On Nov 2, 2007, at 10:41 AM, Joseph Selbie wrote:
> > I would love to see standardization also, but I don't expect it soon. > Architecture has had 5 millennia to evolve some standards, software > has had > 5 decades :).
Agreed that we won't see it soon, but I don't agree that I would ever
love to see it.
As I've argued many times before (http://tinyurl.com/35gqm8),
standardization of design documentation in our business is in many
ways a sure-fire recipe for stagnation, especially if the products or
sites you are designing are intended to do wholly new things, or if
you intend to build a UI convention that improves on the product or
the industry's status quo.
Sure, standardization is good for "we don't need to re-invent the
wheel" products. But do such products really exist? How many products
do you work on where there is no new UI thinking required? If your
design challenge requires any innovation at all -- and in our
industry a great many design challenges do -- then we must constantly
innovate our design documentation, too.
(As an aside: The field of architecture isn't as wholly standardized
as we might think. Architects often devise new ways of showing and
explaining their ideas. From collage to painting and drawing, 3D
rendering software and animated flythroughs, physical models,
conceptual sculptures, full-scale room mock-ups, films and audio
renderings of the experience, and much more. The concept work and
schematic drawings by great architects are often prized as works of
art because they show new ways of perceiving structure and experience.)
If we in the UXD field had, say, strictly standardized our
wireframing process in 2000, we'd never have page-less Ajax
interfaces today. I even suspect that many sites that can and should
be using Ajax today still can't manage to upgrade their UI simply
because the information architects working on the site are hampered
by their primitive -- but standardized! -- IA toolsets. They cannot,
for example, document a registration form that displays the validity
of your new username as you type because the standardized wireframe
template requires all server interactions to be documented as a new
page/view -- their wireframe templates are page-based, so as a result
their sites are, too. When I hear someone ask "how do I document an
Ajax interaction", my answer is "just make it up".
What's more, I do not agree with the idea (posited by some here) that
engineers cannot understand documentation unless it is somehow
standardized. They're not robots -- we don't need to make our
documents machine-readable. The documents simply have to be good,
detailed, and responsive to the customer's (the engineer's) needs. If
every design document you make looks completely different from the
last, uses a whole new design vocabulary, this need never be a
problem as long as you work closely with the engineers to explain
your objectives and respond to their needs and questions. If
anything, a non-standard document is one that can be a great way to
bring everyone to the table, designers and engineers, to discuss a
Reinventing the wheel is a waste of time? If you look at any industry
that actually makes wheels, they are essentially reinventing them all
the time. From carbon-fiber bicycle disk wheels to the Segway's
computer-controlled single-axle balancing act, assumptions about the
wheel are always being challenged.
Just say no to design document standardization. Reinvent the wheel.