I would like to tell you a little bit about the subject of HCI.  Many people either have never heard of it, or if they have, think it is about making graphical user interfaces (GUIs).  So I would like to use the space here to tell you a little bit more about it! HCI is a very multidisciplinary subject.  Worldwide it is taught mostly in University Computer Science Departments and Psychology Departments.  There are a few faculties which are based around HCI, teaching as well subjects like cognitive science and ergonomics, but on the whole it is mostly known as a post graduate subject, or an ‘add on extra’.  Yet it is interesting to note that the first course of HCI was actually taught in the Design department of Carnegie Mellon University in the U.S., the feeling being that it was a subject that should be taught to designers.

HCI concerns itself with questions of usability, accessibility, and the design of products and systems so that they are fit for purpose.  You probably don’t need telling that there are hundreds of examples of unusable computer designs around.  It is important to realise that these are not just at the interface of the human with the device, be it a computer screen or a remote control for a television or an air-conditioner.  Everybody has their favourite story to relate of how they could not tape a programme using the video recorder or manage to switch off the burglar alarm.  The ‘I’ in HCI refers to interaction, and this points to the fact that very often the problem lies not with the interface but with the underlying information system.

A simple example of poor system design was in the first ATMs (Automatic teller machines) outside banks.  People used very often to forget to retrieve their cards, because their main task, of getting or depositing cash, had been achieved.  This became such a nuisance for the banks and customers that they changed the system, now the card pops out before the money, and in some machines a beeper sounds continuously until the card is removed. The problem was pinpointed with task analysis, which is a standard part of systems design, but it seems the designers of the early ATMs had not noted this part of the task because when the task was performed with a human teller, the card/ account evidence was passed back and forth between the teller and the customer in no specific sequence.  The teller just made sure he did not keep it. Thus the ATM made sure it popped the card back to the customer, it didn’t appear to matter that this took place at the end of the transaction!

The early phases of HCI as a discipline corresponded to that changeover from human-human interaction to human computer interaction - to the time when ‘computerisation’ meant automation of certain tasks, and computer use was often studied on the basis of one user /one stand alone PC. This meant that often users were categorized into ‘novice’ and ‘expert’, the latter being those with some computer literacy. 

This tendency to regard users as being of different types, was in itself a follow on from work that began in the sixties and seventies concerned with control systems.  These were often military systems, such as SAGE (Strategic Air Ground Environment) but also civic systems, such as air traffic control.  In control systems, it was always assumed that operators would receive extensive training in the use of the systems. However, the early HCI psychologists, known as Human Factors specialists (the term HCI was not yet in use) could clearly predict that however much training operators received some mistakes would be made again and again, because systems were just not designed to give logical feedback.  A case in point was in a project I worked on in the early nineties.  The task was to evaluate the part of a new air traffic control system to plan and sequence incoming flights that allowed air traffic controllers to have a visual list of all incoming flights.  With this they could see rearrangement of flights if, for instance, one plane was low on fuel and needed to land before others in the ‘stack’.  One of the problems we discovered was that flights that landed would just ‘drop’ off the screen and the space on the list would close up. This was alright if flights were landing in sequence on the screen, but if they were not, it was visually very hard to keep track of.  Once again, the problem was one of the system design, and not just the interface.

HCI is evolving, just as our world is changing, with computers embedded in hundreds of devices, in everyday use, in our homes and workplaces, (which very often nowadays, occupy, part of the time at least, the same physical location!).  With computers all around us, we are all users, all with some knowledge, but few with enough knowledge, or time, to really understand and programme complex machines.  We expect to use these machines without prior training, and without making mistakes. Part of the problem may be that we really don’t need twenty different programmes on the washing machine- that is ‘overkill” on the part of the designer to distinguish one machine from another and create a ‘top of the range’ cult.  However, part of the problem is that sometimes that the designer wants to ‘cash in’ on the familiarity of users with a certain device, and so enriched functionality has been grafted onto a basic design that was never intended to take such ‘upgrading.’  My last but one mobile phone was such an example.  It could do everything but make tea!  My problem was that when I was in a hurry, I could not tell if I was in ‘phone mode’, or something else.  I changed my phone (back to Nokia!) and to a simpler functionality, but one that was perfectly adequate for most of my needs.

The power of device familiarity is strong.  Sometimes we do not realise how we ‘port’ our knowledge of one device to another.  I recall with fondness a colleague who phoned up complaining that her mouse wouldn’t work. She was a new computer user, but had absorbed a lot of knowledge by watching and asking colleagues. At length, we discovered that the mouse wasn’t working because she was holding it in the air, and pointing it at the screen, as though it was a remote control! 

Let’s just stay with that familiar gadget, the TV remote control.  (If there is one household that has never been in that position of having temporarily lost the remote control!)

Now are used to zapping through 40+ channels to see what is happening, but as more channels are available, this will not be at all an efficient way of checking out the night’s viewing!  And while we are on the subject of TV - what about Interactive TV?  Its proponents point to the fact that TV is a well-known device, familiar to users and found in most homes in the western world.  They want to offer all the e commerce services, banking, shopping, distance education, chatting now available on personal computers via this device.  Will the handheld remote we know suffice, or will we need something the size of a computer keyboard?  And what about this, such handheld devices are difficult or impossible to use for some of the elderly and disabled, - precisely those groups whose problems make them the ideal candidates to benefit from using the banking, shopping etc services from their home.  Thus it is possible to see that unless some serious HCI work is undertaken, there will be a lot of mistakes repeated. 

Now that our world is network based, with computers with connecting us to our colleagues and friends and, the possibility to connect devices with devices, our dependence upon buttons, switches, and displays is unquestioned.  The information systems design that lies behind the interfaces will be crucial to the success of the product and/or service. 

It is true that, as always, humans will learn and adapt, some more quickly than others.  If products and systems are hard to use, they will have to be very essential to users for them to make the effort to learn to use them, if they are difficult to use, or easy to make mistakes, it could be at best annoying, at worst dangerous.  This is as true now as it was half a century ago.  What has changed is that now the population of users is much greater, and more varied, both in their needs and their abilities.  They also hold computers in less awe, and as market reports show, look for usability over functionality. 

HCI offers a range of techniques, methods and methodologies for helping capture requirements, translating those requirements into technical specifications, for prototyping and for evaluating and validating products and systems. It is applicable to already existing systems and products, as well as to conceptual designs. It takes input from psychologists, cognitive scientists, designers, computer scientists and engineers, and of course, users.  The common theme that runs throughout is that it should be the technology that needs to adapt, -or die, not humans!