Do you remember the days when only experts ever touched a computer interface? Mature IT people speak fondly of a period when you could learn the whims of a machine and not worry whether the people whose work you were feeding through it had any idea of its operation. Then it got complicated - and it is about to get worse.
People have always gone ahead and done their own thing. Usability is the business of making a system that non-experts can work with, acknowledging all the contexts in which it will be used. It is supposed to reduce the tendency for every pocket of staff to work around the system in a different way.
There are two tiers to this in a work environment: making the technology easily supportable by IT staff, and making it functional for end-users. Products pose interface challenges to both groups when the model used in their development takes no account of the contexts of use.
This may result from a lack of engagement with users during the development process, or just with the particular context of use of concern to the commissioning team. Or it may be a factor of passing time.
Users quickly become expert in the whims of a machine. But not with any enthusiasm: just as a difficult colleague is not so much indulged as accommodated, the system gets squeezed willy-nilly into the tasks to be completed. The result is usually a series of ingenious work-arounds, which IT staff have to unpick to support. By contrast, if a system can be modified locally, it can grow with the needs of the workforce and work-arounds remain an integrated and predictable part.
This is the current state of play, but greater challenges are upon us. Wireless networks and pervasive systems are taking usability problems to new depths, while providing the opportunity for everyone to run their own digital kingdom.
There is no way, in the world of networks, that remote developers will be able to predict use either at IT management or end-user levels.
If workable systems cannot be developed remotely, then the trend towards the development of adaptive designs capable of deep and appropriate customisation will have to speed up.
The new usability will be flexibility: not systems geared up to do what the organisation does, but systems that resemble factories for constructing locally the system that meets evolving needs.
Customisation will be in the hands of middle managers and end-users, challenging both groups to learn more about each others' requirements.
However, with the increasing penetration of networked devices into domestic spaces, and the merging of telephony and computing, the distinction between work and non-work is diminishing and new flows of digital information are filling the workspace, belonging to no dedicated domain.
"Work" systems connect to a wave of non-work-related data, linking staff to other people and places. Nonetheless, much of this wave loosely supports the professional dimension where output is created, stored, distributed and communicated about.
So how do you keep things running smoothly as the distinction between domestic and work systems start to break down?
As networks move from the preserve of the workplace, how will all these systems dovetail to yield an organisation that produces what it should? What are the employer's responsibilities for connection and support when work becomes only one of the contexts of a person's connectivity, but one which cannot be separated from other network functions?
Increasingly, end-users will be configuring systems of their own and attempting to integrate work in a way that remains meaningful both domestically and professionally. In effect, they will become their own system administrators.
How much responsibility is it appropriate for organisations to take for supporting users in learning to design, as well as use, their local digital environment?
Decisions on this will vary. As now, some organisations will be more tolerant of local pockets, while others will continue to fight to maintain a consistent IT policy, as work-arounds move from the exception to the norm. Either way, flexibility will be the essential component of new systems to allow choices about both function and control.
What do you think?
How much power should your end-users have? Tell us in an e-mail >> CW360.com reserves the right to edit and publish answers on the Web site. Please state if your answer is not for publication.
Ann Light is editor of UsabilityNews and a consultant and researcher in networked media.