December 22, 2008Doug Hadden
A new software version, and more features. Our computer screens are buried in the rubble of features. New features, complex features. Features you will never use, features you will never know how to use. Making the software harder to use.
Users are overwhelmed with ‘featureitis’. Software that once was easy to use has become difficult. . Software once easy to use now boggles the mind, and those few features that might have been useful are drowned in the din of “feature noise.”
There was a time in financial software when new features were valuable and welcomed. But fast forward to today, and users are understandably reluctant to upgrade to new versions—after all, no one appreciates unnecessary complexity. Alan Cooper, in his book, The Inmates are Running the Asylum, explains why software engineers are keen to include so many features in new software releases. Clayton Christensen in Seeing What’s Next explains how the addition of these new features leads to diminishing returns. And, because software is often designed to accommodate many industries and many types of users with many objectives, software is consequently packed with many burdensome and unnecessary capabilities.
Usability and TLAs
Technology companies tried to overcome this dilemma by introducing the concept of ‘usability’, but in doing so they ended up making software that was ‘usable’ rather than ‘easy to use’. Vendor usability labs test various methods to mitigate the feature noise—employing ‘wizards’ to assist in a particular task or by hiding functions that are not often used. But the software remains complex and presents a steep learning curve.
Meanwhile, the industry has become enamoured with the use of Three Letter Acronyms (TLAs) to complicate software. Vendors talk about SOA, CRM, SLA, PPM and CPM. Vendors do not talk about capacity building and sustainability—they all talk about TCO – “Total Cost of Ownership.”
Tools and Applications
The window-style user interface of Microsoft, Apple and UNIX vendors is the dominant metaphor for software today. These interfaces, while effective for software tools, are less effective for software applications. A software tool (like a spreadsheet application) can achieve many tasks. The user requires functions to be arranged in logical sets: ‘file’ functions are located in one place, ‘formatting’ in another. The software engineer, who cannot accurately predict how the tool will be used, is thus tasked with providing a reasonable and logical functional design.
A software application like IFMIS, however, is considerably different. It is not a general tool. There is an established workflow, and user objectives are predictable: requesting a requisition, approving a purchase order, receiving goods, performing bank reconciliations, and the like. A design focused on a logical presentation of functions can in this case be confusing. The user is performing a task and the software should understand this, just like good airline reservations software which communicates the process to the user without presenting the visual noise of unnecessary embellishments.
Software should be designed to meet user goals and should be simple to use for the task at hand. Perhaps new software should take a cue from the Philips Corporation, which is attempting to make electronics simple, or from Apple, which greatly contributed to the simplicity of a new generation of portable digital multimedia players.
We at FreeBalance believe it’s time to bring the ‘simple’ back to software. It’s straightforward: software that is easy to use is also easy to sustain. Software that is simple can be useful for capacity building, and software that understands the users’ goals can implement much faster, enabling governments to more rapidly accrue benefits and achieve its goals of modernization.