: interface vs interaction [was: How to Get Into ... Best Job Title]

29 Jan 2004 - 2:22pm
1396 reads

This discussion, which seems to have hit a soft spot with many of us,
reminded me of something that I'm sure will put a smile on your face.
It is from around 4 years ago, when it was trendy in the design community to
oppose Jakob "Useit" Nielsen. (Might still be trendy...)


And an applicable one (though not as 'usable') :

/not trying to make a point



From: Todd R.Warfel [mailto:lists at mk27.com]
Sent: Thursday, January 29, 2004 10:53 AM
To: Bob Baxley
Cc: Interaction Designers
Subject: Re: [ID Discuss] : How to Get Into ... Best Job Title


Thanks for your insightful retort. I would agree that Google is hardly a UI
masterpiece from a visual design perspective, which is precisely why I
prefaced my statement with "'work' well." Google works. And from an
interface and interaction perspective, it's a success. Could it have a
better visual design that is more appealing? I don't think any of us would
argue against that.

On Apple - from what I have read, they dissolved their UE group. However,
they still perform focus groups and user testing on their applications. I've
had direct conversations with some of their leads on several of the
applications mentioned in my post. So, while it might seem like a tyrannical
run, it in fact is not. Yes, Jobs has heavy influence, he is after all the
captain of the ship. But I can assure you that the designs are not all based
on and validated by his personal opinions. There is real user research and
validation going on.

Whether we like it or not, visual design is typically equated to style.
Which is why it is typically considered subjective. There are elements,
which can be tested, like the size of icons. That you are correct in.
However, items like colour, which tends to be a significant driver in visual
design, are subjective. I won't debate that colour theory is valid, I'm a
firm believer in it. But at the end of the day, if the CEO hates blue, then
you find another colour. That is why I say it becomes subjective.

That's not to say the IA and interaction design cannot be subjective, as
they can be, technically. However, this is less of an issue for these
disciplines, as we have reliable, accepted methods for testing and measuring

On Jan 29, 2004, at 1:34 PM, Bob Baxley wrote:


A few retorts to one of your recent messages.

You said...

--- snip --

Perfectly, I can't really think of many. That "work" well, I can think of
dozens. Just a couple: the automobile, my Asco washer and dryer, the wrist
watch, Google, Apple's Mail application, iTunes, iPhoto, iDVD, iMovie, etc.

-- end snip --

Although I could take issue with the automobile and the wrist watch
examples, I'll leave those for another day. As for Google, it's hardly a UI
masterpiece so much as it's simply a more effective search engine. I'd put
it more on par with Walmart, McDonalds, and Dell. Successful to be sure, but
more in spite of design than because of it.

Which brings us to your last 5 examples, all of which are from Apple. The
notable thing about the work coming out of Apple right now is that every
last pixel, click, screw, and cable is reviewed, examined, and decided on by
a single individual: Steve Jobs. Not unlike the Lord of Rings trilogy, what
you are seeing from Apple is the large-scale expression of one man's
tyrannical, dictatorial, exceptional taste. To my knowledge, Apple has
closed their usability labs and does not perform focus groups for product
concepts. Steve and the designers that work for him are following their own
vision and bringing that to fruition in the same way as a novelist and

Clearly not a reproducible method for most organizations but definitely a
data point that multi-disciplined committees (er, I mean teams) do not
necessarily trump the efforts of talented individuals.

Later you said...

-- snip --

Visual design is much more subjective than interaction and IA. Interaction
and IA can be tested, proved, etc. Visual design is much more difficult to
(dis)prove, since it is so subjective.

-- end snip --

I definitely take exception to this statement on two fronts. First, visual
design is NOT subjective -- unless it's decoration, in which case it's not
design. Design as a discipline and a grand tradition, is unimpeachably a
practice of analytical problem solving. Visual design, if it is to be called
design, has to solve a stated and understood communication problem. For
example, Paul Rand's original UPS logo successfully communicated meaning
about the company and their purpose. Compare that to their new logo which is
little more than a meaningless shape, a fashion statement, a multi-million
dollar junk of eye candy that's likely to be replaced as soon as the
executives grow tired of it.

Any time you hear someone evaluate a design by using the phrase, "I don't
like..." you can be assured that you have left the realm of design and
entered the transient, commoditized, and soon to be off-shored, world of
style and fashion.

Similarly, I would argue that the belief that IA and interaction design
solutions can be "proven" is one of the greatest lies and liabilities ever
foisted upon the design profession. While I'm a great advocate of validating
solutions with actual users, I also know that there is a lot of nuance and
subtlety to those designs that can never be effectively tested or evaluated.
The vast majority of usability tests are performed on users who are seeing
the product or a particular piece of functionality for the first time. As a
result, those tests are really about the discoverability and learnability of
the design. For some products, perhaps even most products, those two
dimensions of the design are indeed the most important. However, for a host
of other products, Photoshop for example, the more important dimension is
efficiency of use, a dimension that I've never seen tested.

At the end of the day I'd put it like this: design is a discipline devoted
to analytical problem solving. The degree to which a design is successful is
the degree to which it solves a stated and understood problem. Similarly,
the value of an individual designer is determined by their ability to
successfully solve such problems in a repeatable, predictable, and
observable manner.

Not to preach too loud but... Design is not guesswork. Design is not magic.
And design is not subjective.

Finally, thanks for sending out the summary. A nice job of thinning the
underbrush and improving our view of the trees.



Bob Baxley :: Design for Interaction

design :: www.baxleydesign.com

blog :: www.drowninginthecurrent.com

On Jan 29, 2004, at 7:22 AM, Todd R.Warfel wrote:

On Jan 29, 2004, at 2:09 AM, Andrei Herasimchuk wrote:

[...] I have yet to see a project succeed on all levels without that person.
If you know of a project that has, please let me know.

AT&T Wireless eCommerce site. It was done with team of:

(2) Information Architects

(1) Usability Specialist

(4) Visual Designers

(1) Art Director

(1) Creative Director

(10) Developers

A very successful model, which as increased conversion rates, decreased
support costs, decreased acquisition costs, reduced bail-out rates. The list
goes on. And there are more:

the automobile, Nokia's cell phone interface, the original Apple GUI, etc.

[...] Too many teams are heavily weighted towards one side of the equation,
or front loaded with visual people only. Or interaction people only. Or
information people only. And when this happens, if the manager is only
experienced in one area, they will implicitly or unconciously weight their
decisions in favor of what they know.


Mac OS X is a classic example today where this imbalance is playing out.
It's a gorgeous looking OS, but it severely lacks in the behaviors
department. All the interaction people were let go. Jobs is pushing
eye-candy over function, and what you have now is a bunch of pretty looking
interface widgets that behave very badly. The dock anyone? The new Finder in
Panther? The Open dialog? All of them are train wrecks.

The dock and finder work. And work well. They're very functional. Both are
great for increasing productivity (allow users to quickly access regularly
used items), keeping users aware of system status (ability too show how many
new mail messages you have), etc. They each have their pitfalls, as every
system and solution does, but they are by no means a train wreck.

"Train wreck" wreck is a personal opinion. I would be interested in seeing
some evidence that they are "severely lacking in behaviors" or that they
don't function.

I agree they could be improved, as all systems could be. But they're very
functional, productive, and usable.

And if it's the "trash can" isn't in a stable location argument - please!
It's always at the right side of the Dock. We've tested this "theory" in
some work I'm doing here at Cornell and people don't seem to have a problem
with it. They find the trash can just fine.

But can you honestly tell me when you use Amazon that they have their act
together across the board? That the visual appearance is as up to snuff as
their ability to deliver product in more personalized forms? That the
navigation scheme they use makes any bit of sense, and so little that
everyone pretty ignores it and falls back on the search function? That the
manner with which you browse items really makes you feel as good as browsing
items in the store?

Different medium than a store, but Yes, Yes, Yes, and Yes. We've tested
eCommerce sites as well here at Cornell, as well as large index like Yahoo!
And users are able to satisfy your questions above. Again without a great
deal of problems (usability or interaction).

Most of my factual evidence would come from getting into specifics about the
companies I have worked for, and that would do nothing but get me in
trouble. Further evidence would come from knowing what I know about how many
companies operate, and that would not make me many friends.

I can respect confidentiality, but that really doesn't help us here. Are
there other non-confidential supporting elements you can provide?

So instead, I ask you this: How many technology products out there are well
designed? (I can count the number on one hand.) I mean really well designed,
that hit all cylinders, cover all bases, and hum perfectly?

Perfectly, I can't really think of many. That "work" well, I can think of
dozens. Just a couple: the automobile, my Asco washer and dryer, the wrist
watch, Google, Apple's Mail application, iTunes, iPhoto, iDVD, iMovie, etc.

But there are far more that fall short: Flash - very powerful, but could be
easier to use. It's gotten much better over the years. Illustrator -
seriously, why don't we have underlined text and multi-page support with
master pages (real multi-page support). Word - where do I begin?

The definition of the interface is stated in my article. Given that an
interface comprises of visual, information and interaction design solutions,
how does that defy my argument?

American Heritage Dictionary:

in*ter*face n.

3. Computer Science. a. The point of interaction or
communication between a computer and any other entity, such as a printer or
human operator.

b. The layout of an application's graphic or textual
controls in conjunction with the way the application responds
to user activity: an interface whose icons were hard to remember.

WordNet 1997 (c) Princeton University

2: (computer science) a program that controls a display for the user
(usually on a computer monitor) and that allows the user to interact with
the system [syn: user interface]

Both these definitions focus on the visual display element - the point at
which a user interacts with a system. They do not infer anything in relation
to information architecture. WordNet doesn't include interaction, but rather
states that it "allows the user to interact" with the system. American
Heritage infers a relationship between visual display and the way an
application responds to user activity (interaction design). So, by
definition, visual display is included, interaction has a relationship with,
but isn't necessarily included, and information architecture is not
included. That's why I said that the very definition of interface defies
your argument.

It is possible that we need to update the definition, if you are correct
that interface design includes visual display, interaction, and information
architecture. I don't think it does, but if it's determined that it does, we
should update the definitions.

A very good friend of mine went to Cornell as well in the architecture
school. His opinion on the matter of multi-disciplinary seems to be in line
with mine on this subject, and a large part of it was due to his experience
at school. I'll have to ask him for more specifics on his opinion.

I guess this is the part that needs clarification. Are you proposing that:

a) each unique, yet interdependent discipline have knowledge and
appreciation for the others

b) they are not unique, interdependent disciplines, they are all one
discipline and here's why...

c) something else

I'm in agreement with A, but not B for reasons previously stated. If it's C,
I'd be interested to see...

I am not suggesting they are one discipline. I said an interface designer
needs to master all three. I said, "Interface design is the larger role,
comprised of three smaller categories, areas of specialty that we are
already familiar with: graphic design, information design and interaction

It's the master all three that concerns me. Jack of all trades, master of
none. Visual design is much more subjective than interaction and IA.
Interaction and IA can be tested, proved, etc. Visual design is much more
difficult to (dis)prove, since it is so subjective.

In the decade or so that I've been doing this, I've come across only a
handful of people who are truly exceptional visual designers, information
architects, and interaction designers all in one package. I've come across a
dozen or so who are truly exceptional at information architecture and
interaction design in one package. I've come across hundreds of individuals
that are truly exceptional visual designers, truly exceptional information
architects, or truly exceptional interaction designers as individual

Possible? Yes. Probable? Not really. I think Cooper explains the reasoning
why rather well - we each have different, yet interdependent goals. So, our
focus is different. Our goals are different (same high level goal of great
product, but different granular goals).



Todd R. Warfel

User Experience Architect

MessageFirst | making products easier to use


Contact Info

voice: (607) 339-9640

email: twarfel at messagefirst.com

web: www.messagefirst.com

aim: twarfel at mac.com


In theory, theory and practice are the same.

In practice, they are not.


Interaction Design Discussion List

discuss at interactiondesigners.com


to change your options (unsubscribe or set digest):


Questions: lists at interactiondesigners.com


Announcement Online List (discussion list members get announcements already)





Todd R. Warfel

User Experience Architect

MessageFirst | making products easier to use


Contact Info

voice: (607) 339-9640

email: twarfel at messagefirst.com

web: www.messagefirst.com

aim: twarfel at mac.com


In theory, theory and practice are the same.

In practice, they are not.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.interactiondesigners.com/private.cgi/discuss-interactiondesigners.com/attachments/20040129/77b9c609/attachment-0001.html

Syndicate content Get the feed