Are good IxDers like good chess players?

9 Jan 2006 - 10:16pm
8 years ago
5 replies
606 reads
Gabriel White
2005

In my morning malaise on the subway this morning I thought this:

Some of the best interaction designers I've worked with are like good
chess players. They are the kind of people who can think 39 actions
ahead, for each of the 27 different possible use scenarios. And they
are able to do this in their head.

As designers we try to focus on designing for the most important
stuff, and leaving feature creep on the roadside.

But when it comes down to actually making a design work, I believe
that the ability to imagine the 387 contingent possibilities is
critical to making the difference between an excellent and merely good
design.

Gabe

(Don't challenge me to a game of chess, I'm absolutely miserable at it)

www.smallsurfaces.com - mobile interaction design resources

Comments

9 Jan 2006 - 10:41pm
Jared M. Spool
2003

At 10:16 PM 1/9/2006, Gabriel White wrote:

>(Don't challenge me to a game of chess, I'm absolutely miserable at it)

So, being "like a good chess player" isn't like being a good chess player?

Inquiring minds wanna know...

10 Jan 2006 - 7:00pm
Dave Cronin
2005

I agree that like a good chess player, a good IxD needs to be able to
imagine into the future and run through scenarios in their head. I think
another interesting factor, though, is that (while I'm a pretty poor
chess player myself), I'm guessing that grand masters don't actually run
through 387 contingencies in their head because "intuition" or pattern
recognition helps them quickly gravitate towards the best one *without*
having to fully explore each permutation.

I would suggest that a good IxD behaves similarly.

Incidently, there was recently a fantastic article in The New Yorker by
Tom Mueller about computer chess, where "players" do run through each
permutation of the upcoming moves and use computational techniques to
determine the "best" move. There was some good commentary about the
differences between this and the way a human might play that are
probably relevant to this topic...

__________

Cooper | Product Design for a Digital World

David Cronin
Director of Interaction Design

office (415) 267 3504
mobile (415) 699 3036
dave at cooper.com | www.cooper.com

> From: Gabriel White
> Sent: Monday, January 09, 2006 7:17 PM
>
> Some of the best interaction designers I've worked with are
> like good chess players. They are the kind of people who can
> think 39 actions ahead, for each of the 27 different possible
> use scenarios. And they are able to do this in their head.
>
> As designers we try to focus on designing for the most
> important stuff, and leaving feature creep on the roadside.
>
> But when it comes down to actually making a design work, I
> believe that the ability to imagine the 387 contingent
> possibilities is critical to making the difference between an
> excellent and merely good design.

10 Jan 2006 - 7:39pm
mariaromera
2005

>Some of the best interaction designers I've worked with are like good
chess players. They are the kind of people who can think 39 actions
ahead, for each of the 27 different possible use scenarios. And they
are able to do this in their head.

It's funny, Gabe, just before I got to this email I was thinking how the solution I had just come up with for one flow breaks another.
Well, perhaps that's only 4 steps ahead ;-)

I think you've hit something... though I may be biased. It goes back to when I had an engineer as a mentor while doing my Masters thesis for a CogPsych/Human Factors degree -- he was all about state machines.

Now, as a quick definition you could think of state machines as a set of rules where each action takes you to a new state -- defining the behavior of a device. For example, I'm in a composition screen, I hit cancel, I get a warning note, etc.

I find this kind of thinking not only good for analyzing an existing system, but in thinking through a design. Flow charts are one very approachable way to visualize this kind of thinking and I use them often. Another way is to actually borrow from computer science the formal language of finite automata. You can check out the way I used this as an annotation for mental models here: http://www.mria.net/papers/mentalmodels.html. But even without getting so formal, you can use this "the system is in this state, I press this key, and this happens" kind of thinking. I'm specifying some of my key press behaviors in what could essentially be called a state transition table... so you don't have to get too complicated with this to be useful.

Cheers,
Maria

---------------------------------
Yahoo! Photos – Showcase holiday pictures in hardcover
Photo Books. You design it and we’ll bind it!

11 Jan 2006 - 8:56am
Josh Seiden
2003

>Some of the best interaction designers I've worked with are like good
chess players. They are the kind of people who can think 39 actions ahead,
for each of the 27 different possible use scenarios. And they are able to do
this in their head.

----

For what it's worth, I'm terrible at the kind of detail oriented thinking
required to discover every case. I'm a pattern person, and find it easy
instead to see the patterns of use. (Many interaction designers that I know
are this way too.) For this reason, I like to work with a partner--often an
engineer, a business analyst, or design communicator--who is good at
detailed case discovery. I find a very productive interaction takes place
when you put pattern thinkers and detail thinkers together.

For me, the lesson for individual designers is to recognize their personal
work style, strengths and weaknesses, and find a way to support weaker
areas.

JS

11 Jan 2006 - 10:56am
Gabriel White
2005

Josh - Interesting comment about patterns.

I've read about chess players that pattern recognition (rather than
explicit lists of possibilities) is important to success.

The good chess player will be able to identify better and worse
patterns, while not necessarily having an explicit description of why
a particular pattern is advantageous.

So, as designers if we can work using patterns (much less brain and
time demanding) to identify anomalous (where things break) or virtuous
(where everything just works) design situations, then we're probably
going to do our jobs better.

Gabe

> For what it's worth, I'm terrible at the kind of detail oriented thinking
> required to discover every case. I'm a pattern person, and find it easy
> instead to see the patterns of use. (Many interaction designers that I know
> are this way too.) For this reason, I like to work with a partner--often an
> engineer, a business analyst, or design communicator--who is good at
> detailed case discovery. I find a very productive interaction takes place
> when you put pattern thinkers and detail thinkers together.
>
> For me, the lesson for individual designers is to recognize their personal
> work style, strengths and weaknesses, and find a way to support weaker
> areas.

Syndicate content Get the feed