Posted by Recently, I have been considering ways to reorganise my course on human-computer interaction (HCI) over on Udemy. I love the course and I add to it as and when I think oooh this is a good idea. Consequently it has, gotten a bit theory heavy and a few bad ratings. So, I am looking to reorganise the lectures to impart knowledge in a lighter way and perhaps drop a lecture or two as it comes in now at a hefty six hours and nine minutes.
Along with this, I also need to manage potential students’ expectations better, as often, when people talk about HCI, they actually just mean user experience (UX) with its relationship to business/digital marketing and persuading people to buy your stuff, so students turn up expecting sexy marketing hype and I give them boring HCI, hence the bad ratings. Even psychologist, Don Norman, who came up with term UX in 1993 as a way of extending user-centred design (UCD) says that the term UX has been horribly misused. For he was really talking about how people feel when they buy something cool like a new phone or telly and bring it home and get excited just by purchasing it even before they get it out of the box.
I got a new phone recently and it’s great, it’s whizzy and all of that, and yet it is like a millstone round my neck. I have to have one these days, society doesn’t let you function well without one. And, as I have said before, I am the PA for my family, a role for which I definitely didn’t sign up which means that I get endless, unwanted, meaningless messages from schools, hospitals, clubs, travel cards, lunch passes, bank accounts, and all the other areas of my young adults/children’s lives. And as they are automated, I get them 24/7 reminding me that it is my responsibility for everyone in my family to keep all these things paid up as well as making sure that they are hitting those ridiculous government targets from attendance to performance, as everything is simplified to the point of nonsense e.g., if you attend school 100% you will get 7 and above at gcses, which conveniently forgets about the ability of student/teacher/other factors. The data point is the goal, who cares about the human?
HCI is way bigger than UX and UCD. Of course, we want our design centred around the humans who are using it. We also want to create software which works well and thus gives users a good experience. We do this by following a lot of the boring guidelines because despite the current UX hype telling you otherwise, you cannot design an experience.
It is simply not possible. No one has the same experience even in the same circumstances as humans bring themselves to the software and all of their experiences regardless of how it is designed, which is why we spend so much talking to others about our understanding of a situation.
There is no such thing as a raw experience
We all see things differently and that’s what makes us human. We never have a raw experience, all of our experiences are filtered and coloured by our past experiences which is why when we turn up looking at some software, we have certain expectations, whether they are helpful or not.
Jakob’s Law states that the more similar your new app/software design is to the existing apps and software that your user group already uses, the more successful your design is likely to be. Unless of course, they hate the apps/software they use, which is why it is so important to talk to your users and get their feedback. And there are many UX and UCD techniques, which have been around a very long time called different things by different domains such as knowledge elicitation when building AI systems and user requirements in software engineering.
However, users will tell you that they want one thing when really, they mean something else. Our job is to listen really carefully to what people are saying whilst observing what people do as more often than not the two do not match up at all.
However, it has been proven over and over that a well-designed user interface, often a graphical one (GUI), can save money which has led to the shorthand belief that good UX will increase your sales and turnover, help you persuade everyone to buy your app and make you famous/sexy/happy or whatever the goal is.
One recent example I have been using in my talk about how the AI at the McDonalds drive-thru worked and why it went ‘rogue’ is that, one of the reasons McDonald’s wanted to introduce technology in the first place is that they found that customers using the app or kiosk would order more food and larger portion sizes when they didn’t have to interact with another human whom they felt might be judging them.
Surprising behaviour around software
The human response and subsequent unanticipated behaviour, for me, is way more interesting than trying to design things to persuade humans to behave differently and to close the deal or whatever the phrase is. I only want to design something if people want it. I don’t want to be forcing people to use things and change the way they work because of what a stakeholder thinks or that it will save money in the short-term. As experience shows long-term benefits will not be forthcoming if users just don’t use what you have created.
I want to support and empower users, not make them miserable. Though, the McDonald’s example I’ve used above is morally ambiguous. Technically speaking, it is serendipitous design but has been adopted as a driver for profit which in turn I guess, proves my point, humans continually surprise us with that behaviour, good and bad.
Whenever I teach HCI, I begin by quoting the definition given by the Association for Computing Machinery Special Interest Group Computer-Human Interaction in 1992 (SIGCHI, ACM). This is because even though much has changed in the world of technology since then, this definition holds up really well:
Human-computer interaction is the design, evaluation, and implementation of interactive computing systems for human use and with the study of major phenomena surrounding them.
SIGCHI ACM (1992).
Whilst I have repeated the above many times and got cracking with designing, developing and evaluating, just lately I have been thinking a lot about the phrase: the study of major phenomena surrounding them.
In part, it is people behaving differently when interacting with software to the way they do with people, and the drive for nice interfaces because attractive things working better, which in turn leads to people being loyal to software. It is also a lot about what we bring as humans to the user experience, as I said above: our expectations, our feelings, our knowledge, and even something as simple as the day we are already having.
Exploring major phenomena
Recently, I have had a couple of experiences both as a consultant and as a user which have really brought into focus this major phenomena. Phenomena means experience: The experiences we have when we are either designing or using software to achieve an end goal. And, regardless of whether you are interacting directly with an app or kiosk like McDonald’s, a human-machine interface in charge of a caustic soda plant, a computer with a command line, or a computer supported collaborative work system (CSCW) one of those socio technological systems, HCI so loves to talk about, whatever it is you are interacting with: effective communication is key.
We need to make sure that the dialogue we are having is a proper dialogue and that it is reflected in the design itself between human and computer, or many humans using one system, and even many humans and many systems.
And, so I am starting a short (well we will see) blog series here to explore further what do I actually mean by this. What sort of interaction do we have between a human and an interactive system? What is the major phenomena? If there are many humans and one system in computer supported cooperative work land (CSCW), what happens here? And, also the impact AI is having on such a dialogue.
I can’t wait to see what I find out.
My course on HCI is available on Udemy: https://www.udemy.com/course/human-computer-interactions
