Interaction Design

The term InteractionDesign as used by AlanCooper refers to two very different things:

a new field of human endeavour
a methodology in that field

Interaction Design vs Cooper's Methodology of Interaction Design

Interaction Design is a field and a process, the essence of which is to do design in a comprehensive and systematic way. The chief failing of most design processes is that they don't examine the value of interactions, the keystone of interaction design. So interaction design is a qualitatively different kind of design process, as different from everyday design as chemistry is from alchemy.

Interaction design, as promulgated by Alan Cooper, is a methodology of interaction design. Confusingly enough, Alan Cooper refers to both the field and the methodology by the same name, and there's really no justification for this. Perhaps it's the case that currently his is the only methodology of interaction design, but there's no reason to believe that his is the only possible methodology of interaction design.

If interaction design is not a new field, then Alan Cooper is just a design methodologist like so many others and he is guilty of hyping his methodology. If interaction design is a new field, then Alan Cooper's methodology of interaction design is a separate thing from interaction design and Cooper is guilty of misleading people about his methodology's relation to that new field. Either way, Alan Cooper is guilty of gross miscommunication which disqualifies him as an authority on the subject. He may have coined the term but he no longer gets any say about what it means.

In order to AvoidNeologisms and because the preponderance of pages on this wiki will be about the field, we shall use the term Interaction Design to refer exclusively to the field and rename the methodology CooperInteractionDesign. As for the former, see just WhatIsInteractionDesign.

-- RichardKulisz

Interaction design isn't a new field. It's an age-old practice of harmonizing form and function that goes back to paleolithic tool-making. Outside of the software industry, getting usability right is normally important enough that there are people who make this work a career (e.g. industrial designers). If you want to manufacture a chair, you get someone to design it first. Most people understand that making a chair and designing a chair involve two very different skill sets.

Interaction design has become an issue in software development because we're finally admitting just how badly it's done:

we've been building interfaces designed by programmers who know nothing about usability, and don't think of it as a factor to be considered;
software development managers believe that all they have to do is make usability an issue and programmers will magically develop the skills to do it well;
the external design of software products is neither practiced nor recognized as a distinct aspect of software development or, more importantly, as the essential component of requirements specification.

Mature industries don't make these kinds of mistakes. What's truly frightening is that the latest fads in "methodologies" exacerbate the problem.

-- MarcThibault

I suppose the drive-by deserves a reply. I solved my first tough interaction design problem in 1965 - the control panel for a 27-axis machine tool control (the first NC to use software). We called it interaction design then, and usability was always the objective.

We needed a push-button interface for a bunch of guys with grade-school education who had spent their lives moving levers and spinning wheels. It sounds strange now, but in 1965 the touch-tone phone hadn't been invented and the only people adjusted to push-button control were Edsel owners. Designing a human interface that implemented an alien paradigm and required only on-the-job training was fun to say the least. The biggest problem was operator stress - it took a while for them to trust that the machine would do what they keyed and not end up shooting bits of broken metal all over the plant.

If it's not about usability, why would we waste time on it? -- mt

I don't know, why did FrederickWinslowTaylor do it?

What are your positions, if any, on:

usability engineering is not interaction design; the former is low-level measurement-based, the latter is high-level conceptual. Agreed. UsabilityEngineering provides information for interaction design.
typical design of software is not systematic enough to be called interaction design; it doesn't question the value of interactions. Sadly, too true.
typical design in other fields is not complex enough to be called interaction design; there simply isn't enough interaction. Disagree. Complexity is not a requirement.

-- RK

My take on the first bullet: Both are necessary. If a (bad) analogy is to be formed; the former is analogous to the role of a structural engineer in building design (or design of other public works); the latter is analogous to the role of the building architect. Structural engineers are necessary to generate sound "low-level" designs for building and to verify/ensure structural integrity; they generally aren't concerned with FitnessForPurpose or aesthetics. The architect is responsible for the overall scope of the project, including aesthetics and functionality. On any complicated project, both sets of skills are necessary.

For software design, a similar principle can be applied. Usability engineers can be used to answer specific questions about an interface design; and assist with transforming a concept to a sound design. Interaction designers come up with the concept designs, and are responsible for the interface as a whole.

With regards to the second bullet, I agree for the most part. Exceptions do exist, of course. In many cases, it doesn't matter; one doesn't commission an architect when one wants to build a tool-shed, after all.

I disagree with the third point. Many other systems (such as the human interface in your car, or the cockpit of a 747) are just as complicated if not more so than a software system. In many cases, these aren't just "interfaces" but control systems, with a human in the feedback loop; and with catastrophic consequences if the human screws up. Some of these interfaces are largely solved problems, or dictated by standardization, but still automakers ship cars with controls in the "wrong place". It should also be noted that operation of some of these systems requires licensure; users are expected to train and become familiar with the systems before operating them unassisted.

-- ScottJohnson

So we agree. The way Marc expressed himself left me with the unfortunate impression that:

usability engineering and interaction design are the same thing
the systematic nature of interaction design isn't significant enough to make it a qualitatively different field
designing for the self-evident and trivial functional requirements (goals, not tasks) of a simple industrial age tool such as a chair is equivalent to designing for the non-obvious functional requirements of a cockpit or 3D editor (a chair's complex ergonomics fall into usability engineering, not interaction design)

-- RK

Some things you interact with; others you merely "use". Chairs, it would seem, fall into the latter category.

However, one thing I would like to emphasize is that while UsabilityEngineering is not sufficient for many classes of tools, it is an error to regard it as an "inferior" or "second-rate" endeavor. There have been comments in this space to that regard, suggesting that InteractionDesign is somehow a "higher" calling. That's like saying that being a building architect is a more noble profession than being a structural engineer - rubbish. Both functions are necessary; for advocates of one function to engage in a PissingMatch with the other over who is better or more important is counter-productive.

It is a higher calling. It's higher level. What other use of "higher" is there? "Higher level" refers to an abstraction hierarchy. "Higher calling" suggests something more worthy of society's praise and attention. I view the two as orthogonal.

As for "noble", please define it in a way that's meaningful to modern society. See above; worthy of praise or attention.

As for interaction design being more worthy than usability engineering, this is a fact I did not allude to or insinuate anywhere, let alone state outright. If there's a pissing war, you started it. But I'm quite willing to fight it; the "let's all get along" idea is really quite orthodox.

Let me make clear that I'm not advocating the opposite; nor do I wish to denigrate the area of InteractionDesign. If we agree that elevating InteractionDesigners upon a pedestal isn't a good thing, then we agree.
Sure.
Assuming this is serious, then first ideas can't be licensed. Second, neowiki supplants your particular vision of C2 wiki. A fascist PoMo wiki could easily be implemented within neowiki, and the exact opposite could be implemented even more easily. Third, screen shots wouldn't show behaviour. Fourth, I have a priority list and neowiki isn't top priority. More like 3rd or 4th. Which is actually too bad since it would help matters if neowiki were released simultaneously with the universal catalog. Fifth, I'll entertain help from concrete people not anonymous sources.
Ok, but ideas can be patented, now you despise intellectual property and anything related, but later you can change your mind.Or somebody may acquire patents for the same things before you do. I don't think my Wcp can be implemented within something that will likely never exist. Look, you seem to have a problem with programming: first you despise it, and second, whatever revolutionary ideas you may keep in your head, nobody will believe it until you have at least one prototype. Programmers can help with that, me included. But you have to show us something. Universal catalog, the new OS's desktop , the 3d object browser, whatever you feel like it. Faster than you program you can create a sequence of screen shots, an animated presentation, anything that can communicate your ideas. This is an InteractionDesign problem, isn't it? You have some ideas you don't seem to be able to communicate them to the rest of the world. -- Costin

WikiPedia's take on the subject: http://en.wikipedia.org/wiki/Interaction_design

It's impossible to get a clear idea what interaction design is from the entry because it's so incoherent. Also, it confuses interaction design with graphics design and marketing, which is simply wrong. Its focus on human users, user experience, GUIs and interfaces, is also wrong. These things are part of interaction design but they are not what it's all about.

CategoryInteractionDesign