 |
 |
Terry (and Gunnar) Thanks for the constructive dialog. My feelings are pretty straightforward. I think that the work that MK French was doing is exemplary and is indeed what the best engineering designers practice. it is an open question whether the formalism (even the simple math) used by French will work for designs with a heavy component of human interaction. Certainly today, except in rare cases, we cannot yet do that. (Mind you, my friends Newell, Moran, and Card moved us forward a tremendous amount, continued by many other people, such as Bonnie John.) I think the most powerful and applicable part of French's philosophy is the move to the proper level of abstraction: move up, but not too high -- stop where there are concrete actions to be considered.) This philosophy is what I have long done and what has been taught at the Institute of Design/IIT in Chicago for many years. (Probably at many other places, it is just that I have heard several faculty there -- especially Patrick Whitney -- espousing this.) Mathematics. Right now I have two major projects just underway. One is to look at automation in automobiles, in particular the communication between automated automobiles and people. (If you want to cross the street, how do you know the automated car has seen you? Yeah, I've seen the Mercedes commercial but that is not very practical. What if there are 6 people crossing from different starting points -- the Mercedes method fails. (Note that in the commercial they project a walkway in front of the car, but they show this in dusk or evening. Would this work in bright sunlight? Nope. And if you were the last person in a crowd crossing in front of the car, how would you know it had seen you -- or if you were coming from the other side of the street? The other project is in healthcare, with a zillion unexpected occurrences, interruptions, and many different specialties who all must interact. How would French deal with these? I do not know how any mathematical formulation would lead me to appropriate solutions. But I can imagine a number of creative designers who might very well suggest directions to try. This, by the way, is NOT a criticism. It is a statement about today's state of the art. --- The clamshell phone. This one could be done through more formal methods, simple abstractions, and simple maths. In fact, most engineering designers would treat this as a simple homework problem. Engineering design by the use of matrices of desirable characteristics crossed with potential solution spaces. Even so, this requires adding weights to the desirability and importance of the many factors, and these are subjective. How do you weight the nuisance of those phanthom phone calls (when your phone gets bumped and so it calls some pre-programmed number without your awareness, so the recipient can hear your breathing your conversation with others, etc. It is usually truly annoying to the recipient who can't turn it off -- except when it is truly intersting, which probably means incredibly embarrassing to the phone's owner.) My experience is that engineers pretend to great precision, but when the answer is not what they were looking for, they go back and change the subjective weights until the answer satisfies them. Hmm. It is Terry's personal view that the clamshell cellphone kept going far too long. For him, I am certain that view is accurate, but not for me. There are times i wish I had a clamshell phone to protect me from those false phantom calls. When we have differing personal judgements which play out in differing weighting functions, all the mathematics in the world will not help. I've always argued that the proper answer is to make a wide variety of devices available -- select the one you want. This is how we make hand tools. Go to a good hardware store and examine the umpteen different kinds of hammers, screwdrivers, cooking utensils. Same with watches. As many in this list will attest, people have strong preferences for writing and drawing instruments: many kinds of pencils, pens, brushes, ... . Why is it that phones and computers tend to follow the same design. (I give Microsoft great credit for pushing their novel tablet/keyboard design. Very clever engineering and thought went into that choice. Some day the software will match the cleverness of the hardware (some of my friends tell me that the latest releases do satisfy.) Note that these have nothing to do with design methods, it has to do with marketing and market forces. And I suspect French would have thrown up his hands in disgust over domination of design by marketing and market forces. But hey, that's life. People do not always buy the "best" product. I bet most people here don't either. We but the one that fits our unique needs, desires, and preferences. Is there a formal science of these aspects? There is a nascent science, one just developing, still weak -- and often quite wrong. But just as in design, I think that science will help, the proper abstractions will help, but they cannot do it all. Let me conclude on a more positive note. I respect the work of people who try to bring more prevision and rationality to much of what we do. It will avoid some of the horror stories we now face in the horrible Electronic health records, or truly dangerous controls for complex machinery and process-control plants. But for situations where there are complex human interactions, we do not have the tools today to use a formal system, nor mathematics. And in many such situations, I suspect we never will. (Technical comment: I include computer simulations, AI, deep learning, and other techniques as part of what I call mathematics, even though technically they are not.) Thanks terry. This is an instructive discussion. Don > > On the clamshell, phone, I suggest the clamshell paradigm kept going > longer than it perhaps should have because their wasn't an easy algorithm > for use by phone designers that showed the benefits and limitations of > different configurations in human terms. French's approach to using > abstractions could have provided that kind of information > Don Norman Director, DesignLab, UC San Diego [log in to unmask] designlab.ucsd.edu/ www.jnd.org <http://www.jnd.org/> ----------------------------------------------------------------- PhD-Design mailing list <[log in to unmask]> Discussion of PhD studies and related research in Design Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design -----------------------------------------------------------------