To mark World Usability Day today (10 November 2011) Prof Ann Blandford's Lunch Time Lecture (given on 3 November 2011 at UCL) focused on the ways in which design can help or hinder error.

We live in a world where people make errors and most people will make more of them when they're in a pressured environment. What can designers do to take account of this and make their technology more error-proof?

An everyday example is that used by cash machine designers. When we visit these machines our main goal is usually to 'get cash' but in order to do this we need to follow a sequence of events first.

Card in - enter the PIN - select an amount of cash...

Cash machines used to give people their cash first and then return their card, but not unsurprisingly this led to plenty of people walking off with their cash (their main goal of getting cash having been successfully completed) but leaving their card behind.

A simple swap of events in the sequence: card first, then cash - has more or less solved that problem.

Errors in a medical setting can have much more serious consequences though and Prof Blandford began the lecture with the real-life scenario of Jamie Merrett whose nurse mistakenly switched off his ventilator machine and struggled to switch it back on again. Although the nurse and nursing agency were blamed for this error there seems to have been little consideration of the role of the design of the life support technology.

One way of protecting against accidentally switching off critical machinery is to use a  type of 'interlock' that prevents the user from stopping a device that is in use. This can be physical (eg having to lift a flap to access a switch) or the machine might ask the user to confirm that they did mean to switch it off - most computers ask you if you really want to delete that file.

Was the nurse solely at fault here, or could improved ventilator design have protected Jamie Merrett from 'her' error?

"Remember" by Flickr user The Doctr

Errors aren't going to go away but the way that systems are designed can change the outcome of an error - for example making errors easier to spot or helping people recover from them.

Further reading

• When technology design provokes error - UCL blog post by Nick Dawe
• Resilient medical systems - CHI+MED's research into the ways in which people and systems can be protected against errors.
• Errordiary - a diary of daily failures: a project offshoot from CHI+MED which collects examples of everyday errors - add yours on Twitter with the hashtag #errordiary