If you want to know when new articles go
online,
subscribe to the WebWord.com
Usability Newsletter!
Evolution,
Usability, and Web Design
Article by John
S. Rhodes
Abstract
The purpose of this article
is to explain how evolution and natural selection relate to the web
development process. It is suggested that it is wise to encourage
designers to create many quick and dirty designs over many short
intervals. This is in contrast to asking designers to create a lower
number of better designs over fewer intervals. The ideas of failure,
prototyping, usability testing, and iterative design are explored.
Background
Darwin's Origin
of Species is arguably the most important book on biology ever
written. Just for a moment, consider that web sites have similar
characteristics to biological organisms. In other words, consider that web
sites adhere to the ideas outlined by Darwin. What can we learn? Before we
answer that question, we need to review the concept of natural selection.
Natural selection is
driven by three core ideas. First, organisms create too many
offspring. There will be a struggle for survival. Second, organisms are
different. They vary. Uniqueness reigns. Third, organisms are most like
their direct ancestors. When you combine these three ideas it becomes
pretty clear that evolution by natural selection drives the success and
failure of organisms in an environment. In the long run, only the best
species survive.
The organisms most likely to
survive are the ones that best fit the conditions of the environment.
Stated another way, even if an organism is strong and fit it will not
simply survive and reproduce. Instead, the organism must be strong and fit
in a way that makes it likely to succeed in a particular environment. Drop
a shark onto dry land and it will die. It simply doesn't matter how much
of a killing machine it is in the water.
Darwin repeatedly stressed
the process of variation, followed by selection, followed by more
variation. This is the way the species change and adapt and evolve
over time. Change is the constant. Continuous improvement, in light of a
particular environment, is the key.
In light of Darwin's ideas,
diversity is a good thing. You create many types of similar organisms, but
not exactly the same, and then you let them compete. The environment might
crush them all but it is more likely that a few will survive and continue
the evolutionary process. The process is merciless in some ways, but it is
terribly efficient too.
Keep in mind that the
"best" do not always survive and prosper. Nature is cruel and
sometimes makes mistakes in the short run by rewarding the
"wrong" organisms. For example, Betamax was superior in many
ways to VHS but the market did not select it for survival. Nevertheless,
in the long run, evolution works quite well. It maintains a continually
shifting balance over the long haul.
The Obvious Importance of New Ideas
Shifting gears, if we examine
web design, it is obvious that many people have been experimenting on the
web. For years we've been seeing new and interesting designs. With each
new technology, such as Flash, web designers feel compelled to squeeze out
their creative juices. Interestingly, soon after this happens, usability
professionals rush in and declare that the technology and the designs are
terrible. I admit it, I've done this. Cutting-edge is often seen as an
acronym for anti-usability.
Remember when Flash
was 99% bad? For a moment, consider that almost all Flash web sites
are terrible. The very important point is that 1% of Flash sites are
outstanding. When you count the total number of web pages available, this
is a lot of success. I sincerely applaud the designers and developers that
have done great work with new technology.
However, we have to face
facts. Most new ideas suck. Really. Most new ideas, services, and
products are useless. Worse, they are often dangerous and they consume
precious resources, including time. Yet in spite of failure after failure
it is critical that we allow designers to have room to experiment. We need
people to design things that fail. Failure clears away the clutter.
Failure is just part of evolution. Let it happen because in the wake of
broken dreams and dashed hopes there will be survivors. Some designs will
win in the marketplace and that is exactly what we want. I'll take 99
failures and 1 amazing success.
Does Usability Try to Short Circuit Evolution?
I've been worried for a long
time that usability gets in the way of good web design. Let me say that a
different way. I'm concerned that usability doesn't allow designers to be
creative enough and it doesn't allow them to properly fail.
Deep down there is an
assumption that a representative sample of the user population will allow
usability testing to capture design problems. In large part I think this
assumption is correct. The core issue is that not enough variations are
tested. Usability does a great job helping designers select the right
features. It is also great at making sure the market will like those
features, and that the design will be useful. But, designers and usability
professionals do not put enough emphasis on making sure there is enough
design variation to work with in the first place. Usability professionals
sometimes rush in to save the day when it doesn't need saving.
Usability is a mechanism of
refinement. In the current context, usability is a selection mechanism.
Unfortunately, usability is dangerous because it can act like a design
contraceptive. It tends to hinder and limit variation. At a minimum,
usability is about refinement not about idea generation. Sometimes
designers need room to experiment. Let
one thousand flowers bloom.
The Burden of Design Choices and Iterations
There is a need to encourage
designers to design. Ideas are fuel for growth and choices are important.
Usability should not be brought into a project until it is clear that
enough ideas have been generated. While budgets don't allow for too many
prototypes there is a rule you can follow. It is better to create many low
fidelity prototypes for testing than to create just a few high fidelity
prototypes. Furthermore, as many people will agree, the earlier you are in
your design process, the more important it is to focus on creating many
low fidelity prototypes versus few high fidelity prototypes.
Simply stated, allocate your
design resources such that you produce as many competing designs as you
can, as early as you can. Usability testing can sort out the good from
the bad. A representative sample of users (i.e., the market or the
"environment") will help you select the best material to use.
The trick is to allow usability to guide your design decisions.
Don't neglect to generate enough good ideas to test and review.
Darwin helps us understand
why iterative design and iterative testing are important. It is better to
do a lot of quick and dirty testing over many design cycles than it is to
a few big tests over fewer design cycles. In nearly all cases, this style
of testing will be superior. This simply follows the evolutionary process
as Darwin explained it.
Nature isn't clean or neat.
Neither is web design. Balance the needs of your users with the creativity
of your design team. Users often don't know what they want until they see
it. Along the same vein, realize that new and cool ideas aren't usually
the first ideas. They are the result of repeated experimentation
and testing. You need ideas. Many ideas. Then bring in the usability
tests to kill the bad ideas. Then iterate.
Learn the tricks of evolution
and you'll learn how to design better web sites.
Summary
- Diversity is good.
Encourage creative thinking. Don't let usability limit your thinking.
- Generate a lot of ideas as
early as possible. Encourage competition between many designs.
- Kill off the bad ideas.
Keep only the best. Let your ideas have sex; you want offspring.
- There is nothing wrong
with 1% success. A high level of failure is inevitable.
Comments?
Please send them to me: john@webword.com
I want to know what you think about this article.
What next?
|