Usability = Predictability

24 May 2008 - 7:54pm
5 years ago
61 replies
3048 reads
Robert Hoekman, Jr.
2005

"Usability equals predictability."

As in, if you can accurately predict what's going to happen next in an
interaction, it's because the action you're taking is understandable, clear,
logical, makes you feel confident, etc. If you can accurately predict what's
next, the interaction has high usability. If you can't accurately predict
what's next, the interaction has low usability.

Shoot holes in that statement.
-r-

Comments

30 May 2008 - 7:01am
Todd Warfel
2003

On May 30, 2008, at 1:45 AM, Steve Baty wrote:

> Hmmm, this reminds of the automatic lights in the bathrooms where
> I'm working currently.
>
> 2008/5/30 Gavin Burke|FAW <gavin.burke at futureaudioworkshop.com>:
>
>>
>> You walk up to a door way, you put out your hand to push in the
>> door and it opens automatically.

First time use, but after that, you come to expect automatic systems
like lights, doors, and toilets flushing.

Cheers!

Todd Zaki Warfel
President, Design Researcher
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
Twitter: zakiwarfel
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

30 May 2008 - 11:37am
Robert Hoekman, Jr.
2005

>
> You walk up to a door way, you put out your hand to push in the door and it
> opens automatically.
>

Another good example.

That said, I'm now wondering if the statement that started this thread
really implies exclusivity. As in, yes there are times when something that
is unpredictable is still quite usable, but that doesn't nullify the
equation. Predictable interactions are, it seems, still usable interactions.

So, while predictability isn't the only way to deem an interaction usable,
it does still seem to stand true that predictable interactions are usable
regardless.

Arguments?

-r-

30 May 2008 - 11:42am
Robert Hoekman, Jr.
2005

>
> Usability is a scale from extreme frustration to extreme delight.
>

Been thinking about this statement. Is it really true? Isn't "usability"
only a measure of the degree to which something can used?

At the worst end of your continuum, "extreme frustration" doesn't
necessarily mean "unusable". Take the Word example, in which an interaction
was frustrating all the time, but because there was a hack to get around the
issue, ultimately, it was still usable.

Seems to me that frustration and delight are measures of "enjoyability", not
"usability". I think there's a difference, and that the difference is
important.

Yes, highly usable interactions do generally seem to be more enjoyable, and
less usable interactions do seem to be generally more frustrating, but a
less usable interaction doesn't have to mean frustrating, and "frustrating"
doesn't have to mean "less usable". The two are not mutually exclusive.

-r-

30 May 2008 - 4:15pm
Jared M. Spool
2003

Now you're killing kittens.

There are two ways to measure usability.

You can use a black-and-white binary measure, where something is
either usable or not. To conduct the measure, you ask 'n' people to
perform tasks. If it crosses your acceptance threshold of success, you
have created something usable.

However, that's not a really useful measure, beyond it just basically
working. You have to get there first (something that only works for
people 25% of the time probably needs a lot of work). But, once you
get past the threshold, you're definitely not done.

That's when the scale comes in as the second measure. What makes
something "more usable"? First, you have to hunt down all the sources
of frustration and eliminate them. Once you've knocked them off, are
you done? There's probably ways to make something more usable by
increasing the delight -- reducing work flow, adding features,
creating experiences that make the user pleased and passionate about
the product/service/brand.

We can argue over semantics of whether something that isn't
frustrating has met basic use and therefore can't be more usable, but
I'm not sure there's value to that. In addition to saving the kittens,
avoiding that discussion has the advantage of helping teams focus on
what's important: creating a great experience. Who cares if the great
experience is coming because it's more usable or more enjoyable?

I didn't make this crap up. It's actually the work of Kano. You can
learn more about it here: http://tinyurl.com/65fv56

Or you can see this unfortunate rendition of it here: http://tinyurl.com/66qnfo

Hope this helps.

Jared

Jared M. Spool
User Interface Engineering
510 Turnpike St., Suite 102, North Andover, MA 01845
e: jspool at uie.com p: +1 978 327 5561
http://uie.com Blog: http://uie.com/brainsparks

On May 30, 2008, at 12:42 PM, Robert Hoekman Jr wrote:

> Usability is a scale from extreme frustration to extreme delight.
>
> Been thinking about this statement. Is it really true? Isn't
> "usability" only a measure of the degree to which something can used?
>
> At the worst end of your continuum, "extreme frustration" doesn't
> necessarily mean "unusable". Take the Word example, in which an
> interaction was frustrating all the time, but because there was a
> hack to get around the issue, ultimately, it was still usable.
>
> Seems to me that frustration and delight are measures of
> "enjoyability", not "usability". I think there's a difference, and
> that the difference is important.
>
> Yes, highly usable interactions do generally seem to be more
> enjoyable, and less usable interactions do seem to be generally more
> frustrating, but a less usable interaction doesn't have to mean
> frustrating, and "frustrating" doesn't have to mean "less usable".
> The two are not mutually exclusive.
>
> -r-

30 May 2008 - 4:24pm
Jared M. Spool
2003

On May 30, 2008, at 12:37 PM, Robert Hoekman Jr wrote:

> That said, I'm now wondering if the statement that started this thread
> really implies exclusivity. As in, yes there are times when
> something that
> is unpredictable is still quite usable, but that doesn't nullify the
> equation. Predictable interactions are, it seems, still usable
> interactions.

Frankly, I think we've just proven that (a) the terms are orthogonal
and (b) nobody really knows what they mean when others use them.

If you're trying to assert that predictability is a predeterminate of
usability, I think we've showed that you can have usable, non-
predictable instances. There are times when something that's perceived
as predictable will be perceived as more usable than something that
isn't predictable. However, there are also times when the opposite is
true. So, I think they are truly orthogonal attributes.

Jared

Jared M. Spool
User Interface Engineering
510 Turnpike St., Suite 102, North Andover, MA 01845
e: jspool at uie.com p: +1 978 327 5561
http://uie.com Blog: http://uie.com/brainsparks

30 May 2008 - 5:47pm
Eugene Chen
2004

>> Seems to me that frustration and delight are measures of
"enjoyability", not "usability". I think there's a difference,
and that the difference is important.

I agree. I would not push "usability" too far to become an umbrella
term. Frankly, it's kind of a dry, uninviting term. To me describing
a design as "usable" sounds like describing a meal as "edible".
As others have said on this list before, "usable" would seem to be
the minimum goal for a design. Of course, as someone who does this
every day, I acknowledge how surprisingly difficult this minimum is
to achieve - and that making food edible is certainly the first
priority.

But assuming we're maintaining reasonable "usability", there ought
to be many qualities we can aspire toward. Take for instance,
"interesting", "intriguing", "amusing", "fascinating",
"absorbing", "fast-paced", "punchy", "diverse", "soothing".

I agree with the reformulation If predictable > Then usable. But
notice that "predictable" would be considered an insult in a
restaurant review.

Eugene Chen Design
User Experience | Strategy Research Design
http://www.eugenechendesign.com

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=29451

1 Jun 2008 - 9:13am
Jim Hoekema
2004

I think of a usable interface as one that meets user expectations. Some
expectations are pre-existing -- hence predictability. Others are
learned through interacting with the interface -- they might be
unpredictable at first, but if the interface is consistent, they soon
become predictable.

- Jim Hoekema

HDE . www.hoekema.com <http://www.hoekema.com> . (845) 401-7466 .
www.linkedin.com/in/hoekema <http://www.linkedin.com/in/hoekema>

Jared Spool wrote:
>
> On May 30, 2008, at 12:37 PM, Robert Hoekman Jr wrote:
>
>> That said, I'm now wondering if the statement that started this thread
>> really implies exclusivity. As in, yes there are times when something
>> that
>> is unpredictable is still quite usable, but that doesn't nullify the
>> equation. Predictable interactions are, it seems, still usable
>> interactions.
>
> Frankly, I think we've just proven that (a) the terms are orthogonal
> and (b) nobody really knows what they mean when others use them.
>
> If you're trying to assert that predictability is a predeterminate of
> usability, I think we've showed that you can have usable,
> non-predictable instances. There are times when something that's
> perceived as predictable will be perceived as more usable than
> something that isn't predictable. However, there are also times when
> the opposite is true. So, I think they are truly orthogonal attributes.
>
> Jared
>
> Jared M. Spool
> User Interface Engineering
> 510 Turnpike St., Suite 102, North Andover, MA 01845
> e: jspool at uie.com p: +1 978 327 5561
> http://uie.com Blog: http://uie.com/brainsparks
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>
>

--

<http://www.linkedin.com/in/hoekema>

1 Jun 2008 - 10:58am
Chauncey Wilson
2007

> I agree with the reformulation If predictable > Then usable. But
> notice that "predictable" would be considered an insult in a
> restaurant review.

Actually, predictable or some synonym, could be either an insult or a
compliment. A restaurant might have something that is an incredible
delight, but if the quality from day to day is not predictable, a
review might very well note that as a negative as Frank Bruni, The New
York Times food critic did with a review of a 3-star New York
restaurant:

Under those terms there's a promise of unwavering transcendence, and
Ko in its early months serves a few dishes that merely intrigue along
with others that utterly enrapture. It also falls prey to some
inconsistency.

"Under those terms there's a promise of unwavering transcendence, and
Ko in its early months serves a few dishes that merely intrigue along
with others that utterly enrapture. It also falls prey to some
inconsistency." (Frank Bruni, May 7, 2008)
http://events.nytimes.com/2008/05/07/dining/reviews/07rest.html

With some exceptions, the best chefs do strive for culinary processes
that make the quality of each serving quite predictable, though the
particular recipe itself may be a culinary delight and tremendously
innovative. A steakhouse that doesn't have a predictable medium-rare
steak could easily be skewered in a food review.

Predictable could be an insult if the reviewer notes that the menu
barely changed in the time between reviews spaced a few years apart.
So, predictable could be good or bad in a restaurant review. It could
also be neutral comment or a general selling point for reasonable
mediocrity. MacDonalds, for examples, tries very hard to have their
fries taste the same everywhere so people who are traveling have a
predictable fry. MacDonald's also has regional foods -- when I was at
the CHI conference in Florence, there was a MacDonald's a block or so
away and they had Fries that tasted like those in Boston, but they
also had breaded shrimp as an appetizer to local flavor.

Chauncey

5 Jun 2008 - 8:09am
John Vaughan - ...
2004

Belated comment on this thread:

"Consistency : Predicatbility : Usability" might be expressed as more of a
Venn diagram?

Or perhaps more nuanced, with "Expectations" and "Anticipation" somewhere in
the mix.

I've always felt that the most successfully generous UI's are the ones that
provide a sense of "what's down the road":
* A set of search results that provide some context for the list
* A stepped process that tells you how many pages, questions or minutes are
required
* A sense - not just of opportunities - but of scope

Always a balancing act, of course.

5 Jun 2008 - 8:26am
SemanticWill
2007

Some good points here,
and I wanted to get back to the original concept which was
Predictability=Usability blatantly ignores some key tenants from what we
should all remember from our HCI and Cog Sci classes - considerations for
usability foremost in my mind is cognitive workload of our user. How a
good designer's design take into account working memory and design from the
perspective that it is a limited resource; it is transient and is limited to
capacity and time. It is affected by fatigue, context, motivation, anxiety,
pre-formulated mental models, age, the time of day (and the day of the
week), technological aptitude, disabilities, tacit domain knowledge,
information structure, sensory interference from Dave M's Tweets, tasks,
framing, priming, good beer, the Celtics tonight and many other factors that
are in the literature. I think the problem is complex but not surmountable -
but reductio ad absurdium pronouncements make me wince and contemplate forms
of escapism because we should address all those variables when designing and
not think that a simple formula like making a system predictable will ipso
facto make it usable (simple, enjoyable).

- W

On Thu, Jun 5, 2008 at 9:09 AM, John Vaughan <vaughan1 at optonline.net> wrote:

> Belated comment on this thread:
>
> "Consistency : Predicatbility : Usability" might be expressed as more of a
> Venn diagram?
>
> Or perhaps more nuanced, with "Expectations" and "Anticipation" somewhere
> in the mix.
>
> I've always felt that the most successfully generous UI's are the ones that
> provide a sense of "what's down the road":
> * A set of search results that provide some context for the list
> * A stepped process that tells you how many pages, questions or minutes are
> required
> * A sense - not just of opportunities - but of scope
>
> Always a balancing act, of course.
> ________________________________________
>

5 Jun 2008 - 9:58am
Adam Connor
2007

Using Todd's definition of predictability above:

"Predictability means you before you perform an action, you can
assert with a good degree of accuracy what will happen next."

I'm wondering how the following scenario stands against the notion
that usability = predictability.

Consider the scenario where a user cannot determine what action to
take to initiate a task/process.

Is the user really predicting anything at this point?

Would you argue that the user has made a prediction as to the fact
that they need to perform some action to initiate the process? And
that since they cannot validate their prediction by identifying the
action to take within the interface the argument predictability =
usability holds?

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=29451

Syndicate content Get the feed