Dear Ken,
Thank you for your thoughts.
I suggest a better reasoning is different and includes at least five other factors:
1. In life, design activity is a recursive layered activity that includes designing the tools that are used within it, and these have implications for design quality if you assume at least some of the tools can kead to better design outcomes than human designers achieve without them;
2. That it is important to avoid the biases of seeing and assessing the quality of design activity purely through our opinions as humans, i.e avoiding an obsessively human/person-eyed view;
3. Realising that human beings are terribly incompetent at design, and that this relatively huge lack of competence has unhelpfully shaped our language and ways of thinking about design activity in order to avoid facing up to this fact (or rather normalising it in ways we find personally comfortable). Don's compromise position is a small example of this.
4. A widespread and common lack of awareness and basic knowledge (especially in academics) of the everyday range and limitations of human skills.
5. Evidence from real world design processes outside academically-defined design activities
This leads to a different sort of reasoning about the limits of the higher aspirations of design research and design practices.
And that's my second post.
And, in my previous post, what I wrote was different from what you inferred...
Best wishes,
Terry
--
Dr Terence Love
PhD (UWA), B.A. (Hons) Engin, PGCE. FDRS, PMACM, MISI
Love Services Pty Ltd
PO Box 226, Quinns Rocks Western Australia 6030
Tel: +61 (0)4 3497 5848
Fax:+61 (0)8 9305 7629
[log in to unmask]
--
Warm regards,
Terry
-----Original Message-----
From: [log in to unmask] [mailto:[log in to unmask]] On Behalf Of Ken Friedman
Sent: Sunday, 6 September 2015 8:26 PM
To: PhD-Design
Subject: [SPAM] Re: How can we build technology so that people plus technology is better than either alone?
Dear Terry,
Your recent post offered a powerful logical assertion that it should be possible to demonstrated or to test by logical proof. I want to ask a serious logical question about this assertion.
The background to this assertion was André Neves’s question, “can machines design?” Lubomir Popov replied to André and Don Norman replied to you with comments suggesting that they share a common position. Lubomir states that computers can support decision-making and design. Don states that humans and computers working together can do better in many ways than either working alone; therefore, we should find ways to augment human capacities and skills with computers.
My view is that both these positions are correct. If we restrict our definition of design to computable functions, then machines can design. If we consider the many issues that we must consider in skilled or advanced design, this kind of design is not necessarily skilled or effective. Don phrased it this way: "Can machines design? Of course, I say. Can machines design in a way that inspires us, excites us, and makes us envious. No, I say.”
Rather than limiting the question to design that inspires us or excites us, I am suggesting that this may even apply to reasonable, responsive professional design. In my view, the skilled professional design process requires the designer to choose and specify a preferred state, either directly or together with the problem-owner on whose behalf the designer works. Second, an effective professional design process requires that the designer should work with and understand the needs of legitimate stakeholders — customer, clients, end-users and perhaps others. So far, the evidence suggests that machines can meet these requirements as part of a larger system involving human beings, but they cannot do so based on current programmable capacities.
Like you, I would like to see greater rigour and clarity in the design process. This includes conceptual, logical, and philosophical rigour in the way we think about design, as well as experimental and mathematical rigour in developing, working with, and testing design.
You answered Don by asserting that “if you think through the logic of getting the best design out of humans and designing systems together, in the limit you finish up at my question.” Your question is: "How can we design automated design systems to design BETTER than humans can?”
Like all serious research questions, this question involves assumptions, pre-suppositions, and entailments that attend the varying definitions and meanings of the word design. The answer to the question depends on which of those assumptions pre-suppositions and entailments apply to your use of the word “design.”
You say that the logic of the question in the header of this threads leads to your question. This is a way of saying that [Premise A] “How can we build technology so that people plus technology is better than either alone?” reduces through a logical syllogism to [Conclusion] “How can we design automated design systems to design BETTER than humans can?”
There must be one or more intermediate logical steps that bring us from the first question to the second. These steps must contain terms or premises required for a valid syllogism.
Turning these questions into positive statements for a syllogism involve of saying that the logic of [Premise A] “We should build technology so that people plus technology are better than either alone” reduces through a logical syllogism to [Conclusion] “We should design automated design systems to design BETTER than humans can.”
As I see it, the syllogism requires one of at least two sets of logical claims. One set of claims would demonstrate that [Premise B] automated design systems CAN choose and specify a preferred state, either directly or together with the problem-owner on whose behalf the designer works, and [Premise C] automated design systems CAN work with and understand the needs of legitimate stakeholders — customer, clients, end-users and perhaps others, leads to the [Conclusion] “We should design automated design systems to design BETTER than humans can.” There are still missing terms, but these would appear to be necessary.
Another syllogism would state, [Premise B] automated design systems CANNOT choose and specify a preferred state, either directly or together with the problem-owner on whose behalf the designer works and it does not matter that they cannot do so, and [Premise C] automated design systems CANNOT work with and understand the needs of legitimate stakeholders — customer, clients, end-users and perhaps others and it does not matter that they cannot do so. These premises, together with missing terms, lead to the [Conclusion] “We should design automated design systems to design BETTER than humans can.”
My question to you is to ask which of these syllogisms is correct or closest to correct. What terms are missing?
If neither is correct nor close to correct, then I ask you to state the logically necessary syllogism that leads from [Premise A] “We should build technology so that people plus technology are better than either alone” to the [Conclusion] “We should design automated design systems to design BETTER than humans can.”
If this is for some reason unsatisfactory, then I ask you to demonstrate the logical necessity that leads from the question [A] “How can we build technology so that people plus technology is better than either alone?” to the question [B] “How can we design automated design systems to design BETTER than humans can?”
Yours,
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 | URL: http://www.journals.elsevier.com/she-ji-the-journal-of-design-economics-and-innovation/
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
--
-----------------------------------------------------------------
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
-----------------------------------------------------------------
-----------------------------------------------------------------
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
-----------------------------------------------------------------
|