Mobilizing Enterprise Applications In A Fragmented World
By Faraz Syed
The mobile enterprise has experienced incredible growth in the past two years. Most IT departments are supporting mobile e-mail--and not just for executives, salespeople, and field technicians. Extending your IT systems onto the mobile platform is a necessity brought by the quickly changing demands of enterprise mobility and has the potential to change the way we do business for the better; however, it also faces an array of unique challenges that go far beyond what we previously have had to contend with.
A Forrester Research report, "Technology Populism Fuels Mobile Collaboration," finds that today, almost half of all enterprises support the use of personal mobile devices, and about 40% of employees ask their IT departments to install productivity apps. Now that smartphones have entered the enterprise, IT is struggling with the mobile information workers' pent-up demand for accessing business applications beyond e-mail, including ERP, CRM, and other often-used systems.
Unlike the PC world where the WINTEL "standard" provided a common platform, companies looking to extend IT services onto mobile platforms are often presented with a plethora of protocols, standards, operating systems, and hardware configurations. Today's mobile ecosystem is no longer fragmented but splintered. With both handset manufacturers and even mobile operators all launching their own "unique" offerings, there seems little sign of a single platform or framework that will be universally accepted.
The sky's the limit
This is why some developers are looking to a new model based on cloud computing in which applications are hosted remotely from centralized servers--a promising alternative, especially for enterprise solutions in which multi-platform support is essential.
While a good number of consumer mobile applications have tapped into the cloud, so far only a handful of enterprise mobile applications have done the same with any success. However, Juniper Research has predicted that the number of enterprise workers using mobile cloud-based applications will rise to more than 130 million by 2014.
As cloud providers are increasingly opening up their APIs to developers, an opportunity is emerging for companies to reduce the costs associated with porting apps across multiple platforms. This approach is attractive for both developers and enterprises alike: developers get a wider potential audience, whereas enterprise customers reduce costs and increase their flexibility by outsourcing application management on a scalable, pay-per-use basis. However, those thinking that this is the golden panacea to mobilizing enterprise applications are mistaken.
Golden panacea or head in the clouds?
Similar to its PC-based cousin, mobile cloud computing presents unique issues that require risk assessment in areas such as data integrity, recovery, and privacy, and an evaluation of legal issues in areas such as e-discovery, regulatory compliance, and auditing. When you consider the high-profile cases of lost data devices and laptops by government professionals, the risk for a company can be even greater than simply losing a $400 handset.
Apart from the intrinsic risks surrounding cloud computing, there are particular challenges faced when looking to offer cloud-based applications on the mobile device. For example, IT departments have faced a dizzying assortment of devices and platforms, which are challenging enough on their own. But now we're finding that even devices based on the same OS or from the same manufacturer can also present challenges.
Splintering is not limited to the diverse range of platforms available; it is even present within operating systems themselves. A recent example is the launch of the Twitter application for Android that only works with devices running Android version 2.1 or higher. This means it currently will run only on the Motorola Droid, Google Nexus One, or the new HTC Incredible and is therefore accessible to less than one-third of Android customers.
This is why it's important not just to consider fragmentation in terms of software, but also the various hardware specifications in the mobile marketplace. With a new flagship handset launched seemingly on a weekly basis, the mobile marketplace is littered with a vast array of different screen sizes, resolutions, processors, graphics cards, and input methods that need to be considered when developing a multi-platform strategy. Even the ergonomics of the phone can have a major impact on the ability of the workforce to input data on the go.
Another key challenge faced by developers looking to mobilize applications is how they can ensure quality of service coverage and capacity in a fluctuating signal environment. The mobile channel can vary significantly based on location, both from a domestic and international standpoint, and could therefore leave bandwidth-hungry applications unpractical outside the "laboratory environment."
So, if we can't get it right on smartphones, what about when we add other devices to the mix? When you consider the colliding worlds of tablets, e-readers, netbooks, and other Web-enabled devices coming to market and the introduction of Generation Y -- the "connected" generation -- this undertaking becomes increasingly complex. A single platform that spans all "Screens of Life" is the ideal. Delivering a consistent user-experience seems almost impossible.
Finding the solution
However, mobile enterprise developers shouldn't despair. They can draw on the experiences of their consumer counterparts who have been wrestling with these issues too and are arriving at practical solutions.
The number of enterprises mobilizing their applications is going to increase. The opportunities are too attractive especially as organizations are driving to improve operational efficiencies and quality of service.
So the mobile apps revolution has an enterprise dimension. And the same rules for mobile application testing also apply to all types of mobile content -- consumer-facing, business, or enterprise applications. They need to be tested.