load. Moreover, because you're now invested in the relationship, you're probably not going to bring in a competitor to set up a similar system.
Crowell and Rogers also don't necessarily agree on the future of CRM. Rogers thinks companies will reserve one-to-one marketing for their best and most profitable customers. "The Web-based grocery delivery company does not want all customers the way its brick-and-mortar counterparts do," she says. "It wants the 12 percent of busy suburban families with two jobs, pets or a pool," noting that these are the people who replenish their sundries and groceries regularly, the kind who don't count pennies but do eat prepared foods-those with the biggest profit margins.
Crowell, however, believes CRM will go mainstream. "Customer expectations are going higher and higher at the same time markets are becoming more competitive," she says. "The mainstream customers want service that's personalized and customized. Technology is one of the ways you can do that."
But she takes the idea one step further, suggesting that eventually customers will be able to drive the creation of products they want through CRM interactivity. "When companies can do a better job communicating with the customer, they'll be able to create better products and expand their business.
It will be a competitive advantage to do so," Crowell says.
Crowell also envisions Web site visitors eventually choosing digital shopping companions, similar to avatars, to help them through a Web site. And they'll have a choice of avatars to match their personalities: career woman with kids, bachelor with pets and so on. "That becomes the personalization tool you give the customer," notes Crowell. "If it's nonintrusive and you're not going to misuse it, and you're going to use it to save the customer time, the customer isn't freaked out by it, and she doesn't have to fill out a form. It shows you're a company that respects privacy."
OUTSOURCING It all began with time sharing. You didn't buy a computer; you bought time on a computer, logging in and running whatever applications you needed. When computers were expensive, it was an inexpensive way to access computing cycles. Soon computers became less expensive, but some CIOs believed computing wasn't their core competency. That led to outsourcing, which took off just a decade ago.
"With outsourcing, you're talking about OPR-other people's resources-which have become more valuable than other people's money," says Frank Casale, president of The Outsourcing Institute in Jericho, N.Y. "It's easier for most companies to get cash than [it is to get] resources."
But even outsourcing is morphing. Its future is the nascent application service provider (ASP) category-in which companies pay on a subscription basis for someone else to run their applications and store their data, which they can access over the Internet. "Now you have startup companies-e-businesses that have no back office, don't want one and will never have one," says Casale.
"These companies are establishing relationships with service providers to build and own their infrastructures from the very beginning." In essence, he says, their first employee is really somebody else's employee who's in shipping or logistics or IT. This leaves the e-business-or any business-free to focus on the front end, the customers.
Casale says the ASP trend has been dubbed "back to the future" because it so closely resembles time sharing. But it has a brighter future than time sharing.
"Companies are realizing that process is king. The IT infrastructure, the software, the database, the ERP technology-they're the underpinning, the drivers. But it's not how people will make decisions. The technology becomes virtual because you don't care whether it's offsite or onsite, as long as the deliverable is good and the pricing is right."
"ASPs are doing for applications what the Internet did for data," says Traver Gruen-Kennedy, chairman of the ASP Industry Consortium (a group of 200 companies led by Citrix Systems Inc., and includes AT&T and Sun Microsystems Inc.). "They're providing affordable and universal access and ubiquity." The other trend driving ASPs is the advancement in the voice and data network. "As we move from desktop to net-top, people are going to be able to use the network the same way the backbone works in a PC," adds Gruen-Kennedy. "All the complexity of the networking is removed from the end user. The network is instant-on, affordable, and I can use it like water."
The trend toward ASPs is especially prevalent at the new dot-coms, says John Puckett, CIO of Toysmart.com in Waltham, Mass. Small companies want to be on the Internet without worrying about business processes, but they aren't interested in building the technology to get there. There's a paradox, though, Puckett notes. If you're a small company, you can't afford an ERP application, though you may want to use an ASP to get a piece of one, the same way you might lease a piece of a T1 line for telecommunications. "But ERP takes an awful lot of thinking about an individual company's processes, whereas economies of scale are geared toward a standard implementation."
In the future, expect more highly interwoven outsourcing partnerships. "It used to be that you got low pricing [on an outsourcing contract], and a year later you're nickel-and-dimed to death in the upgrades," says Casale. He foresees situations where the ASP will have its "skin in the game," in which if the company's profit soars, the ASP is compensated appropriately (beyond what it would get for average returns). Casale thinks ASPs have a right to ask, How come if we miss a deadline there's a penalty, but if profits double, we don't get another dime?
PERIPHERAL CONNECTIVITY Joe Beyers has been with Hewlett-Packard Co. since the Nixon administration, so when you talk to him about peripheral connectivity, he can recall some horrific stories. Back then every proprietary computer had its own plug-compatible peripherals, but the challenge and expense didn't stop there. "There were challenges even with the subelements of a single processing system," he notes. "In the HP 3000 minicomputer, if you wanted one element to talk to another, you had to write custom code and do major integration. Now we have object technology and communications protocols, so all that complexity is hidden."
Recognizing that, Microsoft Corp. has attempted to endow Windows with plug and play capabilities, but it hasn't been entirely successful: The company first announced it for Windows 95, but that operating system was no better than earlier plug and "pray" systems. Then it announced those capabilities for Windows 98, and the jury is still out on that.
In the meantime, Hewlett-Packard and Sun Microsystems are tackling the problem of peripheral connectivity on their own. Beyers, general manager of HP's Internet business unit, is in charge of the Chai (like high with a soft ch) Appliance project at HP, and Jim Waldo-now a distinguished engineer-was the chief architect of the Jini project at Sun. Both projects are aimed at enabling peripherals to automatically connect without device drivers.
"The intelligence should be in the device being talked to, not the device doing the talking," says Waldo. "The PC doesn't need the intelligence to talk to everything it might need to talk to." Besides, the more innovations you have, the smarter the PC needs to be in order to talk to peripherals. For CIOs, it's easier if you don't have to update everything in your network when you update a new device.
Naturally, Beyers and Waldo disagree about the utility of each of their technologies. Beyers notes that Chai is based on standard Internet protocols,whereas Jini is proprietary to Sun. Waldo argues that Jini is an evolution beyond Chai because as it becomes widely adopted in the next two years, it will encompass not just peripherals but services like e-mail and spellchecking.
David Usechak, program manager and chief software engineer for the Program Executive Office (PEO) of Command, Control and Communications Systems (C3S) for the U.S. Army at Fort Monmouth, N.J., looked at both technologies because such intelligent connectivity would be a boon in the field. "Mobility is a key factor for us. Continuously and in a random fashion, workstations will shut down, go away, come back and plug into a network." That brings complexity the Army doesn't need in battle conditions. For example, the Sun workstations the Army currently has carry about 50 printer drivers. "Wouldn't it be nice not to carry any of those?" asks Usechak. "If somebody had a new printer, and the Army decided to buy it, all I'd have to do is deliver it to the field and not worry about whether the drivers were there and [whether they] worked with that workstation." The last thing Usechak wants to do is deploy technical support people in the middle of a battle.
The Army is currently testing Jini, with plans to deploy its first "digital division" in September 2000. The decision to go with Jini was based on the fact that it came out first and that it works with the Army's existing workstations.
With the stoicism of a long-suffering technophile, Usechak says, "[Jini and Chai] don't work together. It's going to take a little time for these technologies to sort themselves out." But eventually, he says, they are going to have to interoperate.
Former Senior Editor Howard Baldwin is now a features editor at Upside. He can be reached at hbaldwin@upside.com.
YOU CAN TOUCH THIS Using software developed at MIT and hardware created by Suzuki, virtual reality becomes even more real Today's virtual reality (VR) systems can stimulate the senses of vision, hearing and motion but still leave users feeling untouched.
This may soon change. Researchers at MIT's Rapid Autonomous Machining Laboratory in Cambridge, Mass., have developed software that allows VR users to touch virtual objects with full three-dimensional realism. The advance has the potential to bring VR one step closer to fully representing all the intricate aspects of the physical world.
The MIT software, when combined with "haptic" hardware developed by engineers at Japan's Suzuki Motor Corp., creates a system that could someday let doctors touch replicas of body tissue or allow engineers to feel spatial relationships among parts in a virtual vehicle assembly. Some computer games already include simple "force feedback" haptic systems that simulate physical touch through special joysticks, steering wheels, yokes and other control devices. The MIT-Suzuki project aims to take the technology to the next level.
Sanjay Sarma, an associate professor at MIT and leader of the 3-D haptics project, explains that current haptic interfaces limit VR touch presence to just a single point, allowing users to experience little more than a VR poke.
The new technology extends the capabilities of haptic interfaces by simulating the interaction between three-dimensional bodies-a user's hand and a teacup, or a foot and a football, for example. "Approximating a haptic interface as a point is computationally easier, but the full geometry of the tool is ignored and much of the realism is lost with it," says Sarma. "Our methodology enables full three-dimensional modeling in haptic environments."
Not surprisingly, haptic systems require plenty of processing power. MIT's software was designed to make the best use of the fastest processors available.
"Computer graphics have become more sophisticated thanks to advancements in rendering techniques and computer processing power," says Sarma. "In a similar way, haptic development will benefit from greater processing power as well as advancements in software technology."
The first real-world application of the new technology is a hand-and-arm-operated haptic system-with motors and gears-that lets users create machining tool paths for rapid prototyping. The user manipulates a cutting tool in virtual space to teach the machine how to cut a specified object. Sarma expects even more sophisticated haptic systems to become widely available over the next several years to handle a range of industrial, medical and entertainment applications. "As we crack the speed barriers in haptic rendering, I expect more applications to crop up, more sales to emerge and lower prices as a result." -John Edwards