15 May 2018

The knack of interaction design

A slightly refined version of a Twitter thread I ranted in March 2017. I should turn this into a proper article someday.


Via @vgr I learn that some people cannot visualize things in their “mind’s eye”.


I am reminded of something I realized years ago about the Secret Talent Behind Interaction Design.

When I present an interaction design (IxD) solution to clients or colleagues, they often challenge me with “but what if the user does XYZ instead?” So after showing slides with a walkthrough of some key behaviors, I answer questions about other paths with quick sketches at the whiteboard.

“Oh, if a user clicks here in X situation, then A, B, and C light up, plus D appears in Y case, or E is shown in Z case.”

For a while I thought clients were so often astonished at this because of the obvious brilliance of my design solutions.

Not so.

Then I thought they were astonished at my facility at quickly communicating IxD solutions through simple whiteboard sketches.

Not so.

(For the record, my whiteboarding hand is inelegant, though I know a lot of little tricks for using whiteboards well.)

Then I saw Alan Cooper’s early dialogue with Kent Beck, with its astonishing disjoint of ideas. I finally realized that what astonished my clients was that I had the behaviors of the system in my head at all!

Most people simply cannot picture the interaction design of a software system which does not exist; they must build it to “see” it. As someone who has knack, it had never occurred to me that most people in the software industry could not “visualize” IxD. @exiledsurfer concurs:

took me years to understand what i could see in my head & drew / explained to others was invisible to them until it was manifest.

If you don’t have the knack for picturing IxD, it would never occur to you that someone could. I think much of the skepticism about IxD and UXD in tech comes from a resulting reasonable-but-wrong assumption that this work is just impossible; countless software industry practices are predicated on the assumption that attempting to do too much planning of projects inevitably fails.

Nor do I want to overstate how much planning I think is possible; software development is an unruly process!

Nor do I want to make IxD visualization sound easy. It is hard work. It can take weeks to lock down good IxD solutions in my mind, and I think many UXD projects are stillborn by just not having enough time committed to them, leading to We Tried Baseball.

Keeping even a moderately complex IxD solution in my head “fills” it; I cannot remember the details of past IxD projects because they get crowded out. But I can really have a whole unbuilt system in my head; part of how I know an IxD solution is good is that its logic is coherent enough to make this possible.

One commentator on the original thread, @archslide, suggested that visualizing IxD is more skill than talent. There is definitely a skill one cultivates doing the work; it is not simply a matter of talent. Good balance doesn’t make you a tightrope walker; but if you don’t have good balance, no amount of practice will make you one.

If UX designers’ “knack” were properly understood, I think it would radically transform the entire software industry.

Chris Doyle <@archslide>observes:

I have the same skill wrt code. Probably v difficult to succeed as a dev w/o abstract visualization/organization ability

Consistent with what I have seen with many developers. Ability to visualize deep software logic and interaction are similar skills, but don’t seem to be coupled.

No comments: