|
Last
month, I wrote that it just didn't make sense that Web Content providers
were having to provide pre-formatted WML
rather than XML for Wireless content, rather than just providing
data. A data driven approach which allowed the Wireless Carrier
to format the data to their own taste made a lot more sense. However,
the reality is that WML (or HDML) Content is out there and it will
be used. This month I question whether it spells the death of User
Centered Design, forcing the Content Aggregator to return to the
bad old days of Function Oriented Design, or whether XML might be
the saviour.
So
WAP Phones are coming to North America (HDML phones are already
here) and the bigger carriers are scrambling to provide Content.
At this point they just need something, indeed anything, to show
that you can access the internet from your phone.
It's
the Content Stupid
"It's
the Content Stupid", is the cry. Forget the User! Just get
some Content to market. That's partly the driving factor behind
the current Wireless Content Supplier model, first seen last month.
Diagram1.
Internet Content Providers doing the WML, Content Aggregator is
just a portal and conduit to the client wireless device
In
the current model, the Content Providers such as Amazon.com,
Weather Channel, CNN,
Reuters, etc. have to do the
hard work and configure output in WML format. The portal is really
only providing a set of menus and perhaps some searching and filtering
of Content. Continuing to supply Content in this fashion reinforces
that current model as the Aggregators / Portals / Carriers get lazy,
making it a hard model to break later. After all, what is in it
for the Carrier to change?
It's
the User Stupid!
Well
the real driver for change ought to be the User. The Carriers ought
to come round to the notion that the way they supply "compelling"
Content and gain market advantage is to tailor the Content to the
User's needs. Do I hear you say, "User Centered Design"?
Well frankly, we're a long way from that now and I wonder if it
will ever be possible to recover the situation.
The
Bad Old Days of Functional Design
Once
upon a time, there was the PC and there were fledgling software
companies that knew just a little of how to make a PC do some stuff.
They could Open a File, maybe Edit some Text, maybe Cut Text, Paste
Text, and later it might even let you Save your File or Print your
File.
Early
Applications were functional and Function Oriented. The whole concept
of the Menu Bar driven interface was a Function Oriented Design.
The Application Developer provided an ever growing number of functions
and then exposed those functions through a Menu Bar Interface. In
order to perform any task, the User was required to learn which
Functions were required and in which Order to perform them. This
technology centric approach was hardly User focused at all.
User
Centered Design
As
time passed and the PC software industry evolved, some of the higher
forms of life began to use User Centered Design. The precise approach
varied from life form to life form but by and large two approaches
persisted: the Task Centric Approach; and the Goal Centric Approach.
With
Task Centric Approach, you spend time to understand the User base
and the types of User. You learn the tasks that the User needs to
perform and then you design the software to make these tasks easier
to perform. This could be thought of as a Procedural Design. The
interface provides the ability to follow procedures, so that the
User doesn't have to remember all of the functions and in which
order they need to be performed. Specific Interface Controls emerged
to make this approach easy to implement. Controls such as Wizards
which store the procedure of functions and prompt the User for the
correct functions in order. This was, by and large, considered an
improvement in usability.
So
where are we at in the Wireless world?
Well
at first glance we are already past the Functional Design stage.
But are we? For example, if I hit the correct URL, I can look up
the weather in my city. The URL will deliver some deck of cards
to my phone (a deck of cards is like several small pages of HTML,
you can think of it as the WML equivalent of a Wizard). Perhaps
the first page will prompt me for the name of the city. Later it
will deliver results, perhaps even 5 days ahead allowing me to flip
one day to the next through several cards in a deck. Let's call
it the Weather Wizard.
So
we have a Weather Procedure. Great! Due to the highly modal, conversational
format of WML, this is the only reasonable way that the Content
Provider can package up the material for delivery. However, what
they have delivered is effectively a low level task, interaction
flow, or Use Case, for a very simple function: Tell me the Weather
in a given city. Our Weather Wizard is a Function - not a Task!
Low
Level Tasks are not Compelling Content
So,
it can tell me the weather. Perhaps another source can tell me the
road conditions. And yet another source can tell me the snow conditions
at given resorts in Colorado. Hmmm. Let's see now - I sense a Persona
Definition coming on.
David
is a 21 year old college under-grad in Denver. He and his buddies
are thinking of going skiing on the weekend. Between the four
of them they can muster: a Jeep Cherokee; 3 pairs of skis &
boots; enough food and drink for the trip to the mountains and
a few hundred dollars.
The
Goal is to go skiing for the weekend. Not to sit in a snow drift.
Not to find the resort is full, or empty due to no snow and (oh
yes) they would prefer to stay in a motel rather than sleeping
in the car.
The
task is to achieve the goal with the minimum of cognitive friction
(that's hassle in plain English).
So,
if we're going to do this at WAP speed, we want our carrier to do
the hard work and provide us with a convenient way of achieving
our goal.
First
we need the snow reports for the resorts. We need to pick a resort
and we need it to tell us the road conditions between Denver and
the resort. We need to know whether the roads will still be open
in the smaller hours of Saturday morning. We also need to know about
Motels and we need a ski and boot rental store which is on our route
and open when we are passing it by.
To
deliver this User Centered Goal, we need a fully integrated User
Centered Design. To do that we need to do Usage Scenarios and turn
them into high level, Essential Use Cases (or an equivalent approach).
It's a top down design. User Centered Systems require Top down design.
Now rewind.
Back
at the Content Provider we can source Function Oriented low level
Use Cases for accessing simple functional Content. That's bottom
up design.
Fundamental
mismatch!
Getting
to User Centered Design
In
the old days, it was easy. You just took your old functional design
application, threw out the design but kept the functions. You went
and spoke to the Users and found out what they do. You listened.
You wrote it down. You went back to the office and you produced
a Navigation Model which represented those functions you coded in
a new Task Centered approach. Easy. Well, yes it was because you
had control of the source code.
Now
put yourself in the shoes of the Content Aggregator or Wireless
Carrier. Let's look again at the Going Skiing for the Weekend Goal.
It requires Content from 5 different providers and the Aggregator
doesn't control any of the source code.
Delivering
User Centered Design from aggregated content will be almost impossible
unless the current Presentation Oriented delivery mechanism of WML
is broken and the focus returns to Data Oriented delivery - XML.
And yes, the Carriers and other Aggregators will have to do the
hard work.
And
why should anyone bother?
Well,
it seems to me that there is one company and only one which controls
enough Content and has control of enough WML source code, that it
can deliver Compelling User Centered, Goal Oriented solutions. That
company is America On Line. Through
its Time Warner and EMI subsidiaries AOL has the Content to provide
compelling integrated solutions.
If
you are not AOL, XML is the only way you are going to get
control of that Content so that you can integrate it properly to
deliver compelling User Centered Content.
|
