|
[Dallas,
May 17th 2000]
In
the recent tradition at uidesign.net of flying in the face of the
popular wisdom, we would like to suggest that making screens bigger
just isn't the answer for the problems of interacting with smaller,
wireless devices, or even big wired devices for that matter. Making
the screen bigger is just the current fashionable red herring to
distract you from the plain and simple fact that the vendors can't
get the design right. Let's talk about screen real estate!
In
his recent book, "Designing
Web Usability", Jakob Nielsen writes, "Get a big screen....
the $1,000 or so for a high-end 21" monitor may be one of the
best investments you ever make. ... it is much easier to perform
most tasks when you can take in more information at a single glance."
What exactly did he mean by this?
Well,
firstly this extract from page 31 was addressing web developers,
whom he argues require a bigger screen so that they can have several
windows open simultaneously e.g. a Dreamweaver window, an email
window, a web browser window. This allows the web developer to conduct
his daily business without having to constantly open and close,
minimize or expand windows. The argument is that you must have a
big screen to do this.
Modal
vs A-Modal
Let
us return briefly to the early 1980s and consider the days when
WIMP environments were new. GUIs it was claimed were the "Silver
bullet" for usability. They would allow people to perform tasks
graphically rather than through clumsy command line interfaces.
At the root of this claim was the belief that GUI environments freed
the machine from the highly modal command line style interface and
provided an a-modal graphical one. [ Put aside momentarily the argument
that minimizing and expanding windows is a form of modality ]
The
benefit of the a-modal GUIs was simple, the User could choose how
and when she wanted to do things and would not be restricted by
the current mode of operation. For example, when using a word processor
on a command line screen, you would first invoke the word processing
program which would take control of the screen. Everything you typed
would be input to the word processor. You could not for example
type "dir *" and expect a file directory to appear. First
you would have to close the word processor returning the machine
to command line mode and out of word processor mode, then your "dir
*" command would work.
For
the designers of operating systems for general purpose desktop computers,
a-modal interfaces presented huge advantages. They could provide
a general purpose environment which could support many applications
running at once [ at least visually, as not all implementations
were genuinely multi-tasking ].
With
an A-modal environment clearly a large screen is an advantage. Why?
Simply because a large screen maximizes the possible number of modes
which can be run or accessed simultaneously within the a-modal environment
of the operating system.
Be
aware of the messenger's background
Much
of the literature of UI Design published in the 1980s was written
by guys who were doing research on operating systems. Research was
producing results and results were published as facts. Just like
today with web design, the mistake that was made then was to overly
generalize the results. Results for general purpose use of a-modal
operating systems were published as "gospel" and generalized
to all applications for computers. Some of these results are still
gospel today, such as "A-Modal is better than Modal".
Well in truth that is generally true but it is specifically untrue!
A-Modal
was only better when the User task could not be clearly defined,
or determined or was one of many simultaneous tasks, the order of
which could not be pre-determined. If for example your target user
was someone who was frequently interrupted and had to process several
enquiries at once, across several lines of business, perhaps a stock
trader, then it made sense to provide an interface which allowed
him to switch between tasks easily - an a-modal interface.
On
the other hand, if your User is a telephone operator who processes
insurance quotes on an 8 hour shift then it makes sense to design
a special "insurance quotation system" which operates
only in that mode.
The
question is, "When we really do know what, when, how and why
of the system, do we still need that user to have the biggest screen
affordable?" Interesting when you think about it a bit, isn't
it?
As
Nielsen correctly points out, in the case of a web designer or developer,
a 21" screen with 1600 x 1200 display is a genuine boon. For
other jobs, other users and different applications, this need not
always be the case.
Making
screens bigger and providing supposedly intuitive GUI environments
did improve usability but not by nearly as much as was required.
Meanwhile, increasing the screen size and increasing the complexity
of what was offered on that screen raised a whole new set of problems.
There are still a whole bunch of fans of DOS based word processors
who argue that productivity went down with Windows based alternatives.
A DOS based word processor never needed a 1600x1200 display in 1987.
Why should it need one now?
Fallacy
amplification
Often
in the IT industry much of the nonsense which is talked gets amplified
by others and then built upon until the original nonsense is now
a given fact and everyone has forgotten the history of it. These
myths are often perpetuated by the vendors who have a vested interest
in selling you what they have now. They don't want to hear that
you would rather wait for a better design. And so it was with PCs,
until about 4 years ago when Larry Ellison proclaimed a new era
of Network Computers. Until then PCs we were told would do everything
that we needed. Everyone would have one and it would keep getting
bigger and bigger. We would keep adding more functionality. We would
all have multi-media, speakers, microphones, scanners, bar-code
readers, credit card swipers, video capture, video editing and so
on. The list was endless. All of this made possible by a single
general purpose a-modal device which would need more and more power
and more and more screen real estate. The Network Computer ended
all of that!
Strangely,
Network Computing is seen as a failed technology which never happened.
Well I would argue that it did but we call them PDAs and WAP Cell
Phones, and Information Appliances.
Applying
this lesson to wireless Information Appliances
So
bringing us right back to the 21st century naughties (00's), we
are now in a decade where the promise of wireless, handheld and
ubiquitous computing is rushing over us like a tidal wave. There
is a billion pages of internet content out there and it can be made
available for your use anywhere, anytime if you just buy one of
these great new devices.
Already
we have been reading the words from the prophets of doom that all
this ubiquitous computing stuff will be just great once the screens
get bigger, until then it just isn't going to work. Sound familiar?
Usability will be better when we have real graphic screens; Usability
will be better when we have more screen real estate for more windows
and icons; the web experience will be better when we have bigger
screens with which to browse. That is a brief history of the last
15 years. What is the common thread here?
Bad
Design!
Several
commentators have weighed in on this subject. Mike Banaghan writing
at GBDirect
in April said
"Screens
the size of typical mobile phones are well-nigh unusable and devices
with more real estate will have enough power to run HTML browsers."
Jakob
Nielsen writing at useit.com back
in October 99 said
"Experience from many other user interface platforms indicates
that a bigger screen leads to better usability than a small screen
and that a graphical user interface adds even more usability.
Obviously, in order to be mobile, the device must have a smaller
screen than a desktop computer. A compromise somewhat along the
lines of the Palm Pilot can combine mobility with a decent-sized
screen."
[Notably,
Nielsen has edited this article since it first appeared. The words,
"WAP will fail" have been deleted.]
So
this latest round of angst towards the smaller screen is based on
the principle that the vendor can't adequately predict what you
want to do and when you want to do it, or how you want to use your
device, what you will use it for and when you want to use it.
Anthropology
holds the answer
This
website is becoming popular amongst anthropologists [welcome!].
Why should that be so? Well User modeling and anthropology are essentially
the same thing but coming at the problem from two different angles.
User and Task modeling is becoming a popular day job for anthropologists
[ Americans have a tendency also to call this ethnography, I do
not believe that the use of the term is accurate in this case, but
I am sure that some ethnographer will write in to correct me ].
Anthropology
is by and large, the study of how people live. If you study how
people live then you can better understand how technology can help
them and what effect it will have on their lives.
In
order to design effective small wireless devices, you have to understand
people. The devices are by their very nature Modal devices. So you
are stuck with a modal design. In order to solve the puzzle you
have to figure out which modes will best suit your Users. Screen
size has nothing to do with it.
Screen
size versus utility
Jakob
Nielsen also states in the "WAP will fail" article, that
the answer probably lies in a compromise device. A Palm form factor
with a reasonably large screen, an a-modal interface and a wireless
communication device (cell phone) built in to it.
Or
to put it another way, the answer is a small general purpose device
- a small handheld PC! Microsoft will just love it if that happens.
What
is the alternative?
A
heterogeneous set of custom designed modal devices, designed using
anthropology techniques to better fit your lifestyle. What do I
mean by that?
Well
for a start, how about an in-car system which uses voice browsing
and talks to you, rather than a small screen and a touch keyboard
which you have to fumble with as you drive head down?
Or
how about the Information Appliance equivalent of a Sony Walkman,
which reads the news headlines to you as jog along and allows you
to surf through information by voice control?
The
reality is that a Palm form factor device is a two handed device
which requires head down attention. You hold it in one hand and
you wand it with the pencil with your other hand. That has its uses
but it also has its problems
A
WAP Phone, on the other hand, is a one handed device. Yes it has
a smaller screen, which is necessary in order for it to be a one
handed device. The challenge for WAP designers is to figure out
the anthropology and determine when you need a one handed device,
what you might be doing which prevents you from using your other
hand, and then designing a system which delivers the correct modal
offerings, appropriate for the anthropology in the study.
Human
Factors
There
is a place in the world for devices which require, no hands; some
which require one hand; two hands; two hands using all ten fingers;
and perhaps even beyond that to feet, arms, heads and so forth.
Everything man has ever invented before computers works for modalities
outwith the basic two hands, ten fingers. Why should computers be
any different? Would you rather brake your car with your right foot
or by moving a mouse with your right hand and then pressing the
"brake" button?
The
point here is that a heterogenous set of modal devices used appropriately
and designed for how people live will always be better than a single
general purpose device. Afterall, you can't take the home entertainment
center with you when you walk the dog. But you could take a Walkman
or a suitable designed one-handed information appliance.
With
the advent of ubiquitous computing, the need for Human Factors in
Information Appliance design has never been greater. Would you rather
that your Information Appliance vendor was using anthropology to
treat the cause, or merely a big screen to treat the symptoms of
bad design?
The
need for a big screen is simply a poor excuse for bad design.
[
Postscript May 23rd, 2000 : Since publishing this article, Business
Week (May 22) has published an
article featuring Toshiro Iizuka of Panasonic, who is responsible
for Product Design of many forthcoming wireless internet devices.
Iizuka firmly believes that devices must vary in form factor to
fit the lifestyle and activities of the user. Suggestions such as
a glove controled wireless internet device and other devices suitable
for in-line skaters and joggers are just some of the ideas being
floated by Panasonic for next generation wireless internet access.
Japanese
companies generally, Sony, Sharp and others seem to share the view
held by Panasonic. This varies significantly from the American view
of the future which revolves around the notion of a multi-purpose
device with a 4" (or larger) screen in a Palm Form Factor running
Palm OS or PocketPC/MiniME.
During
the Industrial Age the Japanese took the market from Americans,
who had perviously invented it, in industries such as automobiles
and consumer electronics. It will be interesting to see whether
the Japanese can once again steal an industry, the wireless internet,
with a different vision of the user experience.
In
the meantime, this webzine remains firmly in the Japanese camp.
We remain of the belief that a heterogenous approach to wireless
internet design will lead to a better overall user experience. ]
|
