Foundations of HCI
Foundations of HCI
Before graphical interfaces, computers required users to memorise commands, syntax, and cryptic abbreviations. A wrong keystroke crashed programs. Interfaces were designed for the machine's convenience, not the human's. Human-Computer Interaction emerged as a discipline to reverse that — to make the computer adapt to the human, not the other way around.
Why HCI Matters
Bad interfaces cost real money. Studies consistently find that interface problems cause 50–80% of software project cost overruns — not through bad code, but through rework driven by unusable systems. Poor usability drives abandonment: a user who cannot find what they need in three interactions leaves and does not come back.
The inverse is also true. Amazon's 1-Click purchase removed one step from checkout and generated an estimated $2.4 billion in additional annual revenue. Google's search box — one field, two buttons — became the most used interface in history by removing every possible friction point.
The History That Shaped Modern Interfaces
HCI as a formal discipline began at Xerox PARC in the 1970s, where researchers combined computer science with cognitive psychology to invent the graphical user interface, the desktop metaphor, and the mouse. Their Alto computer (1973) introduced windows, icons, menus, and a pointer — the WIMP paradigm that every desktop interface since has inherited.
Apple commercialised PARC's research with the Macintosh (1984). Microsoft followed with Windows. The web introduced hypertext navigation and forms. Touchscreens removed the pointer entirely, replacing it with gesture-based interaction. Voice interfaces (Siri, Alexa) removed the screen. Each transition required reconsidering every design assumption from the prior paradigm.
The Three Pillars of HCI
1. Understanding Users
HCI starts with the recognition that users are not rational agents who read manuals. They are busy, distracted, operating under cognitive load, with varying mental models of how things work. A core HCI finding: users do not behave the way designers expect. This makes empirical research — watching real users use real interfaces — non-negotiable.
2. Design
Good interface design is not decoration. It is the application of principles about perception, cognition, and behaviour to make systems that support human goals. Visual hierarchy, feedback timing, error recovery, and workflow organisation are design decisions with measurable usability consequences.
3. Evaluation
Claims about interface quality require evidence. HCI developed a toolkit of evaluation methods: usability testing (watching users attempt tasks), heuristic evaluation (expert inspection against principles), A/B testing (comparing design variants on real users), and analytics (measuring actual behaviour at scale).
The Shift from Usability to User Experience
Early HCI focused narrowly on usability: can users accomplish tasks without errors? The field expanded in the 1990s and 2000s to encompass User Experience — the full perceptual, cognitive, and emotional response to a product. Don Norman's The Design of Everyday Things (1988, revised 2013) is the canonical text bridging the two. Norman introduced concepts that now define the discipline: affordances, feedback, constraints, mapping, and the gulf of execution and evaluation.
The distinction matters practically. A product can be technically usable but feel frustrating, cold, or confusing. A product can be delightful to use despite minor inefficiencies. Both dimensions matter — efficiency for productivity tools, emotional resonance for consumer apps.
HCI in a Practitioner's Toolkit
Whether you are building software, directing a design team, evaluating a product, or writing briefs — HCI gives you:
- A vocabulary for articulating interface problems precisely ("affordance mismatch" rather than "it looks wrong")
- Evaluation criteria to assess interfaces before they go to users
- Research methods to understand what users actually need
- Design principles that translate cognitive science into practical guidance
Understanding HCI doesn't require becoming a designer. It requires knowing enough to ask the right questions, identify real problems, and direct improvement with evidence rather than opinion.