In the first years following the millennium, the future appeared bright indeed for mobile technologies. As the sun started to set on e-commerce, venture capitalists and investors turned to what they saw as the wireless wave, driven by the newly established infrastructure for the General Packet Radio Service (GPRS) (which essentially extended the Internet onto mobile devices), the exponential growth of Simple Messaging Service (SMS), and the Europe-wide frenzy over the "broadband" wireless protocols such as 3G. Many companies started promoting applications that exploited the Wireless Application Protocol (WAP) and its multidecked presentation format, Wireless Markup Language (WML). Sun's Bill Joy talked about swarms of devices interacting using Java and Jini, and location-sensitive services looked set to take center stage with the promise of using cell phones as credit cards, teller machines, and points of sale combined.

How very exciting it was all going to be - but here we are just a few years later, still wondering what it was all about. Not ironically, it was the overspending in the telecommunications industry that some say catalysed the technology downturn, and nearly sparked a global recession. Media focus turned to the fact nobody was buying IT at all, never mind extending it into the wireless ether. The industry was thrown into turmoil, and there have been numerous casualties - most IT companies are still trading at a fraction of their peak value.

As with any crisis, it has not all been bad news. Not least, the downturn provided an opportunity for many bleeding-edge technologies to catch up and stabilize a little. Enter mobile technologies - with several years of development behind them, and a long way from their release 1.0 incarnations. We're now looking at Windows Mobile 5.0 from Microsoft, running on devices with high-resolution screens, and Bluetooth and WiFi issued as an integrated standard and supported without a fight over drivers. Things aren't perfect yet - but like many things in the computer and communications industries, the consensus is that mobile technologies are now good enough for mainstream applications. As we reenter the mobile mainstream, we want to avoid the mistakes of the past. What can we learn from the current uptake of such technologies, to drive how we develop and deploy them?

It is with a sense of a little more humility that vendors are approaching the mobile space this time around. Deployments tend to be led by the mobile operators, who are keen to act as supplier, service provider, and application manager for the enterprise. Initial projects tend to be more about pilots than full rollouts, with everyone involved being keen to try out the gadgets; however, the move from pilot to full rollout is often limited, and such technologies are being used unofficially more often than not.

Results of a recent survey from Quocirca show that there is a major upsurge in the use of wireless devices in the enterprise, for access to corporate systems and other uses. The survey group comprised 240 interviews from a broad set of industries across Europe, with those responsible for managing their organization's relationship with mobile operators. Over 70 percent of companies were active or starting to be active in this calendar year, and while the majority of activity was from adding data cards to laptops, there is a growing use of mobile devices such as the RIM Blackberry or handhelds running Windows Mobile. There was significantly less corporate adoption of smartphones phones, which should not be unexpected as many smartphone form factors are designed to appeal more to a consumer audience than to business.

When moving from pilot to full rollout, companies are faced with a number of obstacles based around the impact on the business infrastructure and impact on the individual. It should come as no surprise therefore that one of the main activity areas for mobile rollouts is in mobile e-mail. Indeed, we have noticed what we term "the Blackberry effect" in which roughly double the numbers of Blackberry-toting respondents saw mobile e-mail as likely to be a standard element in the future, compared to users of other devices. We can learn a number of lessons from this, both for e-mail and other applications. In the old days, back at the dawn of PDAs, Palm Computing founder Jeff Hawkins coined the phrase "the Zen of Palm" which referred to the simplicity and effectiveness of the device. A major reason for the failure of WAP was cited as being its lack of usability, further emphasizing the importance of keeping things simple.

A second finding was how mobile e-mail was seen as a catalyst for other applications. Again looking at the Blackberry findings, almost 54 percent of BlackBerry users felt it is a suitable platform for a broader wireless application strategy, but nonusers felt it is more limited. Extrapolating from this, it seems sensible to take any company through a "mobile e-mail" evolutionary phase, if for no other reason than to help manage the change process. Individuals have a chance to get used to mobility principles through using mobile e-mail, and then they are more likely to welcome other applications as they are developed. In addition, some of the infrastructure and operational issues can be ironed out, for example how devices are provisioned, secured, connected, and paid for. Incidentally, Blackberry is not the only platform under consideration - indeed it is neck and neck with a Microsoft-only solution.

Mobile email gives us a good starting point for understanding how other application types can be developed and deployed. We've already mentioned the aspect of simplicity, but there is also the fact that e-mail offers a "hard stop" - in general, the sending or receipt of an e-mail coincides with a hand-off between two actors in a business process. In addition, we have found that mobile applications work best when only those elements of the process that need to be mobile are actually mobilized - this differs from the idea (espoused by a number of vendors in the past) that the creation of a mobile application need involve only the creation of a mobile front end onto an existing app. The best mobile applications are those that extend the application environment and make their users' lives easier, rather than those that try to do everything and create a new set of problems (for example, in data synchronization) in the process.

We've termed the early adopters of mobile technologies "pink collar" - a reference to the fact that many stockbrokers and attorneys sport pink-collared shirts in the city of London. These are the early adopters, and the current applications tend to be generic. However we expect to see the growth of specialized applications for white- and blue-collar workers with the next wave of mobile application development. While there have been a number of blue collar mobile applications deployed over the years, these have often used specialized hardware and heavily bespoken and specialized application development. The upswing in increasingly standards-based mobile platforms, and the availability of more off-the-shelf devices that are capable of supporting more sophisticated applications, broaden the opportunity for building cost-effective applications that really do service a business need. One such example is the EZtouch point of sale system, which exploits standard Pocket PC PDAs and WiFi for order creation by waiters and customer service staff.

By keeping things simple, fitting in with the business processes of their users, and recognizing that the adoption of mobile technology is a change cycle just like any other, the development and adoption of mobile applications stands the highest chance of success.

• Figure 1
• Figure 2
• Figure 3
• Figure 4