Wednesday 8 September 2010

Two more types of design choice, and no more? Do I know what I am torcing about?

My blog story so far (even if this sort of stuff isn't proper blogging), in eight moves:

(1) Any one centre for design is often going to be the wrong centre

(2) There are many ways to be human-centred, as three waves within HCI research have already shown (HCI = Human-Computer Interaction, the study of how Humans Interact with Computers, and how we can improve this). Claiming to have a world exclusive on the one true human-centred path is a mug's game (with no limit it appears to the teams of 1 that fill its leagues)

(3) Designers make choices, and we can associate these with types at different levels of abstraction

(4) Choices of type at one level of abstraction may become the same type at a level above

(5) there could be a hierarchy of design choice types here, but working out exactly what it is, and eliminating overlaps is a mug's game (see 2 above, there's no room left for more mugs)

(6) At the highest level of abstraction, choices about artefacts and choices about evaluations are very different things. After all, someone can choose to evaluate a designed artefact long after the designer has departed this mortal coil (e.g., the Snettisham torc is lovely, but we've no idea who designed it)

(7) I wonder if there are any other types of choice that are distinct at this highest level of abstraction

(8) I know the answer, so I'm not really wondering at all

OK, time for prose, it's beginning to look like Powerpoint up there (I'm so unbloggerly).

Artefacts, Evaluations - anything left? Well, I suppose we could try designers (in HCI actually, we regularly try designers, and they are generally found guilty).  Seriously though, a designer's intent is something quite distinct from their designed artefact that (fails to) realise(s) their intent.  Let's call this purpose.

That's now three distinct types of design choice: Artefacts, Evaluations, and Purposes. In design situations where there are clear choices of purpose (albeit ones that may evolve alongside artefacts and evaluations), it is much easier to focus evaluations. With just artefacts and evaluations, it's very easy to come up with arbitrary evaluation criteria (e.g., the Snettisham torc is too soft and pliable). But it's clear that you were never meant to chop wood with it on, so such an evaluation criterion ignores the likely original purpose of the torc, e.g., to communicate and confirm wealth and status in social gatherings, to be an object of awe and contemplation, to demonstrate the amazing skills of its designer maker, and perhaps more. As a further twist, the Snettisham torc is today associated with a different range of cultural and heritage purposes at the British Museum. Nothing appropriates as well as a national cultural institution.

Once we have more than one high level type of design choice, we have to connect between them to maintain design coherence and integrity. We can (and should) connect evaluation to purpose. Similarly, when reflecting in/on design, we can connect the artefact to its purpose: how do choices for the artefact relate to intended purpose? What is this feature or quality for? What does or can this feature or quality achieve?

So that's three types of highly abstract design choice, and two types of connection, which begs the question, what about connecting between the artefact and its evaluation?  Can questions asked at this level of abstraction get us anywhere? For me, the answer has been yes. Digital artefacts in particular can be given self-instrumenting features. A web page counter is one simple example. It's not necessarily a good example, because it begs the question as to what the counter is actually evaluating, but it could nevertheless serve a useful evaluative function.  However, there are also examples from the non-digital world, such as wear indicators on brake pads and electric toothheads, or power level indicators on dry cell batteries. What's intriguing here is how direct instrumentation can have a sort of Heisenburg effect and change the object, adding value through its own self measurement.

So that's three types of highly abstract design choice, and three types of connection, so let's make this Abstract Design Situation more complex by looking for a fourth type of (very abstract) design choice.  Actually, we can get more complex without that. If we consider interaction as a connection between an artefact and one of its purposes, then we can evaluate the costs incurred in interacting relative to the benefits achieved (a.k.a. worth). If you thought things couldn't get more abstract, then tough. What we have here is evaluation connecting to a connection (interaction = artefact connected to purpose through use). Who would have thought we could have such fun with just three abstract design choice types?

For choice type three, we introduced a human, albeit a baddy, the designer (or design team) with their chosen purposes for making the world a better place, or so they think. We need some good guys to balance out designerly malfeasance. We're talking now about the people who are being designed for, as and when that is, speciific people are being designed for. Such lovely beneficiaries are clearly a further distinct type of design choice at the highest possible level of abstraction.

For doubting Thomases with a second hand philosophical vocabulary from wikipedia, artefacts, evaluations, purposes and beneficiaries are ontologically distinct. Artefacts and beneficiaries are corporeal, material, substantial, here in the now, and in this sense are ontologically similar. However, artefacts are manufactured, whereas beneficiaries are socially constructed, and in their role as beneficiaries rather than physical bodies, they are less substantial and incorporeal. They are selves and others overlaid over physical  bodies.  However, they also bind together many facts, things are true of both stuff and folk.

Evaluations and purposes are not bundles of facts, but respectively judgements and intents. The former are axiological rather than existential, that concern values and not facts of being. The latter are intentional, mental phenomena in the mind of a designer (or shared across the minds of a design team).

Further probing produces further ontological distinctions. Purposes refer to futures. Designing pulls conceived artefacts from the future into the present.  Beneficiaries persist in the present, but have meaningful significant pasts and anticipated futures.  Evaluations however are often one night stands: some forethought and planning, a lot of frenzied and sometimes anxious activities, a closing "how was it for you?" and  suddenly it's all behind you.  I could go on. You know I could go on.  Kindly though, I won't.

Abstract Design Situations combine one or more highly abstract types of design choice, as many connections between them as are committed to, and any useful connections to connections, and connections to connections to connections, ad infinitum (unfortunately, closure will be forever elusive in the absence of designer assent - the situation's as big as you want it to be).

Try as I might, I can't get beyond (or below) four possible highly abstract types of design choice. There's a good home for any others that anyone else can identify, so do let me know if you encounter some massive overlooked generality on your design travels.