Preamble

**Figure: Tom's Diner**. "Tom's Diner" is an a cappella pop song written in 1981 by American singer-songwriter Suzanne Vega. Vega wrote the song based on a comment by her friend Brian Rose, a photographer, who mentioned that in his work, he sometimes felt as if "he saw his whole life through a pane of glass, and ... like he was the witness to a lot of things but was never really involved in them" ---Image Credit: Wikipedia. — **Figure: Tom’s Diner**. “Tom’s Diner” is an a cappella pop song written in 1981 by American singer-songwriter Suzanne Vega. Vega wrote the song based on a comment by her friend Brian Rose, a photographer, who mentioned that in his work, he sometimes felt as if “he saw his whole life through a pane of glass, and … like he was the witness to a lot of things but was never really involved in them” —Image Credit: Wikipedia.

Welcome to User Experience! For many of you, this will be your first encounter with the discipline of Human–Computer Interaction (HCI) and Human Factors. Some of you may have already experienced some basic HCI work as part of the software engineering lifecycle. However, this text takes a far more in-depth look at the tools, techniques, and knowledge you need to understand the user experience within software engineering.

This text goes deeper than introductory HCI material you may have seen before — but it’s not trying to teach you everything about human-facing software engineering. Indeed, entire undergraduate degrees are built around the subject of human factors and ergonomics, and you should not confuse this high-level overview of the domain with the knowledge you would acquire in a full three-year degree programme. That said, the aim of this text is to give you, the reader, the tools, techniques, and the mindset necessary to competently approach your first user testing and user experience job. The text is designed from a practical perspective and will enable you to take a junior role in a user experience department or usability company, and will give you the broader knowledge to communicate with colleagues and make sensible suggestions about UX work. Further, it will provide you with a basis for future self-study within the UX domain or the wider human factors world. With all of that in mind, I’ve decided to call the text, more fully, ‘User Experience from 30,000 feet’¹.

As we’ll see later, the UX domain is still very young and in the process of formation. But, it does bring together many already established areas within the human-computer interaction field. As a placeholder, you may wish to think of UX as the practical application of research knowledge repurposed from other domains into the user-facing software engineering process. Recommending a single UX text would risk skewing your education towards one view of the domain (although you may wish to check out some UX books as well: ²; ³; ⁴). I feel it is better for you to understand my view while realising there may be different ones out there. In the end, you will need to decide, after reading this text, the sort of UX you wish to do and how you think about the area. Therefore, this text has been written from scratch and represents an overview of UX as seen by the human factors and web science researchers I work with.

You don’t need to panic! The text is intended to provide you with an overview of the UX field as well as an understanding of the key concepts within this field. Plus a method of assessing your understanding of the topics covered, and finally, a reference to reading material which will enable your deeper study, both now and in your future work.

Finally, I’ve tried to keep the jargon (and explain any I introduce⁵) to a minimum. And I hope you’ll see that the writing style is ‘formally academic’ but clear and easy to read.

Origins

To give you an idea of how this text arose, it may be useful to give you an idea of its origins.

Similar to many Computer Science Departments, the School of Computer Science at Manchester has its roots in the Computer Group of the Electrical Engineering Department and the Computer Laboratory of the Department of Mathematics. After its formation in 1964, it continued to focus on computational hardware and low–level or embedded software. This focus resulted in the world’s first stored–program computer; the Manchester Small–Scale Experimental Machine (SSEM), nicknamed ‘Baby’. The School, run in the early days by such luminaries as Frederic C. Williams, Tom Kilburn, Geoff Tootill, and Alan Turing, continued to focus on the numerate, applied, and engineering aspects of Computer Science.

In this environment there was little focus on the human aspects of computer science — aspects that would inevitably grow in importance as the computer moved from the laboratory to the desktop and into everyday products and devices. However, some of the researchers at the School became increasingly aware that we were not addressing areas which would – we thought – become increasingly important.

In light of this perception a user experience text was developed to serve as a cohesive starting point for us to be able to teach HCI and UX to technical Computer Science, Software Engineering, Hardware Engineering, and Artificial Intelligence students. It is therefore based on the human factors research activity already established within the School over the past 15 years, in support of the embryonic teaching activities arising after 2010.

Before we go any further it may also be useful to make one more point. You will notice that I use the term User Experience Specialist, usability specialist, UX’er, etc. throughout the book when referring to you the reader. This is because I’ve found it useful to reinforce your belief that this is what you can become as you learn more about the field. I think using this terminology also makes you more likely to think of yourselves in a different way to the ‘developers’ that you probably are, and the engineering domain you feel comfortable within.

Text Structure

UX is an empirical discipline, whereby established principles and guidelines are applied to software development and the correctness of that software is established via experimentation in the form of user feedback. The UX specialist is therefore mainly interested in questioning the development to ascertain its correctness.

This correctness normally exists along many dimensions. But briefly, can be thought of as the primary utility of the development; its effectiveness; its efficiency; and its emotional impact (both tangibly and intangibly, represented by the concept of engagement). The text structure is therefore intended to give you the principles, tools, and techniques needed for UX work along with a critical and questioning mindset to enable you to apply these techniques effectively.

Therefore, you will learn the tools and techniques needed for effective interaction with each dimension. These include the scientific and empirical techniques to join and apply these principles and tools to a new bespoke development. And the mindset to be critical of, and to question the outcomes of, this development to establish the user experience.

Secondary ‘Text’

“Zen and the Art of Motorcycle Maintenance (ZAMM) ⁶” may seem like a strange text to use for a Computer Science based UX course… and it probably is. I don’t expect you to memorise anything in ZAMM — but the critical and questioning skills you’ll develop from analysing it will serve you well. ZAMM is not really about Zen or indeed motorcycle maintenance, it’s about science, quality, and rhetoric; the USA’s Library of Congress describes it as:

“Acclaimed as one of the most exciting books in the history of American letters, this modern epic became an instant bestseller upon publication in 1974, transforming a generation and continuing to inspire millions. This 25th Anniversary Quill Edition features a new introduction by the author important typographical changes and a Reader’s Guide that includes discussion topics, an interview with the author, and letters and documents detailing how this extraordinary book came to be. A narration of a summer motorcycle trip undertaken by a father and his son, the book becomes a personal and philosophical odyssey into fundamental questions of how to live. The narrator’s relationship with his son leads to a powerful self-reckoning the craft of motorcycle maintenance leads to an austerely beautiful process for reconciling science, religion, and humanism. Resonant with the confusions of existence, Zen and the Art of Motorcycle Maintenance is a touching and transcendent book of life.”

While Wikipedia says:

“The book describes, in the first person, a 17-day journey on his motorcycle from Minnesota to California by the author (though he is not identified in the book) and his son Chris, joined for the first nine days by close friends John and Sylvia Sutherland. The trip is punctuated by numerous philosophical discussions, referred to as Chautauquas by the author, on topics including epistemology, ethical emotivism and the philosophy of science. Many of these discussions are tied together by the story of the narrator’s own past self, who is referred to in the third person as Phaedrus (after Plato’s dialogue). Phaedrus, a teacher of creative and technical writing at a small college, became engrossed in the question of what defines good writing, and what in general defines good, or ‘quality’. His philosophical investigations eventually drove him insane, and he was subjected to electroshock treatment which permanently changed his personality.”

ZAMM is useful to us because it serves as a framework for criticism while enabling us to apply UX principles, rational thought, and scientific argument to a well understood and well-respected text. As for my thoughts on ZAMM; well I might not agree with everything the book has to say, but I do think it is an important and interesting topic and discussion which needs to be had. At this point, you need not worry too much about the book itself – other than the fact that you should read it. All will become obvious as the text progresses. But you may like to read a longer discussion of “Zen and the Art of Motorcycle Maintenance” and its usefulness for UX Teaching in the Appendix.

AI Assistance

As I began updating these notes at the beginning of 2023, ChatGPT and a slew of other AI ‘Assistants’ were released, to general uproar from thought workers, academics, and people in the digital space alike. What will this mean for learning, what will this mean for thought? Well, it is difficult to tell without trying to integrate them into your workflow and using them properly; not just as tests (no matter how extended). It’s hard to know what to make of AI assistants, so I thought I’d ask one.

How should I think about ChatGPT?

ChatGPT is an AI-powered language model developed by OpenAI, which uses a deep learning algorithm to generate responses to user input. As an AI language model, ChatGPT is designed to understand natural language and provide helpful and informative responses to user queries.

You can think of ChatGPT as a tool to help you find answers to your questions or provide you with information on a particular topic. While ChatGPT is not a human expert and may not always provide the exact answer you are looking for, it is capable of providing useful insights and pointing you in the right direction.

It is important to remember that ChatGPT’s responses are generated by an algorithm and may not always be accurate or completely up-to-date. Additionally, because ChatGPT is an AI language model, it may not be able to provide nuanced or context-specific responses in the way that a human expert might.

Overall, ChatGPT can be a useful tool for gaining insights and information, but it is important to use its responses as a starting point and to continue your own research and investigation to ensure that you have a complete and accurate understanding of the topic at hand.

The most telling parts, after my own experience, are these: ‘it is capable of providing useful insights and pointing you in the right direction’ and ‘it may not be able to provide nuanced or context-specific responses in the way that a human expert might.’

I used AI assistance to update about 15 pages of this text. Of those, roughly five could be used more or less as generated, five needed extensive reworking, and five needed very heavy correction because the concepts and elaborations were simply wrong. Overall, I found it useful for producing a starting draft to revise, or for straightforward work that was mainly about connecting existing concepts. In general, it produces adequate but surface-level detail of concepts, does not provide context or nuanced discussion based on an understanding of the real world and so needs human oversight. However, it was very useful as a naive collaborator that provided initial writing waiting for the domain expert (me) to make subsequent ‘passes’ and edits. So if you are looking for text which is not expert but is a starting point then it is a very useful tool akin to spellcheckers or more recently intelligent grammar checkers. The most useful skill a human needs to learn when collaborating with an AI assistant is how to formulate the questions presented to it to generate the best answer.

UX and the GenAI Bubble

GenAI/LLMs and the bubble that surrounds them seem like great news for employers who are currently (circa 2026) still in a rush to convert everything to ‘AI’ so that they can cut jobs. This is particularly relevant to software engineering and computer science in general. While I (and many others) may not believe the hype, and think this is cynical marketing by companies quick to realise profits, GenAI can still be a useful tool.

This scepticism is shared by a growing number of credible voices. Cory Doctorow, in his 2026 book The Reverse Centaur’s Guide to Life After AI⁷, argues that management sees AI primarily as a way to replace workers, turning skilled employees into mere assistants to machines — what he calls “reverse centaurs.” He has described the AI boom as a bubble built on unsustainable economics, noting that AI companies are effectively selling their services at a fraction of the real cost⁸. Ed Zitron, technology critic and author of the newsletter Where’s Your Ed At, has spent years documenting the gap between the AI industry’s spending and its actual returns, calling the entire financial structure around generative AI unsustainable⁹. His reporting on OpenAI’s audited financials revealed net losses of $38.5 billion in 2025 against $13.07 billion in revenue¹⁰. Gary Marcus, cognitive scientist and Professor Emeritus at NYU, has consistently pointed out that LLMs still struggle with reasoning and reliability, and that few companies outside of chip makers have seen real profits from AI¹¹. His predictions about the limitations of AI agents and the fragility of the AI business model have been broadly borne out¹². Brian Merchant, former Los Angeles Times technology columnist and author of Blood in the Machine¹³, has been documenting through his AI Killed My Job series how real workers are being affected by AI-driven redundancies, drawing parallels with the original Luddite rebellion against factory owners who deployed machines without regard for the livelihoods they destroyed¹⁴.

Whatever their particular angle, these writers converge on a shared concern: companies are cutting staff to chase an AI narrative without fully understanding what they are losing. Research from Challenger, Gray & Christmas shows that AI became the number one stated reason for job cuts in 2026, with over 100,000 US positions attributed to it in the first half of the year alone — nearly double the total for all of 2025¹⁵. Gartner’s research reinforced this picture, finding no correlation between AI-related workforce reductions and measurable return on investment¹⁶. The outplacement firm Challenger, Gray & Christmas and SHRM have both identified “AI-washing” — overstating AI’s role in business decisions — as a growing trend, with companies using AI as a justification for cuts that are really about freeing up capital or signalling ambition to investors¹⁷. Meanwhile, Forrester’s 2026 Future of Work report found that a third of employers who cut staff in this way ended up spending more on restaffing than they originally saved, once recruiting costs, onboarding time, and higher salary expectations were factored in¹⁸.

However, UX is in a relatively good position here. Activities like user surveys, focus groups, usability studies, contextual inquiries, heuristic evaluations, card sorting, and diary studies all need human effort. They often require face-to-face interactions and a situated knowledge of user-facing issues in a specific context. This gives UX work a degree of protection against current GenAI threats, though not full immunity. AI can speed up transcription, help spot patterns across large qualitative data sets, or generate draft survey questions¹⁹ — but it cannot sit across from a user and pick up on hesitation, confusion, or the unspoken frustration that only becomes visible in person. The Nielsen Norman Group’s State of UX 2026 report put it plainly: human direction, curation, and verification remain essential for distilling insights into good products²⁰. The core of UX research — empathy, judgement, and the ability to understand not just what people do but why they do it — remains well outside what current generative AI can deliver.

Now, let’s get on with understanding and shaping the User Experience!

Up next

Preamble