[Sigia-l] Making assumptions

[Jonathan Baker-Bates]

On Tue, 2009-08-04 at 00:20 -0400, Ziya Oz wrote:
> Yes, this is the key: 90% of the time I'm shown
> tests/validations/controls/research/etc it is a *combination* of multiple
> factors whose inter-relationships are often unknown, unknowable or plain
> contradictory. But because it *may* contain an element of 'truth' the entire
> result is accepted consciously or subconsciously. That's not how scientific
> method works. 
> 

That's another way of describing the problem I'm trying to solve with
this, so I'd agree with that.

> > So am I to understand that with the exception of multivariate tests, you
> > would not consider design research at all?
> 
> I'm all for immersing oneself in the subject at hand in as many ways as
> practical. But using voodoo science to 'prove' a point and calling it a
> test/validation/control is counterproductive, especially if your own money
> is on the line. I've noticed that when people's income is directly related
> to actual results they tend to shy away from aggressively promoting such
> 'research'.
> 

My own money is indeed on the line, so chalk up a point in support of
that particular assumption. (Ah, I feel the power of recursion flowing
through me!)

> What I'm saying is that people, in general, may bitch about Apple's 'opaque'
> design approach, but can hardly question its actual outcome. Results speak
> louder than intent and academic/commercial mumbo jumbo.
> 

Apple may well be the only company in the world were a product can be
brought to market by force of design strategy alone. But until we see
"baby Apples" arising from some (so far postponed) design revolution, I
don't have much to say about the way they do things other than to ask,
as an aside, the opposite question: to what extent could Apple's
failure's have been lessened? 

This is basically what I'm on about here. The "design assumptions"
approach I've outlined is based on the idea that design can be informed
by research and data without being stifled by it. I think we would all
agree on that much at least. So my response is to decouple research from
the ebb and flow of project work - get it out of the "design process"
and into a regulatory role that addresses general issues to do with
customers, markets, products and patterns. With research in this form,
I hope I'd be able to counter the FUD that comes my way every day
without having to demote the role of design. 

Wish me luck. 

Jonathan