Dear Lubomir,

Thanks for an elegant note. I hope it is clear that I amended my earlier views on the potential for this software as a supplement to design practice, in what you label peripheral activities. I have not changed my views on the fact that software cannot replace human designers in core design activities.

Ranjan effectively said that same thing where he wrote, "Designers should watch this space not because they fear being replaced but for amazing new possibilities for research into human behaviour and pattern recognition that is best left to automated tools when tons of data are to be mined."
 
Ranjan’s earlier post suggested that we ought to be alert to the uses of the software in peripheral activities, and I agree. As I wrote, it remains to be seen how these can be put to use in practice and research. If this turns out to be useful, this may have implications for education, but not before.

As I wrote earlier and repeated, this is not an automated image rhetoric system. This software cannot replace human beings, but it may offer benefits in peripheral activities to support narrative and story-telling.

If Ray Kurzweil is right in the world after what he calls “the singularity,” things may be very different. We have not reached that state yet. Most conversation about computers replacing human beings is technological romanticism. In a possible world after the singularity, computers will be able to write our posts to the lists for us — and the rest of us can have our computers read them.

Until then, you are quite right. My amendment only involved the specific range of functions that Ranjan described.   

Best regards,

Ken

Ken Friedman, PhD, DSc (hc), FDRS | Editor-in-Chief | 设计 She Ji. The Journal of Design, Economics, and Innovation | Published by Elsevier in Cooperation with Tongji University Press | Launching in 2015

Chair Professor of Design Innovation Studies | College of Design and Innovation | Tongji University | Shanghai, China ||| University Distinguished Professor | Centre for Design Innovation | Swinburne University of Technology | Melbourne, Australia

Email [log in to unmask] | Academia http://swinburne.academia.edu/KenFriedman | D&I http://tjdi.tongji.edu.cn 

—

Lubomir Popov wrote:

—snip—

I am still on that side of the discussants who think that design software is impossible. I will clarify what I mean. 
 
My observations come from the field of architectural programming and design and are limited to that area. 
 
First, let's see a model of design process that consists of core activities and peripheral activities. Peripheral activities are routine and pertain to technicalities. The core activities are about developing the design concept, etc. This concept is everything for me. The rest is construction/production documents. One of my dissatisfactions with architecture is that the creative time is 1% of the work time. The rest is drafting, construction documents and running around. I know some architects are proud with their mastery in production documents, in the carpentry details, but I evidently am not of this kind. I get bored. So, I differentiate between a core design process and project delivery process. The project delivery process incorporates design and core design activities, but it is much broader and involves a myriad of supplementary activities. The clients pay for the product of the project delivery process.  
 
The first time I heard about design software, it was 30 years ago and the software was called AutoCAD. Well, we all know that CAD is a drafting software, drafting 2D or 3D models. The computer can make 3D models that no human can make. So much about design.  
 
Then, some 20 years ago, someone told me about facilities programming software. There were/are several types of these. One of the software types is basically a spreadsheet where programmers enter information. The software just adds numbers. In the best case scenario, the software makes block diagrams that cannot be fully trusted. Designers need to inspect them, verify them, and select only the option that makes sense. There are a number of options that are plain wrong. The software can also model patterns of human locomotion in a furnished room (surgical suite) in relation to the placement of furniture and equipment. The software calculates the travel time of the operators in the room. Certain furniture configurations can save one full-time equivalent (FTE) staff position. Although this is valuable help, it is not core design. It is still in the realm of the peripheral activities. The movement calculations are based on a behavior model. If the model is wrong, the calculations are wrong. That is it. As I mentioned before, the problem is not in the math; the problem is with the social science models. I can tell you that, this time with my hat of a sociologist. There is software that calculates room adjacencies based on the data that are entered. This is still a peripheral work. 
 
I haven't seen software that is capable to make major decisions. Even if somewhat has developed such software, it will be a spreadsheet where we enter information based on our theoretical models and research findings. If the social models are wrong, everything is wrong. So much for math and software. 
 
Bottom line. Computers may help with technicalities. They can actually do some technical work much better than the humans. Materials and products software works faster and with less errors than the humans. Now they call it BIM software, building information management. But this is not the core of design. This is about drafting and spec writing.  
 
I can imagine that researchers make algorithms based on successful design solutions. Such software might be capable to put together a new successful design, similar to the way lay people work with the design patterns of Christopher Alexander. But the software will always be one huge step behind of the human designer. Such software is in fact a computerized expert system that can produce solutions based on the algorithms entered in it. This is the best case scenario. You can design your house. Right now there are various kinds of "design" software for laymen, sold by retails of construction materials, products, and equipment. Such software is about $40-$50. So much about design. This is something different. This is like entering your information in a virtual medical doctor software and getting your diagnosis and treatment prescriptions. It might work in 95% of the cases. But the remaining 5% might end in the morgue. Would you call this a computer medical doctor? By the way, a friend of mine treated himself his headaches for years on the assumption he has a sinus trouble. The treatment was successful for five years. There was no doubt that the sinuses make the trouble. Until one day he had a major problem,  MRI, and then, there was a discovery that during all these years, a different condition has grown slowly. So much about the virtual medical doctors. They are simple expert knowledge systems that are set on a computer rather than a 1,000 page handbook. They are more interactive and create the feeling of actual interaction and decision making. In effect, they are sophisticated books.  

—snip—


-----------------------------------------------------------------
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
-----------------------------------------------------------------