It seems like not too long ago, many IA/UX designers fought endless battles on mailing lists and Usenet about whether Visio was better than Freehand which was better than Omnigraffle which was better than Excel (no, really, I’ve seen people use Excel to express UI ideas). There was always some software or other that totally rocked while some other tool sucked. Almost as boring and futile as the OS wars. Perhaps I just learnt to ignore it all. But if I remember correctly, didn’t we all reluctantly agree that when it comes to getting to the best execution of an idea, it’s what you do, not how you do it, that counts?
Perhaps not, as there seems to be an increasingly vocal band of people who want to make a point about how wonderful the act of “sketching” on paper is. Moreover, that some people see this as an issue of “sketchists” vs non-sketchists allows me to see this in similar terms to the aforementioned tool wars. There is certainly nothing wrong with a quick scribble to crystallize your thoughts or to demonstrate an idea to somebody. I would also broadly agree with Jason Mesut here (although isn’t it stating the obvious?). But the further you go in this, the less clear the benefits of sketching become.
From what I can tell, those who would eschew all digital modes of expression in the early stages of design say that sketching with pen or pencil (and at this point we’re probably talking about drawing) has the overridingly positive quality of allowing the rapid communication of different ideas to stakeholders. Not only that, but because sketches don’t look final, stakeholders are more likely to feed back on the high-level direction of the idea, rather than get hung up on minor detail.
One observation on these points is that the half life of this value is quite short. Because sketched ideas are necessarily vague and open to interpretation compared to more “pixel precise” media, a lot depends on the subsequent visual design to execute the ideas in the sketches effectively.
But this much I think is not contentious: a good interface or user experience idea is a good idea whether it’s expressed by drawing it with a stick on a cave wall, or with Adobe Illustrator 6. And exactly the same is true of a bad idea. It is also true that you cannot know for sure whether something will be a good idea until you actually build it for real and it’s being used by the target audience. Even a high-fidelity prototype might not allow you to know (for an example, take the Folding Plug – if it wasn’t for those pesky British Standards…).
It is also probably true that any idea of sufficient complexity needs to be thought through from the general to the particular. In the case of UX design, from the low to the high-fidelity expression. Moving from a vague sketch to the real thing implies quite a lot of risk. Why would you not seek the safety of quick HTML prototype mid way? There are similarly vocal fans of very early prototyping as well, I might add.
Ultimately, I think it’s going to depend on the person doing the design, how they think and what kind of project they are on. It will also depend on the people they are communicating to, and the environment they are all in. The issues here seem well summarised by this guy on the above Smashing Magazine article (and note the high number in agreement!). See also his thoughts on teamwork when it comes to this.