Purpose
During today, like most computer conferences, we’re going to see and hear about all sorts of wonderful advances, clever tools, and stories of how IT is the best thing since the invention of fire.
Without wanting to put a damper on the enthusiasm, I want to explain why I think we still have a long way to go. Computers and related systems remain immature technologies that fail more than they succeed.
What do I mean?
First, remember why it is that we use computers. Some of us just love to play with gadgets. Some value the creative reward of problem solving. Some see IT as a means to make money. But the underlying intention that enables all these others is not to develop technology for its own sake, but to create tools that support human goals.
People want better ways to communicate, to exchange information, to find things, to compose and perform music, to impress other people, to write newsletters for their club, to entertain and be entertained, to learn. And people look for tools that enable them to achieve those goals.
But as a tool for achieving those goals, IT sux.
1. Not easy to use
2. Not useful
3. No “graceful degradation”
¨ Fine if it works, but horrible when it doesn’t
¨ That’s almost a fundamental flaw in the digital computational model. The boolean thinking imposed by discrete state machines leads to software in which an action either succeeds or fails.
4. Forces us to adjust our behaviour to suit the limitations/structure of the technology. The technology distracts us and interferes rather than enabling us to get on with the things we really want to do.
5. High failure rates of IT projects
¨ 19% abandoned before completion – that’s basically 1 in 5!
¨ From the Standish Group International’s Chaos Report, quoted at http://www.cio.com/article/124309
6. Technophobia
¨ A splurge of research in late 90’s and early 2000’s (notably Larry Rosen and Michelle Weil) but apparently very little in the last few years.
In summary, a lot of IT fails outright; even when it works it is often hard to use; and even at it’s best, it makes many people anxious and stressed.
What causes it?
1. Too many features
¨ But that’s about all non-experts can use to compare competing products (e.g. article on consumer electronics / white goods)
¨ Lemon market, where consumers cannot judge quality and must rely on something like salesrep’s recommendation, number of features or price.
2. Incompatibility
¨ e.g. my problems setting up Win XP x64: printer, sound card, anti-virus all stopped working
¨ e.g. Mum’s problem getting a printer for an old iMac
3. Changes too frequently –> prevents mastery
4. Prevalence of poor UI
¨ Programming the VCR
5. Immature technology in a state of rapid discovery, with poorly understood engineering principles.
¨ Imagine if 1 out of 5 buildings failed to be completed!
¨ I don’t say that as a deep criticism. IT is exciting and wonderful. But let’s not kid ourselves that it is a well-advanced discipline. We are nowhere near understanding its potential at this stage, let alone achieving that potential.
What can be done?
1. Vast improvements recently
¨ The 19% failure rate is an improvement on the 1994 rate of 31%
¨ Ubiquity of internet connectivity — but still confusion about how to act when disconnected (e.g. GMail, but c.f. Google Gears)
¨ Convergence of computer, mobile, phone, camera, calculator, calendar, watches etc
¨ Simplicity of Google
2. Primary principle: user-centric technology
¨ Not just UI design, but a human focus from conception to post-implementation support
¨ Think about how technology affects human behaviour
* From the physical and psychological aspects of how a single person interacts with technology
* To organisational structures and processes
* And even considering the interactions between social change and technological change
3. Just one example
¨ Build the user interface to match the user’s existing mental model
¨ Affordances (James Gibson and later Donald Norman)
* Toilet door handles
* Five easiest point to navigate to on a screen. Why aren’t they used?
* Another negative example is given by Joel Spolsky (Chapter 4 of User Interface Design for Programmers), who points outthat MS Word toolbars have several icons that include a magnifying glass (Print Preview, Document Map, Research), none of which makes the view of the screen any bigger. To make it bigger, you choose from an unlabelled drop-down list of percentages. Even when that control is pointed out to people, I’ve seen them confused about whether it increases the font size in the saved document or just the current on-screen magnification.
4. How will we know when IT succeeds?
¨ Successful technology disappears (was that also Norman?)