This is a comment from industry. Sketching is very useful and fast - ideal where something needs to be loosely defined. It is a fabulous tool for getting early agreement - and as Ideo suggest; "If an idea cannot be sketched, it probably doesn't exist". Clients sometimes prefer not to have something in concrete before they sign off on something. However, if one is considering a more complex surface design of say, an aircraft sidewall, it can sometimes be faster to work in a dedicated CAID tool - such as Alias, Rhino, Shapestudio or Think Design. For example - if we are dealing with the intricacies of a complex form - such as the light reflectance the window panel possesses. Unless one is of the Syd Mead level of illustration and can depict light very accurately - then a sketch is about useless. The sketch can provide little more than a simple, inaccurate goal of the form. Another time that sketching is useless, is when a designer has run out of ideas. Sometimes it is then better to go and make something. This process puts back in place the train of thought emptied by the idea notation (sketching) process. Added to this, new tools such as Wacom's Cintiq, have enabled a real blurring of the line between sketching and traditional CAID tools. Finally, the latest tools from companies such as Think and UG create surfaces defined from parametric curves. In other words, once a design has been created it can be continuously modified aesthetically. The surface model then becomes an equivalent of a thousand sketches as it were. Sketching captures an idea, but I would caution that sketching as a main idea representation is being overtaken. glenn johnson