HIF / Documentation / Foundation

HIF Philosophy

1. The subject of design

An interface is not a surface between a fully formed person and a fully formed machine. It is part of a wider system in which a person perceives state, forms an intention, acts, receives a response, corrects errors and interacts with other people.

The subject of HIF is therefore not the screen, but a system of relationships:

person
↕
task ↔ representation ↔ rules ↔ data
↕
social and physical context

Visual quality is necessary, but it is only one property of this system.

2. Human model

HIF proceeds from the following propositions.

2.1. People are constrained and adaptive

Attention, working memory, motor precision and time are finite. At the same time, people learn, develop strategies, use external representations and adapt their environments.

An interface should reduce unnecessary demands without depriving people of useful complexity, control or professional skill.

2.2. Action is situated

A plan does not determine every future operation. People refine their intentions as they receive feedback. An interface should support exploration, comparison, stopping, correction and changes of goal.

2.3. Cognition is distributed

Knowledge is distributed across people, the spatial arrangement of objects, documents, signs, histories and devices. A good interface externalises material state and helps people coordinate activity.

2.4. People vary

There is no single “average user”. People differ in experience, language, culture, bodily and sensory capabilities, devices, context and temporary condition.

Variability is a starting condition, not an exception.

3. Ten commitments

3.1. Human agency

The system helps people achieve their goals while retaining authority over material decisions. Automation should not obscure authorship, consequences or the possibility of intervention.

3.2. Understandability

A person should have a sufficient model of what exists, what is happening, why it is happening and what can be done next.

3.3. Causal honesty

A representation should not imply a false causal relationship. A visual response should correspond to the actual acceptance and outcome of an action.

3.4. Reversibility

Exploration, error and changes of mind are normal. Routine operations should be reversible; irreversible operations should be rare and clearly distinguished.

3.5. Continuity

A change of device, theme, view or mode should not unexpectedly alter object identity, data, rights or command meaning.

3.6. Inclusion

A primary function should not depend on a single sense, input device, level of precision, language or mode of perception.

3.7. Respect for attention

Attention belongs to the person. A system should not capture it without proportionate need, or turn notifications, motion and urgency into a constant background.

3.8. Privacy and security

Security does not justify incomprehensible controls, and convenience does not justify concealed data collection. People should understand material access boundaries.

3.9. Pluralism of form

Different interface cultures are legitimate. Floating, tiling, command-line, touch-first, voice and spatial interfaces can express the same system through different grammars.

3.10. Verifiable quality

A design decision remains a hypothesis until its consequences have been examined using a suitable method.

4. Theoretical perspectives

HIF uses several compatible analytical perspectives.

4.1. Human-centred design

The context of use, intended people and tasks are studied throughout the lifecycle. Evaluation is iterative and informs requirements.

4.2. Information-processing models

Fitts, Hick–Hyman, GOMS and KLM are useful for bounded predictions of movement, choice and expert efficiency. They are not universal laws of composition.

4.3. Ecological approaches and direct manipulation

Objects and possibilities for action should be perceptible. Direct manipulation is valuable when its effects are visible, incremental, rapid and reversible.

4.4. Distributed cognition

The unit of analysis is a system of people and artefacts, not an isolated mind.

4.5. Situated action and activity theory

A task is considered within motivated activity, its social environment and actual conditions. An operation is not optimised in isolation from its goal.

4.6. Sociotechnical and ethical analysis

An interface distributes authority, visibility, risk and opportunity among participants. An apparently neutral visual form can have non-neutral consequences.

5. What is not a universal law

HIF prohibits uncritical reliance on popular simplifications.

Click count is not equivalent to difficulty.
Fewer options are not always better.
Consistency does not require visual sameness.
Familiarity does not justify inherited inaccessibility.
Minimalism does not mean concealing state.
Personalisation does not remove the need for system invariants.
Engagement is not an independent human value.
A high conversion rate does not demonstrate that a design is ethical or beneficial.
An accessibility checklist does not replace participation by disabled people.
Aesthetic authenticity does not demonstrate functional fidelity.

6. Complexity

Complexity can be:

domain complexity — inherent in the task;
system complexity — produced by the product architecture;
interface complexity — produced by the representation;
learning complexity — necessary for developing competence.

HIF requires the removal of accidental interface complexity, the explanation of system complexity, support for domain complexity and preservation of productive learning complexity.

Concealment does not eliminate complexity. It transfers complexity to memory, error or recovery.

7. Aesthetics

Aesthetics is functional when it:

establishes hierarchy;
supports grouping;
expresses state;
supports identity;
strengthens trust;
makes sustained use comfortable.

Aesthetics becomes harmful when it:

competes with the task;
imitates a function that does not exist;
uses ambiguous signs;
reduces legibility;
makes different consequences appear equivalent;
turns a person into a spectator where action is required.

8. The criterion for a good interface

A good interface enables different people to achieve meaningful goals:

effectively;
with proportionate effort;
safely;
accessibly;
with an understanding of state;
without a concealed loss of autonomy;
with opportunities to learn and develop expertise.

No single measure can substitute for this whole.