Terry
I wanted to say something on color perception. I was taught by former Bauhaus teachers but became exposed to experiments in social perception which taught me the tremendous cultural differences in perception, including of colors. I worked for a year in an institute for visual perception and presented my views in writing, but recently discovered evidence that color perception certainly is triggered by physical stimuli but the do not explain our culturally coherent sense of colors.
However, now that you clarified your aim of developing a predictive design theory I need to respond differently.
First I side with Davis sless who suggested that your aim is thoroughly (not his word) positivist
Second, for a theory to be predictable requires the pattern that the theory aims to describe to continues as observed in the past. To me design is by definition innovative, ideally transformative (ranging from changing everyday practices to transforming social structures). A design theory that predicts the consequences of a design contradicts (my conception of) the mission of design to introduce unexpected changes in the world. Such a predictive theory can only encourage strengthening what is already know, not to change it.
Third, I am of course cognizant if the need for designers to justify their design to clients, all kinds of stakeholders, including users and advocates for the environment. I my opinion, a predictive theory would not be convincing. What could inspire the stakeholders of a design to become enrolled In a project of realizing it are plausible arguments. In the semantic turn I have explored several rhetorical strategies They include experiments on sub-populations as well as self-fulfilling prophesies.
Science fiction, futuristic novels, as well as ethnographic accounts of how people are struggling in life may well inspire designers. But Google searches provide mostly data of what their authors want you to know exists. They rarely predict the innovations introduced that designers introduce.
Klaus
Sent from my iPhone
> On Feb 21, 2016, at 12:25 AM, Terence Love <[log in to unmask]> wrote:
>
> Hi Don and all,
>
> For design research, one of the biggest and most important challenges, is developing a body of theory for predicting the outcomes in the world of the effects of individual designs.
>
> It's in this context I raised a relatively abstract meta-analytical question about design theory making, relating to decisions about the character of variety in how we represent factors that are elements of design theories. This is an issue important to both predictive design theories themselves and the representation of concpets and factors that we use as professional designers
>
> There are advantages and disadvantages of 'fixed and known' variety versus 'infinite variety in representing the factors used in creating predictive theory about the consequences of designs in the world. I've copied the original post below.
>
> I used colour as an example of such a factor. First, because it is well understood phenomenon on this list (or so I thought), and hence wouldn't need to be explained. Second, colour is clearly a phenomenon that, like any phenomenon, we can represent as an approximation via either a continuous function or discrete stepped function. Also, I thought that double aspect of the representation of colour wouldn't need to be explained as we clearly use both in design.
>
> None of the above, I suggest, is contentious. It is taken for granted and straightforward in much of the discourse.
>
> The real challenge is to look at possible approaches to creating design theories that predict design outcomes as the consequences of design in the world.
>
> These kinds of design theories are difficult predictive models. The most immediate ambition is to make useful projections into the future of the consequences of individual designs. It is not expected that these theories will provide definitive predictions, rather that they will be usefully correct, at a level sufficient to aid with design decision making. Some simplified predictive design theory models are already possible and in use, for example in the realm of environmental and life cycle costs.
>
> It requires new sorts of design theories, however, to extend the ability to predict the consequences of individual designs more fully into other realms such as social analysis, lifestyle change, future innovation, politics, planning, economics, security and geopolitics.
>
> The variety characteristics of factors used in constructing and describing designs is obviously a key issue (e.g. in the manner Purma described about colour).
>
> We have a choice to include any design-related factor in predictive design theories via either continuous or discrete representations or some combination. In the case of colour we have a choice of either
>
> In the post, I asked for comments about which people felt was going to work better in the development of new predictive design theories that identified the consequences of designs.
>
> If we choose to represent design factors and their variety as continuous functions, then the predictive structures embodied in design theories will need to accommodate this. It drives representation of the design theories into, for example, the realms of non-linear n-order calculus, which on one hand offers potential benefits in identifying optimal positions in m-dimensional design space. On the other hand it presents significant challenges to solve the functions in creating such theories in which n and m are large.
>
> In contrast, if we choose to represent design factors and their variety as discrete functions (identified via just noticeable difference or the several other measures of limits of discrimination (which as far as I can see combine vectorially)), then the predictive structures embodied in the design theories can be chosen to accommodate this in a different manner to addressing continuous functions. Rather differently to the above, it enables the structure of design theories to move, for example, into vector space with predictive projection based on matrix-based analyses, and time projections based on finite difference or time step methods. The predictive processing is easier to compute, but identifying optimal directions and points in the design space is less easy. The challenges are in creating such predictive design theories in which n and m are large are less.
>
> I'm aware many will regard these as relatively new areas of design research. In fact, they originated in the earlier parts of the last century and were the foundational issues on which design research was developed. They can be seen as the central concerns of Rittel, Zwicky, Jones, Bertalanffy, systems research, behavioural science, operational research and many others. I would tentatively suggest that the underlying aim of most research in areas such as the perception of colour is to contribute to this larger and as yet not well addressed aim of developing theory to be better able to predict the outcomes of designs in the world.
>
> Again, I hope that clarifies my original post... but I'm aware people might be much more interested in colour theory!
>
> Warm regards,
> Terry
>
> ---
> Dr Terence Love
> PhD(UWA), BA(Hons) Engin. PGCEd, FDRS, PMACM, MISI
> Love Services Pty Ltd
> PO Box 226, Quinns Rocks
> Western Australia 6030
> Tel: +61 (0)4 3497 5848
> [log in to unmask]
> www.loveservices.com.au
> --
>
> Original post
>
> 'Is it better to assume as an axiom in making design theory that colour always comprises a set of fixed colours rather than assuming colour as being a continuous spectrum? There seem to be strong reasons in theory and practice to make this assumption, and that it is possible calculate exactly how big the set is (i.e exactly how many different colours) for each design scenario. The advantage in design theory and research is between fixed and known variety and ‘infinite’ variety.
>
>
> -----------------------------------------------------------------
> 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
-----------------------------------------------------------------
|