When Poor Usability is Fatal

guestblogger Adam Cole (Toronto, Manager Applications and Development for McKesson Canada)

I have always been a proponent of "the user interface is the application". (That is, it doesn't matter what is happening behind the scenes, in most cases an application is judged on it's user interface.)

With this in mind I found the attached article really resonated. The article very tangibly links risk to usability (interface design). In the case of Chernobyl poor design of the control room was a root cause of the reactor's ultimate catastrophic failure.

...Alas I am sorry - one more item to add to your risk assessment checklist.


Comments (1)

  1. Graham Jones says:

    It is interesting that you should quote Chernobyl. As someone who spent over 30 years in the process industries with some interest in plant safety, human "interface" issues, when it came to accident investigation, invariably figured in the causation chain somewhere. So human "interface" issues have been around for much longer than modern software design.

    It became even more important when computer controlled procees plants became more prevelant. Prior to that it was quite common to layout the control panel using a "mimic" diagram of the plant with the instrumentation placed to provide feedback about each part of the plant.

    When computer control came along it wasn’t possible to display a mimic diagram of the whole plant on a fairly small computer screen. It had to be multiple smaller diagrams. This caused many people to lose their "feel" for the plant. Computer supervisory control was a step forward but the operator interface was a step back. This wasn’t recognized at the outset. It wasn’t that the operator couldn’t find out everything that they needed but it required them to ask. In time multiple screens were used to move back closer to the original interface. Today, modern display technology permits both benefits. The lesson, of course, is that you must not forget about the "human" interface when making technological change.


    Graham J.

Skip to main content