2014/06/03 FORMakademisk Ing review

Mr. David Ing:

We have reached a decision regarding your submission to FORMakademisk, "Design Flaws and Service System Breakdowns:  Learning from Systems Thinking".

Our decision is to: Revisions required

See the comments below

My responses to the review comments are inserted in bold.

Editorial comments to the author:

The following reviews suggests changes to the article that would enhance its integration between and among points to support more of an overall argument to its structure.  As these are typical suggestions for article revisions, the editors ask you to consider the suggestions and provide a revision. 

Several approaches to writing style are proposed to better support these suggestions:

•    Reduce the number of bullet points, converting some lists to discursive paragraphs. 

Point taken.  I prefer bulleted points for clarity in reading rather than long blocks of text.  I have used numbering in some places (i.e. (i), (ii) ...) to try to be similarly clear instead of spacing vertically.

•    Do not use indented (second level) bullet lists.

Done.  In section 1.1, I've removed the second level bullets, changing the spacing into paragraphs.  (They were formerly already semicolons).

•    Don’t mix bullets and numbered lists for the same content where lists are used. 

Mostly done.  As opposed to using "Stil1" and bullet lists, I can use "Stil1" and number lists.  This fails under section 3.3, where the numbering starts at 0 rather than 1.  In LibreOffice, I'm unable to set the start number to 0.

•    Use shorter heading titles, as the longer titles (especially where indented) are suggestive of the lecture slides indicated by reviewer comments.

Considered.  This is a style of writing from my consulting experience, following Minto's Pyramid Principle.  Full sentences reduce ambiguity in the meaning of sections.  The sentence that follows the section should not be a repeat of that heading.  I write that way whether I use slides or not.

•    Reduce the use of first names in cited references, just using last names and citing years, unless a specific author is discussed as an historical personage.

Inspected.  The only places that first names are used are in section 1 with three names, and section 4 with two names.  Both are cited as historical personages.

The article discusses several complex frameworks that will not be familiar to all readers. Wherever possible, graphical images or tables might be included to help readers form a more complete representation of the systemic theory or model.

Done.  You asked for it.  Ten figures, one table.  The figures originally drawn in LibreOffice Impress have been transferred into LibreOffice Draw, and then exported as SVG.  Source files are provided.

The article cover significant topics important to the journal issue and it would appear that there is room in the request for revision to develop or present a personal perspective  and argument from the author as well. This might be presented in a brief conclusion or discussion section at the end of the piece.

I've expanded the conclusion a bit, to more strongly call for engagement of designers with the systems thinking community.  I could go into service systems thinking to a greater extent, but that's really a talk for RSD 2014, not RSD 2013.


Reviewer A:

The following is a mix of my own summary and notes and some critical comments / questions:

I've written some responses to the "summary and notes" that haven't impacted the revision of the article.


A very brief and condensed but consistent Introduction with a slight bias towards a mechanistic and systems engineering view of socio-technical systems. 

This criticism is up for debate.  My view isn't a "mechanistic and systems engineering view of socio-technical systems", but instead (i) an outline of systems thinking as applied across physical, biological, social and ecological systems, and (ii) attention drawn specifically to service systems, which are human systems.  The target audience for paper presumes that designers work with human systems and their artifacts.

The self-organizing and autopoietic aspects of social systems are neglected.

In the introduction, a discussion of self-organization and autopoesis would have been too deep.  Later, the language action perspective includes aspects of self-organization and autopoesis embedded in the theory, but that's well beyond the overview provided.

These aspects are introduced in the beginning of chapter 2. And I am not sure if the author argues or wants to suggest that designers are not aware of the systems thinking approaches to deal with these types of problems? Well, some are, some are not.

1 Activity package mismatch:  Theory of the offering

2 Coordination fumble:  Language action perspective

3 Change target discord:  Reactivism, inactivism, preactivism, interactivism

4 Resource scaling collapse:  Supply side sustainability

5 Environmental context shift:  Causal texture theory

6 Pacing layers trap:  Coevolution and learning

7 Regeneration failure:  Panarchy

Yes, in section 2 appears: Designers may be unfamiliar with some of the advances in knowledge gained over the past two decades.  The "may" should satisfy the "some, are, some are not".

Some of the patterns point to the necessity of languaging, to enter conversation, to see the design process as an argument (Rittel, Krippendorff, ...).

I suspect that reviewer has a systems perspective that is language-centric.  Language is certainly covered in the condition on "language action perspective".  The greater emphasis is placed on the inquiring system, i.e. Collaborations between designers and systems thinkers may be formalized as inquiring systems.  How else would the inquiry happen, if there wasn't languaging?

Some other patterns look very mechanistic and OR-like and not really helpful for designers.

I don't understand the criticism of "mechanistic", as four of the seven conditions come from ecologists and biologists.  I can't judge the usefulness for designers, as I'm not a designer, but am willing to have the conversation.

Some - such as 6 - are very specific, but nice (How Buildings Learn).

Okay.  I don't really understand why "How Buildings Learn" is nicer than the others, or how it's specific, but that doesn't impact the revision.

Designers have always known that their expertise comprises episteme - phronesis - techné. They call it inspiration, ideation and implementation or analysis, projection and synthesis or ...

-> a pattern language for communication between systems thinkers and designers... why not?

-> action research is one of the main methodical approaches in practice-based design research...

If the audience already knows the terms episteme, techne and phronesis, I'm preaching to the converted.  My guess is that the average designer doesn't know.

My experience (limited to date) is that designers are not familiar with a pattern language in way that Christopher Alexander means, and saying "patterns" is not a pattern language.  In addition, my impression is that designers rarely formally propose an action research project where the designer becomes internal to the action.

The suggestion of UST comes a bit abrupty and unexpectedly and should be explained a bit more.
When referring to Churchman and Singer the concept of CST (Critical Systems Thinking, Werner Ulrich) comes to mind...

Done, as an expansion of section 4.  Critical systems thinking (Ulrich) is a step further than just the description of the design of inquiring systems.  This is valid, but I don't want to add another reference to already long list.


To conclude:

No objections to this text, it makes sense and triggers debate.

Acknowledged, thanks.  Does this mean that I didn't need to respond to the comments by the reviewer listed above?

Just one concrete question:

Why exactly THESE 7 patterns and no others. There are - for example - the systems archetypes in System Dynamics or other collections...

Fixed, as the just before section 2.1 starts the first of the seven conditions. The number of conditions was arbitrarily chosen as seven, at the maximum capacity of cognitive load (i.e. Miller's number, 5±2).  In an interactive setting, a group could be led to cover a broader body of knowledge, with each participant producing his or her own personal list of 7.  This at least answers the question, and provides a path for others to create their own lists.


Reviewewr B:

This paper has an exciting title and tantalising content.

I use the word "tantalising" deliberately because it means to arouse expectations, but to disappoint them, purportedly from the Greek Tantalos, king of Phrygia, son of Zeus, punished in the afterlife (for an offenses variously given) by being made to stand in a river up to his chin, under branches laden with fruit, all of which withdrew from his reach whenever he tried to reach for them

Agreed.  The article was not ended as a finished product, but instead an invitation to a conversation.  On the path towards designers becoming (stronger) systems thinkers, the interested parties might have a strong appreciation for what they do or do not know about systems thinking, and assess whether they might benefit by an investment in energies towards further learning.

After a good start, this paper reads like a lecture with powerpoint slides, in the sense that there is a series of statements, or even bullet points which the author, were he in a lecture, would have elaborated. Without this “inbetween”  supporting information, the reader goes from statement to statement. These statements are rich in meaning and provoke powerful responses, but without the needed linking information, the reader cannot follow and decipher the author’s messages. Since the whole text, including references, is very short, we would like to encourage the author to supply the missing links in the line or argument, with more information. We would like to suggest, that, if the editorship of the journal allows, the style even be allowed to remain to read in a more informal way, like the transcript of a lecture.

Agreed.  Revision including diagrams should help, at least some.  The article has now expanded from 11 pages to 14.  The reference list is full and complete, so interested readers have a path for following through for more depth.

I hope that this suggestion, if allowed by the editors, will on the one hand enable the author to quickly address the issue, and not lose the power (and immediacy) of the statements already made.