User-Driven Applications—New Paradigm for Interface Design 381
ties. Microsoft decided to save those users (and develop-
ers!) from the hard labour of thinking and began to pro-
mote the dynamic layout. In 2006 Charles Petzold wrote
in his book [2] that “dynamic layout is…an important
part of future Windows user-interface design philosophy”.
Petzold is not only a very good author but he knows very
well, what is going to be cooked at the Microsoft’s
kitchen. That statement was definitely not the Petzold’s
speculation on the item but the Microsoft’s decision. An
extremely wrong one, but when Microsoft invests a lot of
money into something, there is usually some result. At
least, temporarily but not necessarily for best.
The idea of dynamic layout is simple: user changes the
outer sizes of the form (window) and, as a reaction to this
single command, the program changes the sizes of the
inner elements according to the algorithm predefined and
fixed by developer. The result is obvious: there are no
more choices for users; everything is predetermined by
developer. The interface design made the full loop and
now we are back in the situation that we had 40 years
ago: everything is again controlled by developer. What is
really funny is the fact that “big specialists” try to ex-
plain even this as a huge achievement. Two years ago I
got such an explanation in a private letter from one of the
MIT professors: “Without dynamic layout, the end user
would have to manually, one by one, resize and reposi-
tion the elements inside. So dynamic layout does confer
usability benefits b y making the user interface more effi-
cient: one resize action by the user results in many auto-
matic resizes and repositions of dependent objects.” In
the same way the proponents of slavery can declare that
slavery is the best form of social organization because
slaves don’t need to think about food or shelter; slaves
are provided with both and can focus entirely on their
work. This is not the place to discuss the moral aspects of
slavery; I only want to remind that slavery died of the
economical issues. As a result of its economical ineffi-
ciency, it simply lost the competition.
Dynamic layout is the dominant idea of interface de-
sign throughout the last years. Well, it is n ot the first time
in the history of science when the wrong idea becomes
dominant. Certainly, not forever. One of my friends and
former colleagues (an excellent physicist, head of the De-
partment of Mathematical Modelling) perfectly described
the situation: “You can fool one person as long as you
want, you can foo l all people once, but you can’t fool all
and forever”. The dynamic layout is used nearly every-
where now, but it doesn’t mean that it is a good idea.
Dynamic layout is a well constructed highway into the
dead end.
If so, and I am sure about it, then where the wrong turn
was made and where we have to return back to take an-
other road? And what is that other road can be?
3. The Idea of the Different Way
To catch the main idea of another way on which the
whole interface design can be based, let us consider the
situation from abso lutely differen t area. Supp ose that you
have rented for some time a house with the furniture,
fully equipped kitchen, and all other needed things. The
owner tried to do the best and before your arrival put
everything in such an order that, from his point of view,
would be the best for you. On your arrival, you can be
satisfied with all you see and keep everything in the same
order, but chances are high that you will move some
pieces of furniture and other things around the house
according to your own preferences. And throughout your
stay at the house you will continue to move the things
around whenever you feel any need for it. You want to
feel comfortable at any moment and this can be easily
achieved by moving things around. Those movements
can be caused by many different things: rainy or sunny
day outside and your desire to have more or less light, a
home party will require more free space while a private
conversation will need two armcha irs close to each other,
a dog will need some special corner, and so on. When
you leave the house, another tenant will arrive and will
move the things around the house according to his pref-
erences. The same circumstances will result in similar
actions by different tenants; similar but not identical be-
cause everyone has his own taste and estimation of com-
fort. A complicated problem of organizing a comfortable
living for every tenant is solved easily enough and does
not require a special course: everyone knows that all
things are movable and can do it himself. As much as he
wants and whene ver he feels any need fo r it. The solution
to important problem of organizing a comfortable living
is the movability of each and all pieces. The movability
without restrictions and without involvement of anyone
else.
In the same way the movability of the screen objects
can help us in our problem with interface design, but
before announcing any new steps, let us analyse the use
of movability of the screen objects throughout the past
years. Maybe this was already done before?
Let us return back for nearly 30 years. The first com-
mercially successful product to use a multi-pan el window
GUI was the Macintosh, released in 1984; next year the
Windows system was released. Throughout the years
there were different variations, but for the last 20 years
we—users of computers—deal with the two-level system.
At the first level—the level of operating system—we
have only icons and rectangular areas representing the
working applications. Both types of objects and the entire
mechanism of working with them are developed by the
authors of the operating system. As the result, from the
very first versions of those operating systems both types
Copyright © 2013 SciRes. JSEA