Dear Francois,
Whoa. Its obvious there are some misunderstandings.
Let's start with the dynamic modelling.
To avoid confusing things, I'll describe the process of creating a system
dynamic model only. The steps of the process are as follows:
1. Meet with stakeholders and experts to identify what needs modelling, the
bounds of the model, which are significant elements of the model, the nature
of the causal relationships between different elements.
2. Create a 'causal loop diagram' this is a diagram that contains all the
elements described above and the relationships between them. As examples,
see,
http://orgalab.wikispaces.com/file/view/Case_4_CLD.jpg/121252941/Case_4_CLD.
jpg , and
http://www.emeraldinsight.com/content_images/fig/2970010201001.png or do a
Google image search on causal loop diagram. Creating the causal loop diagram
can be done on paper or in software.
3. Review the causal loop diagram with stakeholders and experts to check it
includes everything that is important and the linkages are correct, and
gain further information about the detail of each of the relationships
between elements.
4. Create a stocks and flows dynamic model based on the causal loop
diagram. This stocks and flows model needs to be created using system
dynamics software. Nowadays, I use mainly Vensim (free download of the
academic version). An example of a stock flow diagram for design of mobile
voice services is at
http://origin-ars.els-cdn.com/content/image/1-s2.0-S0308596111001315-gr9.jpg
, and a simple sales/adoption model that could be used by many design
businesses is at
http://www.sg.ethz.ch/media/medialibrary/2012/10/diffusion_mixed.png
5. Once these stock flow models are created in system dynamics software, the
relationships and values of variables can be added to the model in
mathematical form. This creates the *dynamic* mathematical model that can
be run in a speeded up form for the designers to see the outcomes of
different changes over time. Some simple examples are shown at
http://en.wikipedia.org/wiki/System_dynamics (these are not particularly
good as they go past so fast - but they are the easiest to find!)
6. Then, using historical data, check the dynamic behaviour of the model
with real world changes in the past and tune the model as needed.
7. The use the system dynamic model to show how the consequences of your
design decisions play out in time in the future.
There are lots of intro videos and similar descriptions for the above on the
web.
As you can see, the 'causal loop diagram' is simply a diagram of
relationships. If you added figures to it, you would have a single snapshot
in time.
In contrast, the stocks and flows system dynamics model shows over
(speeded up) time how the outocmes in the future of design decisions will
play out.
This is a different way of thinking. It assumes all outcomes vary over time.
It contrasts with common assumptions that the consequences of a design
(e.g. a poster) are fixed and simple (e.g. the number of people attending a
concert) and instead reveals (depending on the model) the implications over
time on all sorts of different planes (e.g. changes over time in people's
attitudes to particular types of music or genres of poster design, changes
in social interactions between people, changes in amounts of disposable
income people might consider allocating to concert attendance etc).
Four other quick points. First, most of my public publications except for
the recent ones are available at
http://www.love.com.au/index.php/publications just download them. Second,
I've described the detail of many of the discussions before on phd-design.
Again just search and download them. For a long time, I've regarded
phd-design as a legitimate way of publishing ideas and analyses, especially
in light of the relative lack of use of journals by most people (though I'd
recommend Transactions of the Royal Society). Third, like many on this list
who do commercial research, the publication outputs are not publicly
available and this often comprises major amounts of research and analysis.
For example, recently I was a member of a team project for which the final
report was over 400,000 words that had to be checked by several dozen
stakeholder groups and went to twenty something revisions. The report was
for a government agency and will almost certainly never be published whole.
Instead, the outcomes will appear as policies and funding decisions. Four
is that much of the evidence is by exemplars that demonstrate points to
support analyses . It needs you to think through the reasoning rather than
have it in the form where it has been pre-processed. For example, some time
ago, I wrote that systems dynamics and similar modelling provided a way to
address issues that were typically considered 'wicked' problems. I followed
this with a list of weblinks to papers that described system dynamic
modelling and prediction of outcomes for various problems that fitted the
wicked problem definition of Rittel and Horst. Then ( if I remember right)
I outlined why I thought these examples showed more generally that this was
a path to identify solutions to wicked problems. I also pointed to a class
of problems to exclude - those intrinsically unsolvable by any means (for
example, they assumed perpetual motion, were logically a null set etc). It
is necessary to read the links to see whether for you they provide
sufficient evidence that confirm the argument, or not.
Finally, at heart, I feel the differences between what I've been presenting
and the reactions from some areas of design are a matter of differences in
perspective on the role of designers in the world. For a couple of hundred
years, Manchester culture has encompassed a particularly Mancunian radical
open perspective on the world that looks forward to global change and
improvement. Part of this has been grounded on the position that everyone
has a right to be involved in creating the world and a part in it. This has
resulted in the past and present in a wealth of public services for workers
and the poor including free schooling for poor children at a time when
schooling was paid for (Chetham bluecoat school). It aslo resulted in early
free provision of education for adults and free access to knowledge. One
part of this was the creation of the Whitworth Mechanics Institute that
became the UMIST that encouraged John Chris Jones and Nigel Cross to teach a
Masters course on design research - resulting in the formation of the Design
Research Society. As a centre for radical politics Frederic Engels and Marx
studied and met together regularly in Manchester. The Cooperative Group
started there in 1862 and with 6 million owner members is the largest
cooperative business in the world. Manchester culture has been characterised
by an open world view of responsible human beings looking for opportunities
for improvements and change. This perspective has spanned from the poorest
unemployed person upwards. Yesterday, I had the pleasure of seeing again
the BBC play Comedians. The dialogue between Eddie Waters and Gethin Pryce
throughout the play illustrates well some of the subtlety of that Mancunian
future-focused responsible perspective, whether engaging with the world as a
comedian or as a designer. Its available on
http://www.youtube.com/watch?v=GUEB4PAZMRk and comprises about 10 parts. I
recommend it.
From that perspective, it is important for designers to be responsible to
themselves, their clients and the world in general. For designers to be able
to be responsible requires them to be able to predict the consequences of
their actions. In turn, that requires forecasting the consequences and
effects of their designs.
Some designers and designers might say this responsibility doesn't matter,
and that the old traditional ways are enough. I don't believe that and I
don't believe that designers at heart feel that. I suggest something deep
in all of us knows we need to feel responsible for what we do.
Then the issue is a matter of how designers can predict the consequences of
what they create - for themselves, their clients and for the world. I've
simply suggested a set of tools that to me seem to work in enabling
designers to take up their responsibility. These system dynamics modelling
tools were designed for that purpose., They were created as a result of
awareness that using other more traditional methods it is pretty well
impossible for designers to predict outcomes of their designs. The proof
these methods work is by seeing how many people have successfully used them
for a variety of purposes, and then using nous to transfer that
understanding to design practices.
Also yesterday, I met with a creative director for one of Australia's
largest and most progressive advertising agencies to ask how Australian
advertising agencies predict outcomes that result from their designs. He
described a move beyond focus groups and the like towards far more
technically-based dynamic modelling approaches to predict outcomes and
these linked to the contract with the client (as well as indirectly with
advertising codes and regulations). These changes appear to align with
the direction I've been suggesting.
In respect of providing more references. On one hand, I have you suggesting
I provide more pointers and on the other hand Ken has been suggesting that
I provide less. Perhaps it is easier if everyone finds and analyses their
own references?
Best wishes, and thank you again for your comments ,
Terry
---
Dr Terence Love
PhD(UWA), BA(Hons) Engin. PGCEd, FDRS, AMIMechE, MISI
Honorary Fellow
IEED, Management School
Lancaster University, Lancaster, UK
Director,
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]
--
-----Original Message-----
From: PhD-Design - This list is for discussion of PhD studies and related
research in Design [mailto:[log in to unmask]] On Behalf Of Francois
Nsenga
Sent: Monday, 23 September 2013 12:15 AM
To: [log in to unmask]
Subject: Re: Maths for Design
Dear Terry
Thank you for sharing the only two references you have provided so far.
The first (as I understand it was suggested offlist by Ken?) is D. Meadows'
book on <Thinking in Systems>, that I am going to purchase and read later.
The second is your own paper, <Design Guideline Gap and 2 Feedback Loop
Limitation: Two issues in Design and Emotion theory, research and practice>,
that I have just finished reading and commenting upon here below.
My first, immediate, and general observation, at last, is that you seem to
(with)hold a legitimate and foundational insight that I am even really
inclined to partly share, and eventually comment upon in my non-mathematical
way.That is on how carefully we ought to conceive and practice our
profession, each in our own respective subfield of practice.
None of us practice "Design" in general. Consequently, we shouldn't be
expected us all to provide one single and similar, mathematically or
otherwise, proof of our professionalism. Let's, once for all, admit that, in
certain Design subfields, personal 'intuition and guessing', you say (p.
7) are highly valid as a proof of good practice. In other subfields,
however, other means may be required to prove one's worth as a professional,
mathematical logic being one among many of those means. In my view, the only
paramount proof that we all must provide, no matter in which subfield of
involvement, pertains to the level of risk and the nature and quantity of
benefits embedded in the artefactual outcomes we 'prescribe'.
And risk as well various benefits can be conceived and conveyed under
various means, including mathematical tools of any kind and any level, as
needed.
My second overall observation, more personal, is that, besides your opinion
held and defended for the last 3 decades you say, up to now to my/our
dismay, inspite of my/our insistance, you still don't provide substantiated
and convincingly proven evidence of the main point you have been advancing
thus far: that there are mathematical approach(es), tools, and method(s)
that would enable ALL professional designers and evenly in ALL subfields, to
provide potent and irrefutable guaranty of corresponding harmlessness,
together with more advantages over inevitable losses incurred through use of
respective design outcomes.
I am - and I guess many of this list are - wanting for more from you. And
yet, you quite overtly seem to be not willing to deliver. For instance, in
addition to the uselessly length of this thread, on three occasions you tell
us, in the only paper you have indicated to us thus far:
1. <This latter topic is the subject of a different paper> - Which and where
is/are your or other sclolars' paper (s) on < neuro-cognitive studies and
(...) simple practical tests of human ability to understand situations with
2 or more feedback loops>? Where is the report on <evidence from the systems
field over a long period and a wide variety of other subject fields together
with deictic empirical testing>? Has the < cross-institutional research
collaboration> you mention ever been funded and findings published, on the
<formal large-scale trial to test...the biological limitation of humans in
being able to predict dynamic design outcomes...the validity of the 2
Feedback Loop Limitation across all humans, regardless of personal skill,
intuition, cognitive ability, emotion, feelings, creative skill or
education>? Have you, or anyone else of you knowledge has ever published a
compelling proof that mathematical <ability to predict behavior is important
in Design>, and the only one important? (all on p. 6)
2. Has your co-authored paper mentioned on p. 9 in relation to <Complex
design situation> modelling, on <Preliminary design relationships affecting
crime in a rail corridor> ever been published? Which and where is its
bibliographic reference?
3. You say, on p. 11: <this issue - Redefinition of wicked problems as being
straightforward to address rather than impossible - has been addressed by
the author - i.e. you - in other papers>. Which papers, and why don't you
indicate them to us so we can benefit from them as well?
Why are you so reluctantly obstinate not, at least very briefly, to instruct
us on the < Three obvious - not so obvious to many among us - approaches to
addressing and resolving the 2 Feedback Loop Limitation>?
(p.11)
Regarding the said <three approaches>, which precisely are they? Reading
your two closing paragraphs - with my admitted poor ability in English - I
am under the - most probably wrong - impression that you elaborate a little
ONLY on the first one, <the use of modelling>. And then, which clearly are
the two others? There seems to be a cofusion here: are the <three
approaches> same as the <three primary modelling tools - causal loop
diagrams, systems dynamic models and agent models -> ? (p. 12).
Even more puzzling is that you recommende the three tools above, and yet you
tell us that:
- < Causal loop modelling is restricted as it provides only a snapshot
in time and cannot provide understanding or forecasting of dynamic
outcomes. This means it will result in faulty designs.>
- <For Systems Dynamic modelling and agent-based modelling of design
situations with multiple interlinked feedback loops, then the prediction
of
the behaviour of the designed outcome is only available by watching the
system play out in real time. These two provide causally-based design
approaches>
First, if you conclude that 'Systems Dynamic modelling' and 'agent-based
modeling' both <provide causally-based design approaches>, that means they
all are the same and all lead to the same conclusion? Meaning that, as in
case of the <causal loop modelling>, the last two leading to <faulty
designs> as well?
And second, do I then and lastly understand quite well that, at the end,
there is no point going through the entire exercise of mdelling, since < the
prediction of the behaviour of the designed outcome is only available by
watching the system play out in real time>?
The second, of the above announced three <approaches to addressing and
resolving the 2 Feedback Loop Limitation> seems to be the <associative
modeling approach>. And you say that this approach is also limited <because
causal mechanisms can change and make associative patterns faulty.> Unless
one <undertakes research to develop an understanding of the causal
mechanisms at a causally more detailed scale.> But even then, referring to
what you tell us above on the first approach, that of 'Causal loop
modellig', even with the 'associative modelling approach', there always is
risk of ending up the entire exercise with no <understanding or forecasting
of dynamic outcomes>, and thus yielding only <faulty designs>, again. So,
what is its worth?
And finally, the third modellig approach, is it missing by inadvertance in
the development of your paper?
Dear Terry, to summarize this thread, I sincerely wonder why you don't want
to deliver, given your highly valuable insight, in my opinion as confided
above; also, given your professional experience, your academic credentials,
and...your maturity in judgement and age!! Would there be, perhaps, more to
find in number 39 of the bibliogrphic references to your paper? ( <Love, T.
Can you feel it? Yes we can! Human Limitations in Design Theory (invited
keynote), In CEPHAD 2010. P. Galle, Editor. 2010, Denmark DesignSkole:
Copenhagen.>)
Unless, as I hinted at in my previous post, the whole scheme is a teacher's
trick to incite students to dig deeper into a provocative <hypothesis> (p.
7). Then, if so, I sincerely hope your goal will be attained, that someone
will soon take it on from here and undertake a PhD to proof or disproof
your assertions.
Best regards
Francois
-----------------------------------------------------------------
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
-----------------------------------------------------------------
|
