Thursday, February 12, 2009

Is Good Design Replicable?

Is there a process that will reproduce good results? ....

"Here’s a question: Is there a process that guarantees good design?

I ask because many people seem to think process is the key to good design. For example, on almost every thread of the IxDA mailing list, there is an argument about which design methods are better…are personas better or is genius design better? What method or process should we be doing to get the best possible product? The implicit assumption is that if you perform some particular UX method then you’ll produce consistently better design: the right process = the right product.

So, the obvious question to ask is: Is there evidence that someone following a certain process produces great design every time?

The only company/group of people I can think of is maybe Apple, who has had a string of three amazing product lines: the iMac, the iPod, and the iPhone. These three things have catapulted Apple far beyond almost every other design company out there. If there is any company who can consistently produce great design, it’s Apple. However, Apple has also produced suspect things like the Apple TV, the Cube, iMovie ‘08, and iCal.

The problem with Apple, of course, is that their design process is completely shrouded in mystery. They may be using different design processes on each project, or on a project-by-project basis. Very few people know what goes on in Cupertino, so fantastic stories of “not doing research” or “genius design” abound. My guess is that the folks at Apple are brilliant, but that their success is achieved in the same way that most success is: through hard work.

Flipping through most portfolios is an exercise in good/bad. Take a look at Pentagram’s portfolio, for example. Some projects are good, some are less so (the XO Laptop user interface Sugar, for example, was widely criticized). Most designers I know are really proud of a small number of projects, and quite ambivalent about the rest. What if design is always hit or miss? Would the world end?

I wonder if the real issue is that most of the time designers simply don’t know if what they’re building is great, and they end up relying on process to get as far as they can. If they go through the right process, they think, then they’ll produce maybe not the best solution, but the best solution possible. This may be true…and it is comforting, in a way, because if you feel like you are doing it right then you can sleep soundly."    (Continued via Bokardo)    [Usability Resources]


Blogger mrfrank said...

I know that there is a connection between design and usability but it also seems that some designs succeed for merely random reasons.

That being said, I believe genetic algorithms will be used to optimize our designs in the future - as this level of computing power becomes available to individuals.

You can see an example of genetic algorithms being used in the design of a solar panel here [ Ted Talks / Bill Gross: "Great ideas for finding new energy" ]

2:48 PM  

Post a Comment

<< Home

<< Home