One Size Fits None. Towards Critical Interface Design By Rosa Llop

From interfacemanifesto
Revision as of 15:39, 26 November 2015 by Andreu (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search


One Size Fits None. Hacia un diseño de interfaz crítico

By Rosa Llop

Interface design is a young, multidisciplinary discipline, based on theory from fields as diverse as psychology, ergonomics, communications, art, engineering, and mathematics. When we look for specific literature on interface design, we mainly find two types of books: declarations of intent in the form of good design principles, and books that describe patterns based on previous experiences. The weakness of both types is that they encourage an innocent approach to interface design. At a time when interfaces are a very important part of our culture, there is a need for much more mature reflection on how we want them to be designed.

The first books on design principles for interfaces were written in the eighties, when the first computers were being developed at Xerox Parc. These texts promoted a hardware-user relationship based on transparent interaction. WYSIWYG (What-You-See-Is-What-You-Get) was the mantra during those early years, and those principles laid the groundwork for the construction of an illusion of objectual correspondence with something as intangible as software.

Later, in the nineties, interface design surveys began appearing, with contributions by academics and professionals as well as companies and institutions involved in software design. (Galitz 1992, IBM 1991, 2001, Lidwell et al. 2003, Mayhew 1992, Microsoft 1992, 1995, 2001, Norman 1988, Open Software Foundation 1993, Verplank 1988, World Wide Web Consortium 2001, etc.). These surveys included principles that favoured efficient interaction. The purpose of interface design at the time was to build a relationship of trust with users of these new devices, so that they could be successfully introduced in our lives.

At the turn of the millennium, computers already formed part of our everyday lives, and that’s when the books on patterns and guidelines began to appear. These are much more practical than the books on principles. Their strong point is the fact that they offer a knowledge framework specific to interface design, based on professional practice, which means that designers don’t need to start from scratch with each project, and can instead apply solutions that have worked previously on similar projects. Their weakness of these is that they provide general solutions, what software designers call “one size fits all”,[1]. solutions that most designers apply without a second thought for whether or not they are the best option.

These two types of specialist reference books have taken interface design along a path that is lacking in reflection and critical thinking. So now we find ourselves with a discipline that does not denounce falsehoods or question the lack of context. A profession that accepts paternalism and does not encourage autonomy. A panorama in which all interfaces are similar, in which the transparency and trust of the early years are taken for granted even though they are actually an illusion.

This uncritical acceptance of interface design models is a serious matter. At a time when we spend approximately 7 hours a day in front of screens (Meeker 2014) [2], , we can’t ignore the fact that graphic interfaces are a very significant part of our cultural experience. The discipline of interface design needs to question the principles and patterns that have governed design up until now, and stop accepting them as untouchable dogmas.

We now need to promote a kind of interface design that seeks methodologies that can generate a more sincere relationship with users. And I don’t mean archetypal users, imaginary people, but users with strong links to a particular context. Users whose intellectual needs go beyond the efficient use of a system, and that designers should not culturally abandon.

This text reviews some of the fundamental principles of interface design. I invite readers to question my suggestions and the principles and patterns that are used in the design process. The aim of this proposal is to awaken the need for reflection in a field of design that is still in its infancy, but has become culturally important with a speed unmatched by any other.


Interface Design Principles that Promote Cultural Development

1. Interfaces should be challenging rather than accessible

According to the principle of accessibility “systems should be designed to be used by the greatest possible number of people, without modifications.” This principle is false, now more than ever. Interfaces are designed with “early adopters” in mind, because otherwise technological progress as we know it would be impossible. Think about touchscreens on mobile devices, for instance, a phenomenon that seemed extraordinary a few years ago and has now become commonplace. When touchscreens were launched on the market, the devices were targeted at a limited group of advanced users. It was these users who evaluated the first versions of the design and acted as trendsetters, so the interface became part the mainstream and the rest of us adopted it. It was precisely the decision to focus on advanced users that allowed designers to introduce new features that disrupted the way we interacted interact with devices. If they had continued to think in terms of a design that aimed to please everybody, none of this would have happened.

Therefore, I propose replacing the term “accessible” with “challenging”. Interfaces should be challenging for advanced users and for the mainstream, so that interaction with the system is an intellectually motivating experience.

2. Interfaces should be contextual rather than obvious

According to the obviousness principle, “the interface should make the system easy to learn and understand.” This principle is based on the idea that the design of the objects in an interface should in itself suggest how they should be used. This idea is usually called “affordance" [3], a term coined by the psychologist James J. Gibson in 1979. According to Gibson, affordances are the potential uses of an object that emerge from observing its properties. For example, a button suggests the action of pressing due to the property of volume. But today, in the midst of the maelstrom of “flat” interface design, most users understand that certain graphics are buttons, even though their physical properties do not indicate that this is so. Our perception of objects is clearly subjective, and the meaning of these objects does not only depend on its intrinsic properties but is largely determined by our previous experiences.

As such, I suggest replacing the word “obvious” with “contextual”. Interfaces should respond to the context in which they are used, and their design should be in line with the needs and specificities of the time and place where it is used.

3. Interfaces should be tolerant but also sceptical

EThere is a principle by which there should be tolerance for errors arising from the use of an interface. This principle seeks to protect users from committing irreversible errors, providing clear, concise information on how to avoid them and on how to resolve them. Nobody wants to accidentally delete a whole afternoon’s work, and we all agree that we should have the right to undo an ill-considered action. This principle is in fact of vital importance in getting users to trust a system. But the other side of this paternalistic attitude is the fact that it encourages acting without really thinking: it doesn’t matter what we do, because we can always undo it.

This lack of reflection on the consequences of actions is not without ethical problems. I propose that interfaces be tolerant, but also sceptical. That they encourage users to be curious about how they interact with them, help to disclose the values that arise from their use, and promote more conscious and responsible behaviour.

4. Interfaces should offer a sense of empathy rather than control

The principle of control encourages users to always have “a sense of control over interaction.” The idea is that the interface should always provide feedback on the actions we carry out, and that this will give us the sense of being in control. Generally speaking, this feeling of control is formalized through the display of simple responses – dialogue boxes, micro-animations, etc. – but these displays are simply an illusion of control over the machine. Interfaces today are so complex that expecting to place users in command is naive to say the least. I propose that interfaces provide empathetic rather than submissive feedback.

That instead of giving signs that they are executing the requested action they send signs that show the complexity of the system, generate curiosity, and encourage an attitude of exploration and learning.

5. Interfaces can offer customisation options, but they should openly communicate limits

According to the configurability principle, “interfaces should provide default configurations for new users as well as advanced customisation options for expert users.” This option to adjust the interface configuration is intended to allow users certain freedom in the way they interact with the system. The idea is to allow users to choose the way they want the information to be presented, how the interface behaves, and how they interact with it.

Nonetheless, this principle is highly questionable when interface design is subject to commercial and corporate criteria, as most interfaces today are. Think of Facebook for example. The service offers us an illusion of customisation in the way we present ourselves: we can include a photo in our profile and change it whenever we like. We can decide which friends we want, and which to block. But that is really all we can customise. We can’t change the colour, or the font, or the structure or the order of the information that represents us as individuals. And we can’t avoid the ads, or stop third parties from seeing our interactions with friends in common. All these decisions were made consciously, in line with decisions based on corporate interest. There is no attempt to guarantee levels of customisation, and there is no honest communication about the limits that are set. I think that interfaces should not be obliged to offer customisation options if the corporation behind them does not wish to. Nonetheless, I think that they have the responsibility to be honest with those who use it, and that they should communicate those limits clearly.

6. Interfaces should be simple, not just appear to be so

The simplicity principle encourages “the removal of all irrelevant information so as to achieve the simplest possible interface.” As I see it, we have reached a point where all systems include so many functionalities that they have to be kept out of sight to prevent them interfering in the main use of the device. An example is the “hamburger” icon included in many of the applications and services that we use. This not very expressive icon – three horizontal lines – is actually used to hide a whole amalgam of functionalities that get in the way, but that nobody dares to get rid of. I see it as evidence that attempting to design simple interfaces is not enough. When we design an interface, we should be bold enough to remove everything that is superfluous, until we get the system to reveal its essence.


Toward critical and socially responsible interface design

Throughout this text I have tried to critique some of the fundamental interface design principles published by Wilbert Galitz in his book The Essential Guide to User Interface Design; An Introduction to GUI Design Principles and Techniques”[4]. The exercise was not intended to pull apart the theories of interaction design experts. In fact, my aim was just the opposite. I still think that the literature on interface design is scarce and not very specialised compared to other areas of design. The idea of the exercise was precisely to challenge a body of theory that was written before smartphones and social media existed. Theory written with the eagerness with which you start something new, but too innocent when judged through contemporary eyes. I believe that interface design should stop repeating patterns without questioning them, and start to generate a critical approach both to the sources and the methodologies that it is based on. I believe that this revision is essential if we are to move towards the creation of culturally rich interfaces that encourage an active role in the construction of progress, and that guarantee the free action of their users.


  1. http://www.eetimes.com/document.asp?doc_id=1309668
  2. Meeker, M. (2014): Daily Distribution of Screen Minutes. BGR Media. http://bgr.com/2014/05/29/ smartphone-computer-usage-study-chart
  3. Norman, D. A. (1990). The design of everyday things. New York: Doubleday.